Full text of "Bulletin of the Museum of Comparative Zoology at Harvard College"

See other formats

hob

HARVARD UNIVERSITY

Library of the

Museum of

Comparative Zoology

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE

Vol. LXXXVIII, No. 1

ADDITIONS TO THE RHEOPHILOUS MOLLUSK FAUNA
OF THE CONGO ESTUARY

By Joseph C. Bequaert
and W. J. Clench

With Two Plates

CAMBRIDGE, MASS., U. S. A.
PRINTED FOR THE MUSEUM
April, 1941

tJBRASH

No. 1. — Additions to the Rheophilous Mollusk Fauna
of the Congo Estuary

(Studies of African Land and Fresh-water Mollusks, No. 13)

By Joseph C. Bequaert
and W. J. Clench

In a first account of the rheophilous mollusks of the Congo Estuary
(1936, Mem. Mus. R. Hist. Nat. Belgique, ser. 2, fasc. 3, pp. 161-168),
we anticipated the discovery of additional species in this peculiar
habitat. Our prediction was fulfilled much sooner than we had hoped,
due to the intelligent and active cooperation of Dr. M. Wanson. Fol-
lowing directions given by Dr. Maurice Bequaert, Dr. Wanson made
extensive collections in the region of Matadi during 1938. The ma-
terial, which he/ very generously sent us for study, contains, in addi-
tion to all previously known species, representatives of four new forms.
Included also were many specimens of Septariellina, with the soft parts
preserved, enabling us to give precise information concerning the re-
lationships of this aberrant genus.

Dr. Wanson's outstanding discovery is the new Potadovia aggluti-
nans. Owing to the peculiar habit of living in narrow crevices of rocks,
immersed in swift current, this snail cements itself to neighboring
objects, particularly to the walls of the crevice and often also to other
individuals of the same species fixed to the rock in the vicinity. Ab-
normal growth of the shell results in all the adult snails being deformed,
often much so.

While somewhat similar cementing of the shell to foreign objects is
by no means rare in marine and fresh-water Pelecypoda, and occurring
occasionally in marine Gastropoda, we know of no other comparable
case in fresh-water Gastropoda. No doubt the adherence of P. aggluti-
nans to foreign objects prevents the snail's being washed away by the
swift current, in which it is immersed. Yet the evolutionary significance
of the process is not quite clear. The view might be taken that the
cementing, as well as the abnormal shape, are merely due to mechani-
cal pressure exerted on the snail while it grows in a cramped space
from which it is no longer able to move. The resulting obvious ad-
vantage given the snail against the current would then be purely in-
cidental, and the cementing would not be an adaptive process. A
ready objection to this view is that, while mere mechanical pressure
might explain the abnormal shape, it seems insufficient to account for
the cementing to other objects. This latter process may require a

4 bulletin: museum of comparative zoology

more profound change, perhaps of a physiological nature, in the
method of secretion of shell material by the mantle. In view of these
considerations, we suggest that it might be more rational to regard the
cementing as the result of selection. Certain snails which developed
the faculty of fixing themselves to foreign objects, were able to become
adapted to the peculiar environment of rocks immersed in swift cur-
rent, and thus had a survival advantage over others. In this connec-
tion it may be noted that P. agglutinans not only adheres to large
objects, where pressure may be the main determining factor, but also
incorporates in the shell smaller particles, where purely mechanical
influences seem to be ruled out.

We cannot urge too strongly further intensive research in the
estuaries of the Congo and other West African rivers (Coanza, Gaboon,
Ogowe, Niger, Gambia, Senegal, and the many smaller coastal rivers),
where equally surprising finds may well be made. In proof of this con-
tention, we cite the recent unexpected rediscovery of Potamopyrgus
ciliatus (Gould) by Dr. E. Dartevelle in the mangrove region of the
Congo Estuary, between Malela and Banana. This snail was origi-
nally described from the Deea River, near Cape Palmas, Liberia. As
it is extremely similar to P. corolla, of New Zealand, and as it had not
been taken again in West Africa for nearly a century, some doubt had
arisen as to the accuracy of the original locality. These misgivings
have now been removed, and we may expect P. ciliatus to occur at
many spots on the Coast of Guinea. Dr. Dartevelle found his speci-
mens living in the burrows made by shipworms (Teredo) in the live
aerial roots of mangrove trees (Rhizophora Mangle), while the burrows
are still inhabited by the Teredo. These mollusks and some others
(Nentina and Cyrcnoida) seem to form a regular biocoenose.1

SYNCERIDAE (ASSIMINEIDAE)

Septariellina congolensis Bequaert and Clench

Plate 2, figs. 1-3

Dr. Wanson found this species living in large numbers, fixed to rocks
in swift current, on the right bank of the Congo Estuary, a short dis-
tance above Matadi.

1 Further details about this interesting association of mollusks will be awaited with interest.
The rediscovery of Potamopyrgus was announced incidentally in a paper on the Teredo of the
Congo Estuary by F. Moll (1939, Rev. Zool. Bot. Afric, 32, p. 373), where it might be readily
overlooked.

BEQUAERT AND CLENCH: RHEOPHILOUS MOLLUSK FAUNA 5

The species was described from a single dead specimen, 2.6 mm. wide
and 1.8 mm. high. The largest specimen we have seen now measures:
6.8 mm. in greatest width, 5.2 mm. in height, 3.4 mm. in smallest
diameter (seen from above), the aperture 4.6 by 5.2 mm.; but it has
only the body-whorl left, the spire being completely corroded and re-
placed by a flat area near the columella. On the larger snails the
growth-striae are cut by numerous, finely engraved spiral lines, which
gradually spread apart toward the outer lip.

The preserved material sent by Dr. Wanson has made it possible to
study the soft parts and to define more accurately the relationships of
the genus. We had placed it, with some misgivings, in the Neritidae,
but this allocation was erroneous.

A most careful examination of many animals has failed to disclose
the slightest trace of an operculum. If this is present at all in the living
animal, it must be very small or thin and drops off after death. We
are convinced, however, that it has disappeared completely. As all
other characters and particularly the radula remove the snail from the
Basommatophora, we regard Septariellina as a fresh-water operculate
which has lost the operculum. We know of no comparable case among
fresh-water operculates, although the loss of the operculum is not so
uncommon in the marine Gastropoda. Goodrich (1939, The Nautilus,
52, p. 140) has noted the occasional loss, followed by regeneration, of
the operculum in the Pleuroceridae, but this seems to be an abnormal
or pathological process.

Other important features of Septariellina are the position of the eyes
at the apex of the tentacles and the large penis placed dorsally to one
side of the animal (Text fig. ID). Males are rare, only one being found
among some 75 snails examined.

The radula (Text fig. 1 A) is decidedly of the Taenioglossate type and
at once removes the genus from the Neritidae. On the whole it is
similar to that of the Bulimidae and Synceridae. Central tooth un-
usually wide, with trilobate base and very elongate basal areas; on each
side two long, sharp inner basal denticles; cutting edge with one large
central cusp and five smaller cusps on each side. Lateral (or admedian)
tooth fairly broad, with four unequal cusps; its base divided obliquely,
setting off a plate adjacent to the base of the inner marginal. Inner
marginal broad, with a prominent inner basal projection and about six
subequal cusps. Outer marginal slender, with many cusps.

Assuming that Septariellina is an operculate which has lost the oper-
culum, the position of the eyes at the tip of the tentacles and the divi-
sion of the base of the lateral tooth into two plates refer the genus to

bulletin: museum of comparative zoology

the Synceridae. According to Thiele (1927, Zool. Jahrb., Abt. Syst.,
53, pp. 113-146) this division of the lateral is the only peculiarity dis-
tinguishing the radula of all Synceridae from that of all Bulimidae
(Amnicolidae). Both families contain snails with and others without

Fig. 1. A, radula of Septariellina congolensis Bequaert and Clench; B,
radula of Valvatorbis manritii Bequaert and Clench; C, central tooth of radula
of Potadoma agglutinans Bequaert and Clench; D, tentacles and penis of Sep-
tariellina congolensis; E, operculum of Potadoma agglutinans.

basal inner denticles on the central tooth. In most Synceridae the ac-
cessory (detached) plate of the base of the lateral is more widely
separated than in Septariellina; but Thiele's figure of the radula of
Syncera microsculpta shows it in about the same position as in S. con-
golensis.

Pseudogibbula duponti Dautzenberg

Dr. Wanson found this species commonly on both banks of the
Congo near Matadi, fixed to rocks in very swift current. How far up
and down stream from the Vivi-Matadi area it extends remains to be
worked out.

Our largest specimen, 8.5 mm. high and 6.3 mm. wide, has a corroded
summit, so that the total number of whorls cannot be determined; only
the last three whorls remain. Shells up to 1 mm. high and 1.5 mm. wide
are complete, of 3 whorls: the first (embryonic) whorl is smooth, with-

BEQUAERT AND CLENCH: RHEOPHILOUS MOLLUSK FAUNA /

out either axial or spiral sculpture; the next third of a whorl shows only
microscopic spiral sculpture ; after which the strong spiral ribbing and
finer axial striation of the adult shell appear suddenly in full strength.

PSEUDOGIBBULA DUPONTI PALLIDIOR subspec. nOV.

A small lot of snails, from one colony on the banks of the Congo near
Matadi, are all uniformly very pale dirty-yellow, instead of the usual
dark chestnut-brown with violaceous tinge. No differences in size,
shape or sculpture could be detected.

Holotype. Mus. Comp. Zool. No. 112265; paratypes, Mus. Comp.
Zool. No. 112295. Dr. M. Wanson collector, 1938.

Valvatorbis matjritii Bequaert and Clench
Plate 2, fig. 9

Numerous specimens on stones of the banks of the Congo, in swift
current, near Matadi. The species is probably as common as P. du-
ponti, but more difficult to find, owing to its small size. No specimen
exceeds 2.2 mm. in diameter.

The radula (Text fig. IB), which we have only studied partly, is not
incompatible with the Synceridae, where we placed Valvatorbis. The
similarity of the central tooth with that of Scptariellina is particularly
striking; but there appears to be only one basal inner denticle on each
side. We are unable to state, however, whether or not the base of the
lateral is divided into two plates.

BULIMIDAE (AMNICOLIDAE)

Lobogenes schoutedeni Bequaert and Clench

Plate 2, figs. 11-12

Dr. Wanson collected many specimens of this minute snail from
stones at the margin of swiftly running water in the Congo, at Kala-
Kala near Matadi. None are larger than the types.

Lobogenes zairensis Bequaert and Clench

Plate 2, fig. 5

Four specimens of this species were found among the lot of P. du-
ponti pallidior, with which they agree in color. They are similar to the
type and about the same size.

8 bulletin: museum of comparative zoology

Hydrobia plena Bequaert and Clench

Plate 2, fig. 10

x\ single specimen of this minute snail was found among a large lot of
Septariellina congolensis, in the Estuary near Matadi. It appears to
be adult, though only slightly larger than the type: 1.8 mm. high and
2.1 mm. wide. The outer lip of the aperture is slightly flaring and some-
what thickened, distinctly sinuate basally in profile, the columellar
margin being again produced.

Hydrobia rheophila spec. nov.

Plate 2, fig. 4

Shell obtusely elongate-conic, thin, translucent, with closed um-
bilicus and very obtuse apex, colored a pale horny-yellow. Whorls
four; first depressed (a little corroded); remainder rapidly increasing
in size; second and third moderately convex; body-whorl strongly
swollen, evenly rounded oft* at the periphery, rapidly tapering down-
ward, taking in three-fifths of the height; sutures deep, simple. Sur-
face shiny, with only weak, much spaced, vertical, slightly curved
growth-striae; no spiral sculpture visible. Aperture simple, extending
below the base of the body-whorl, about two-fifths of the height of the
shell, narrowly ovate, vertical. Outer lip sharp, straight in profile
(probably not fully formed). Columella slightly concave, thickened
and folded back over the closed umbilicus. Operculum unknown.

Measurements:

Aperture

1.0 x 1.5 mm. Holotype; 4 whorls.

0.7 x 1.1 Paratype; 3)/£ "

Holotype (not fully adult). Mus. Comp. Zool. No. 112310, Estuary
of the Congo River at Kala-Kala near Matadi, Belgian Congo. Dr.
M. Wanson collector, 1938. Additional par at y pes (all much younger),
Mus. Comp. Zool. No. 112311, from the same locality and collector.
These snails were found on stones in swiftly running water, together
with Lobogenes schoutedeni.

Among the few described Ethiopian Hydrobia (= Paludestrina), H.
rheophila appears to be related to H. gabonensis Morelet, of the Ogowe
River (French Congo), which is only slightly larger (5.5 mm. high, 3
mm. wide, of 53^2 whorls; adult rheophila may possibly reach that size).
The shape of the shell, however, differs, H. rheophila being more coni-

Length

Width

3.6

1.8

2.6

1.5

BEQUAERT AND CLENCH: RHEOPHILOUS MOLLUSK FAUNA 9

cal, notwithstanding the more obtuse summit, with the body-whorl
much more bulging.

TIARIDAE (MELANIIDAE)

POTADOMA WANSONI Spec. IIOV.

Plate 1, figs. 9 and 11-12

Shell small, solid, though thinner than usual and somewhat trans-
lucent, elongate-conic, with slowly tapering spire (truncate through
corrosion), rounded periphery and attenuate, imperforate base. Adult
shells of 3 to 5 remaining whorls, at least 3 or 4 earlier whorls being
lost (by comparison with the youngest shells in the type lot). Whorls
slightly convex, with moderately impressed sutures, regularly in-
creasing in size ; body-whorl much longer than the combined preceding
two whorls. Color a uniform, dull pale olivaceous-green, somewhat
lighter in young shells. Outer lip thin and simple, strongly sinuate in
profile, retracted near the suture, produced below. Aperture elongate-
ovate, narrowly edged with black all around, not at all produced at
the base. Columella slightly thickened; concave parietal wall thinly
glazed. Sculpture of very fine, sinuous growth-lines; no trace of
rippled spiral striation. Radula and operculum as in other species of
the genus.

Measurements:

Length Width Aperture

11.7 5.4 5.4 x 2.0 mm. Holotype

11.7 5.6 5.0x2.1 Paratype

12.1 6.1 5.5x2.3

These specimens all with truncate summit.

Holotype. Mus. Comp. Zool. No. 112268, Estuary of the Congo
River on the right bank, near Matadi, Belgian Congo. Dr. M. Wan-
son collector, 1938. Additional paratypes, Mus. Comp. Zool. No.
112269, from the same locality and collector. These snails live on rocks
immersed in swiftly running water.

This is one of the smallest adult Potadoma known to us.1
In general shape and smoothness, as well as in the absence of minute
spiral sculpture, P. wansoni comes nearest P. ignobilis (Thiele), of the
northeastern Belgian Congo; but the latter reaches at least twice the
size, so that no confusion is possible.

1 Potadoma mayumbensis Thiele (1928) was described from young specimens, 9 mm. long and
5 mm. wide. There is an adu It of this species at the Mus. Comp. Zool., reaching 26 mm. in height

10 bulletin: museum op comparative zoology

POTADOMA AGGLUTINANS Spec. nOV.

Plate 1, figs. 1-8 and 10

Normal adult shape unknown, as all adult shells are more or less
deformed through corrosion and adherence to rocks, dirt or other
shells of the same species. Our smallest shell, 2.3 mm. long and 1.6
mm. in greatest width, consists of 4 whorls, with the apex apparently
little or not corroded; it is turrited, of normal Potadoma shape, and
strongly ribbed spirally. On the fourth whorl there are four ribs at the
periphery, the upper one very prominent, the others gradually weaker.
The largest normal shells seen (PI. 1, fig. 5) are about 8 mm. long and
3.5 to 4 mm. in greatest width (at the last whorl) ; they are also tur-
rited, but less slender than the younger shells, only four whorls being
preserved; these are sculptured spirally with more numerous ribs,
three very strong at the periphery, five or six weaker ones spaced over
the basal third, and sometimes a very weak one midway between the
suture and the periphery; in addition, there is a very fine axial sculp-
ture of close set, somewhat curved growth-striae. All shells over 8
mm. long (and many smaller ones) are deformed, but they retain at
least part of the sculpture of the normal younger stage. The holotype
(PI. 1, fig. 1) is relatively little deformed, being yet distinctly turrited;
but the sutures are very deep and there is some adherence in spots to
foreign matter. In other shells (PI. 1, figs. 5-8 and 10) the later whorls
bulge out abruptly and show the flattened areas cemented to stones
or other shells. When the earlier and normal whorls persist, they often
slant from the later, abnormal whorls. Aperture usually subcircular or
slightly higher than wide, not produced at the base. Outer lip thin, not
expanded, somewhat arched forward at the periphery. Parietal wall
thinly glazed. Columella uniformly concave, with a thin fold extend-
ing over the narrowly rimate or nearly closed umbilicus. Color a dull
pale olivaceous-brown. Radula (Text fig. 1C) much like that of
Potadoma ponthiervillensis. Operculum (Text fig. IE) of the usual
Potadoma type.

Measurements of adults shells :

Preserved

Length

Width

Whorls

12.5

7.5 mm.

Holotype

11.5

6.3

Paratype

11.0

6.6

10.6

7.1

8.6

6.3

BEQUAERT AND CLENCH: RHEOPHILOUS MOLLUSK FAUNA 11

Holotype. Mus. Comp. Zool. No. 112267, Estuary of the Congo
River, on the left bank at Kala-Kala near Matadi, Belgian Congo.
Dr. M. Wanson collector, 1938. Additional paratypes, Mus. Comp.
Zool. No. 112266, from the same locality and collector.

These snails were found in crevices of rocks immersed in very
swiftly running water. The peculiar habitat explains the cementing of
the shells to neighboring objects and the consequent irregularity in
growth and shape. So far as we know, both features are unique, not
only among the Melanians, but among all known fresh-water Gastro-
pods. The young are free and move about in the crevices in which they
hide; but the strength of the current obviously induces them to lodge
in narrow spaces. As growth proceeds, new shell material deposited at
the edge of the outer lip not only takes the shape of the wall of the
crevice (or any object pressing against the aperture), but adheres to
it. Once the aperture is partly attached in this manner, the snail will
usually be unable to move, so that continued growth will cement it
further and enhance the deformation. If young, free snails were re-
moved from crevices and kept alive for some length of time in a less
restricted environment, possibly "normal" adult shells might be
obtained.

ADDITIONS TO THE LOWER CONGO FAUNA

The following species of land and fresh-water mollusks, collected by
Dr. M. Wanson outside the Estuary, contain some interesting new
records for the Belgian Congo. The sequence is that of Pilsbry and
Bequaert's two volumes on the mollusks of the Belgian Congo (1919
and 1927).

Achatina bandeirana Morelet. Matadi.

Achatina tincta Reeve. Matadi.

Achatina pfeifferi eugrapta Pilsbry. Matadi.

SUBULINA (NOTHAPALUS) PAUCISPIRA MUKONGO SubspeC nOV.

Plate 2, figs. 6-8

Agreeing in most characters of the shell, including aperture and
sculpture, with S. paucispira v. Martens, but consistently shorter and
wider. Body-whorl nearly two-thirds and aperture a little less than
one-third of the total length. The greatest width is about mid-length
of the shell, not in the lower third as in typical paucispira. The shape

Length

Width

16.8

5.6

17.0

5.8

15.6

5.3

''horls

6
6
6

Holotype

Paratype

<«

12 bulletin: museum of comparative zoology

of the summit is as in paucispira, being very much narrower than in
Subulina (Nothapalus) laevigata (Pfeiffer), of West Africa.

Measurements:

Aperture
6.4 x 2.7 mm.
6.8x3.1
6.3 x 3.0

Holotype. Mus. Comp. Zool. No. 112302, Luadi-Soyo near Matadi,

Belgian Congo. Dr. M. Wanson collector, 1938. Additional paratypes

Mus. Comp. Zool. No. 112304, from the same locality and collector.

We have reached the conclusion that S. paucispira xanthophaes

Pilsbry is not separable as a race from typical S. paucispira.

Pseudoglessula strigosa (Morelet). Luadi-Soyo near Matadi.

Opcas gracile (Hutton). Matadi.

Ptychotrema bequaerti thysvillensc Pilsbry. Luadi-Soyo near Matadi.

Ptychotrema (Ptychotrema) pupaeforme (Morelet). Ennea pupae-
f or mis Morelet, 1866, Jl. de Conchyl., 14, p. 154; 1867, Voy. Welwitsch,
Moll. Terr. Fluv., (1868), p. 82, PI. II, fig. 6; Mt. Cungulangulo,
2,000 ft., District of Gulungo Alto, Angola. — One specimen from
Luadi-Soyo near Matadi. We have compared it with two of Morelet's
cotypes and find the Lower Congo snail identical. E. v. Martens
(1876, Monatsber. Ak. Wiss. Berlin, p. 268) suggested that Ennea
calameli Jousseaume (1872, Rev. Mag. Zool., (2), 23, p. 12, PI. II, figs.
3-4 ; Novo Redondo, Angola) was possibly P. pupaeforme. This state-
ment must have been made by an oversight, since E. calameli is too
small for pupaeforme and clearly the same as P. ringiculum (Morelet).

Gulella monodon (Morelet). Ennea monodon Morelet, 1873, Jl. de
Conchyl., 21, p. 330 (Gaboon); Connolly, 1929, Ann. Mag. Nat. Hist.,
(10), 3, p. 167, fig. 1 (type). One specimen from Luadi-Soyo near
Matadi. It agrees well with Connolly's figure of the type, as well as
with many specimens which the senior author collected in ten locali-
ties in Liberia.

Thapsia zambiensis Pilsbry. Luadi-Soyo near Matadi.

Gymnarion sowerbyanus (Pfeiffer). Luadi-Soyo near Matadi.

Succinea congoensis Pilsbry. Matadi.

Lymnaea natalensis succinoides Morelet. Lukunga River at Kim-
pese. This form is intermediate between typical natalensis, of South
Africa, and var. undussumae v. Martens, of the Upper Congo.

Gyraulus misellus (Morelet). Matadi ("barrage Coco-Sambana").
Dartevelle and Schwetz have recently reported this species from the

BEQUAERT AND CLENCH: RHEOPHILOUS MOLLUSK FAUNA 13

island of Mateba in the Congo Estuary (1937, Ann. Soc. Zool. Belgi-
que, 68, p. 53).

Biomphalaria salinarum (Morelet). Lukunga River at Kimpese.
This is probably no more than a race of B. pfeifferi (Krauss) of South
Africa.

There is much to be said in favor of Major Connolly's recent treat-
ment of all larger African Planorbinae as one genus, Biomphalaria
(1939, Ann. South Afr. Mus., 33, p. 484). These have also been placed
or distributed variously in Planorbis (sensu sir.), Planorbula and
Afro planorbis. Probably most, if not all of these African snails may
sometimes produce internal folds or teeth in the young shells, like the
North American Planorbula; but, as F. C. Baker (1940, The Nautilus,
53, p. 106) points out, these lamellae have arisen independently in
otherwise unrelated groups. Several of the young salinarum from the
Lukunga River, up to 5 or 6 mm. in greatest diameter, show the inter-
nal barrier of lamellar teeth.

Bulinus senegalensis crystallinus (Morelet). Matadi ("barrage
Coco-Sambana"), very common on dead leaves. The specimens were
compared with Morelet's cotypes.

Lanistes congicus O. Boettger. Lukungu River at Kimpese.

Caelatura bomae Pilsbry. Matadi, many specimens on the banks of
the Congo, in quiet water. We have it also from Ango-Ango, below
Matadi. In our opinion, this is a distinct species, not a race of C.
stagnorum (as originally described) nor a synonym of C. bourguignati
(de Rochebrune) as claimed by Haas (1936, Abh. Senckenb. Naturf.
Ges., 431, p. 66). It differs consistently from bourguignati, not only
in shape, but also in the presence of many fine, close set, regular
radiating corrugations behind the beaks, over nearly the upper third
of the valves. In C. bourguignati this area bears only a few, irregular,
much spaced, not radiating folds.
On the other hand, we agree with Haas (loc. cit.) that Unio stagnorum
Dautzenberg and Caelatura rotula Pilsbry and Bequaert are synonyms
of Caelatura bourguignati (de Rochebrune).

EXPLANATION OF PLATES

PLATE 1

equaebt and Clench — Rheophilous Mollusk Fauna

PLATE 1

Fig. 1. Potadoma agglutinans Bequaert and Clench; holotype, X 5.
Figs. 2 and 3. Potadoma agglutinans B. and C; young paratypes, X 4.
Figs. 4-8 and 10. Potadoma agglutinans B. and C; paratypes, X 4.
Figs. 9 and 12. Potadoma wansoni Bequaert and Clench; paratypes, X 3.
Fig. 11. Potadoma wansoni B. and C; holotype, X 3.

BULL. MUS. COMP. ZOOL. Bequaert and Clench: Rheophilous Mollusk Fauna. Plate!.

PLATE 2

Beqimeht ano Clench — Rlieciphilous Mollu.sk Fnuna

PLATE 2

Figs. 1-3. Septariellina congolensis Bequaert and Clench; X 8.

Fig. 4. Hydrobia rheophila Bequaert and Clench; holotype, X 10.

Fig. 5. Lobogenes zairensis Bequaert and Clench; holotype, X 10.

Figs. 6-7. Subulina (Nothapalus) paucispira mukongo Bequaert and Clench;

paratypes, X 3.
Fig. 8. Subulina paucispira mukongo B. and C; holotype, X 3.
Fig. 9. Valvatorbis mauritii Bequaert and Clench; holotype, X 10.
Fig. 10. Hydrobia plena Bequaert and Clench; holotype, X 10.
Figs. 11-12. Lobogenes schoutedeni Bequaert and Clench; holotype, X 10.

!ULL. MUS. COMP. ZOOL. Bequaert and Clench: Rheophilous Mollusk Fauna. Plate2.

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE

Vol. LXXXVIII. No. 2

MILLIPEDS COLLECTED IN PUERTO RICO AND THE
DOMINICAN REPUBLIC BY DR. P. J. DARLINGTON

IN 1938

By H. F. Loomis

Bureau of Plant Industry
U. S. Department of Agriculture

CAMBRIDGE, MASS., U. S. A.

PRINTED FOR THE MUSEUM

May, 1941

jj»' ' Zoology

'MAY 16 1941

ilBRA^

No. 2. — Millipeds Collected in Puerto Rico and the Dominican Republic
By Dr. P. J. Darlington in 1938 l

By H. F. Loomis

During the summer of 1938 Dr. P. J. Darlington, of the Museum of
Comparative Zoology, Cambridge, Massachusetts, gathered much
zoological material in Puerto Rico and the Dominican Republic which
included a large number of very interesting millipeds, later sent to me
for identification, and now forming the basis for this paper.

The Puerto Rican collection of millipeds contained 11 species, and
has been exceeded in number only by that of 12 species by Prof. W. M.
Wheeler in 1906, and reported on by Silvestri2. Four species of the
Darlington collection are here described as new, one being made the
type of an unusual new genus of the family Stemmiulidae. With these
additions, thirty species now have been recorded from Puerto Rico,
but the identity of five is in doubt and cannot be settled until much
more collecting and study have been done.

In the Dominican Republic 35 species were found, and these repre-
sent the first extensive collection ever to come from that country,
where previously only six species had been known, one of those also
being reported from the adjacent Republic of Haiti. In the Darlington
collection are six previously described species, one of which was already
known from the Dominican Republic, the remaining five being newly
discovered there but previously known from Haiti. Twenty-eight of
the species, apparently new to science, are described in the following
pages, and among them are the types of eight new genera. A single
remaining form was represented by inadequate material, impossible
to identify specifically.

The Dominican portion of the collection is remarkable in the number
of new species of Prostemmiulus and Micros pirobol us it contains; and
in the many new monotypic genera of the order Merocheta it has been
necessary to erect to fit the animals into the current system of classifi-
cation. These genera are indicative of a large undiscovered milliped
fauna, and future collections in other parts of the country hardly can
fail to add species to some of them. Great localization of milliped spe-
cies already has been noted in Cuba3, and especially in the Republic of
Haiti4, whence 108 species have been reported. The Dominican Re-
public, with double the area of Haiti, remains almost wholly unknown,

1 Published with the aid of a special gift from Mr. George R. Agassiz.

2 Bull. Amer. Mus. Nat. Hist., Vol. 24, pp. 563-578, 1908.

3 Bull. Mus. Comp. Zool., Vol. 82, No. 6, pp. 427-480, 1938.

4 Bull. Mus. Comp. Zool., Vol. 80, No. 1, pp. 3-191, 1936.

18 bulletin: museum of comparative zoology

as far as millipeds are concerned, in spite of the present collection, and
a milliped fauna comparable to that of Haiti may be expected. An-
other unusual feature of this collection, for which no explanation is
offered, is its complete lack of species of Cyclodesmus. Thirteen species
of this genus are known from Haiti, and its distribution certainly must
extend into many parts of the Dominican Republic.

Combining the faunas of the Haitian and Dominican Republics,
141 species of millipeds now are credited to the Island of Hispaniola,
but this number will be increased with each new collection from there,
especially if made in hitherto unvisited regions.

Types, paratypes and all other specimens in the collection are in the
Museum of Comparative Zoology.

GLOMERIDESMIDAE

Glomeridesmus pectinatus spec. nov.

The type, and another male and a 20-segmented female from El
Yunque, Puerto Rico, May 1938.

Diagnosis. Intermediate in size between G. marmorens Pocock and
G. trinidadensis Loomis but readily distinguished from the former by
the more rectangular shape of the pleurae which have a comb-like
border of fine setae along the posterior margin in contrast to the simple
margin found in the latter species.

Description. Length from 8 to 9.5 mm., width to 2 mm.

Fig. 1. Glomeridesmus pectinatus. Last two joints of last leg of male.

Color of head and antennae dark; first segment dark with a median
W-shaped figure of white with the free ends of the figure forward;
laterad of this figure, in the outer angle, is a small white spot ; ensuing
segments dark with a large white spot in front a third of the way to the
lateral margin, the spots much smaller on the caudal third of the body;
segments 2 to 4 or 5 with an additional smaller white spot in front on

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 19

each side at the median line; sides of all segments white or colorless for
a distance above the lateral margin; legs and ventral surfaces colorless;
one specimen, less fully colored than the others, has the head, first seg-
ment, and the outer third of the other segments colorless.

Head with the entire clypeal region swollen, forming a broad trans-
verse elevation, on the posterior limits of which many small erect
setae are evident, especially near the sides of the head, a few larger
setae farther forward; pit behind the antennal socket much larger
than the socket and much more nearly circular than in G. marmoreus
as shown by Pocock's illustration.

First segment much exceeding the width of the head, in the propor-
tion of five to four.

Caudal segments with each posterior corner produced into a small
acute tooth.

Pleurae rectangular, the exposed portion broader than long, surface
smooth, the posterior margin fringed with a comb of fine setae except
near the outer angle.

Coxal joints of the legs broader and less angular than in G. mar-
moreus; the posterior margin simple, neither toothed nor setose.

Last leg of male with the two outer joints as shown in figure 1, the
last joint apparently with a tiny peg-like claw.

Chamberlin reported1, without mention of size, color, or structural
characters, a single specimen of Glomeridesmus from El Yunque,
which he referred to G. eoncolor, described from Haiti as being from
4 to 6.5 mm. long. It now appears more probable that this specimen
should be placed under G. pectinatus rather than under the Haitian
species.

SIPHONOPHORIDAE

Siphonophora platops spec. nov.

Two males (one the type) and two females from rain forest near
Valle Nuevo, 6,000 feet elevation, Cordillera Central, Dominican Re-
public, August 1938.

Diagnosis. A more robust species than any previously known from
the island, its proportions being quite similar to S. robusta Chamberlin
of Jamaica from which it differs in the much shorter antennae.

Description. Body stout, the dorsum strongly convex, densely
velvety pubescent; color light yellow in alcohol but probably white in

i Proc. U. S. Nat. Mus, vol. 61, art. 10, p. 1, 1922.

bulletin: museum of comparative zoology

life; segments 44 and 45 for the males, 55 and 61 for the females;
largest male 11 mm. long and 1 mm. wide, largest female 20 mm. long
and 1.5 mm. wide.

Head rather flattened from the base of the slightly decurved beak
to the moderately swollen vertex and with the pubescence longer but
not. as abundant as that on the segments or on joints 5 and 6 of the
antennae. Head and antennae shown in figure 2, a, the antennal

Fig. 2. Siphonophora platops. a, Head and antenna, vertical view; 6, Seg-
ments 1 to 3 in vertical view, the head deflexed; c, Gonopods, outer lateral view.

sockets far around on the ventro-lateral surface, the first joint and
basal portion of the second one not visible from above; antennae
strongly clavate, the sixth joint longest and broadest, the seventh
joint small and exceedingly short, scarcely projecting beyond the
sixth joint and only half as wide.

First segment with sides sharply diverging from in front, over twice
as wide behind as the head and almost as wide as the greatest diameter
of the body (Fig. 2, 6).

Gonopods shown in figure 2, c, the apex of the stout anterior pair
bent sharply inward and backward.

Legs in advance of the gonopods almost similar in size, the first pair
slightly shorter than the eighth pair.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

STEMMIULIDAE

Prostemmiulus quintarius spec. nov.

The male type and four females from the Maricao Forest, about
2,500 feet elevation, western Puerto Rico, June 2 and 3, 1938.

Diagnosis. Insofar as is known this is the only species having
pleural lobes on the fifth segment of the male. The gonopods associate
the species with the much smaller P. wheeleri Silvestri, from the small
island of Culebra, 20 miles east of Puerto Rico.

Description. Body strongly compressed laterally; moderately subu-
late, the last 15 segments narrowing gradually; number of segments

Fig. 3. Prostemmiulus quintarius. a, Segment from middle of body, dorsal
view; b, Gonopods, anterior view; c, Gonopods, posterior view.

46 to 49; length 27 to 31 mm; males obviously more slender than
females.

Color in alcohol dark slate gray, an interrupted light median line
on the dorsum consisting of a very narrow line on the prozonite of each
segment, part of which shows through the transparent posterior por-
tion of the overlapping metazonite; a small light spot or mottled area
at each pore with another half way between it and the base of the legs;
antennae with joints 1, 6 and 7 light in color.

22 bulletin: museum of comparative zoology

Head with sulcus of vertex very faint or entirely lacking; antennae
long and slender, the joints of about equal width ; joint 2 longest, joint
3 next in length with joints 4 and 5 slightly shorter, subequal, joint 6
two-thirds as long as joint 5; ocelli of moderate size, the posterior one
half as wide as the antennal socket, the anterior one from half to
nearly as large as the other; mandibulary stipe pointed in front, the
upper and lower margin with a raised rim, stipe of female wider than
that of male.

First segment with two or sometimes three primary striae, below
which, on the subventral surface, shorter secondary striae are present.

On ensuing segments the oblique striae are sharply marked, first
reaching the broad, impressed median sulcus on segment 9 or 10, the
striae on a mid-body segment are shown in figure 3, a, as typical of the
genus and for comparison with the striae of the new Puerto Rican
genus Scoliogmus; notch at posterior end of median sulcus deep but
narrow; serration of the posterior margin above the feet strong and
acute.

Preanal scale broadly rounded-truncate behind.

Gonopods as shown in figure 3, b and c.

Second legs of male much like those of P. wheeleri.

Male with pleurae of segment 3 simple, not produced; pleurae of
segment 4 each carried inward as a narrow, mesially rounded lobe,
with a raised rim around the entire margin; pleurae of segment 5
each carried inward in a shorter and broader lobe; pleurae of ensuing
segments normal.

Prostemmiulus iuloides spec. nov.

Three males (1 the type) and 9 females from Pico del Yaque, Loma
Rucilla, 8,000 to 10,000 feet elevation, Dominican Republic, June
1938. Additional specimens collected the same month from between
5,000 and 8,000 feet elevation from Loma Rucilla and mountains
north, Cordillera Central, Dominican Republic.

Diagnosis. Aside from the gonopods, which are of the greatest im-
portance in differentiating the species of this genus, this species shares
with only P. scaurus and the much smaller P. gracilipes the peculiar
character of having the side of the first segment of the male, below the
lateral stria, continuous in contour with the surface above the stria,
not at all bent under as is usual in the female and in both sexes of other
species. From P. scaurus it differs in the caudaliy bent first legs of the
male and the unmodified third legs.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

Description. Body long and slender, much like Parakdus, gradually
narrowing from behind the middle of the body; the first segment of the
males enlarged, wider posteriorly than remainder of body; number of
segments 48 to 49, length 25 to 26 mm.

Fig. 4. Prostemmiulus iuloides. a, Segment 1, lateral view from slightly in
front; b, Gonopods, anterior view; c, Gonopods, posterior view; d, Second leg
of male, anterior view.

Color in alcohol rather dark brown except the first three segments,
which are slightly lighter, especially in the males, and the light brown
median line; the usual light spots on the side of each segment are
faintly indicated.

Head with small eyes, the anterior ocellus about half as large as the
posterior which is less than half as broad as the antennal socket;
antennae with joint 2 longest; joints 3, 4, and 5 subequal; joint 6
two-thirds as long as joint 5; furrow of the vertex short but well im-

24 bulletin: museum of comparative zoology

pressed; mandibulary stipe of male rounded-truncate in front, ventral
fourth depressed above the margining rim, the remainder of the surface
raised in a sharply angular longitudinal swelling; stipe of female wider
but less sharply swollen and more pointed in front.

First segment as shown in figure 4, a, with numerous fine setae
scattered over the surface; in the females there is a single primary
stria along the lower anterior margin, the surface below it, bearing the
secondary striae, limited in area and somewhat bent under, partially
hiding the striae in lateral view; in the male the surface bearing the
secondary striae is much larger and is not bent under but is continuous
with the surface above the primary stria; the intervals between the
secondary striae bear a single series of strongly clavate setae similar
to those on the first male legs; clavate setae also are present on the
lateral intervals of segments 2 and 3 of the males and five scattered
setae are present on the dorso-lateral surface of segment 2.

On ensuing segments the dorsal striae are pronounced, first reaching
the middle of the dorsum on segment 10, 11, or 12; median sulcus well
developed, the notch at its posterior end rather deep but narrow, in-
conspicuous; serrations of the lower posterior margin of the segments
fine, the lower striations little stronger than those on the dorsum.

Preanal scale evenly rounded behind.

Gonopods as shown in figure 4, b and c.

Male with pleurae of the third segment produced inward; the long,
thickened inner margin proceeding obliquely inward from front to
back, the posterior corner acute, slightly raised ; fourth pleura extend-
ing further inward, greatly narrowed and bent sharply away from body
into an acute lobe mesad of the pleura of the third segment.

First male legs curving caudad, crassate, nearly as long as the nor-
mal legs, covered with clavate setae; second legs as shown in figure 4,
d; third and fourth legs decreasingly crassate but increasing in length
toward that of the ensuing legs; setae of the third legs clavate, those
of the fourth legs normal.

Prostemmiulus scaurus spec. nov.

One male (type) and two females, the largest of which lacks the
head and at least the first four segments, collected at Valle Nuevo,
southeast of Constanza, elevation about 7,000 feet, Cordillera Central,
Dominican Republic, August 1938.

Diagnosis. Several characters, including the form of the first seg-
ment of the male and the presence on it and several ensuing segments

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

of clavate setae, might be thought to associate this species with P.
iuloides but the gonopods belie close relationship; greatly enlarged
third male legs are not found in other members of the island fauna
except the much smaller Haitian P. clavipes Loomis which also shows
other differences.

Fig. 5. Prostemmiulus scaunis. a, Gonopods, anterior view; b, Gonopods,
posterior view; c, Second leg of male, anterior view; d, Third leg and sternal
plate of male.

Description. Body relatively slender and parallel-sided as in P.
iuloides, resembling a Paraiulus; male 25 mm. long, the broken female
at least 30 mm. long in life; number of segments 46 to 47.

Color brown in alcohol, a continuous broad light median line the
length of the body, the line slightly broader on the anterior portion of
each segment, its sides less definite than with most species; at each

26 bulletin: museum of comparative zoology

pore is a medium sized white spot below which the color becomes
lighter brown, finely mottled with white, the region near the legs white.
Head with the eyes unusually small, the posterior ocellus less than
half as wide as the antennal socket, the anterior ocellus scarcely half
as wide as the posterior one; antennae of usual proportions, with joint
2 longest, the next three joints subequal in length, joint 6 two-thirds
as long as joint 5. Mandibulary stipe of male much as in P. iuloides
but a little narrower, the front more acute; in the female the stipe is
squarely truncate in front for a short distance.

First segment with fewer scattered fine dorsal setae than in P.
iuloides but the lateral striate regions of the two sexes differ as in that
species, the surface bearing the secondary striae not at all bent under
in the male but somewhat so in the female; the interstrial costae of the
male scaurus have clavate setae in single series as also are found be-
tween the lower striae of the next two segments, there being seven
clavately setose interstrial series on segment 3.

Ensuing segments with oblique striae well defined, first approaching
the median sulcus on segment 8 or 9 ; notch at posterior end of median
sulcus very small, short, and narrow; serrations of the lower posterior
margin faintly obvious only in the pleural region.

Preanal scale evenly rounded behind.

Gonopods as shown in figure 5, a and b.

First male legs slightly heavier than the fourth pair; second legs as
shown in figure 5, c; third legs with the four outer joints greatly swollen
as shown in figure 5, d.

Pleurae of third segment of male produced inward only a little, the
inner edge turned slightly away from the body; pleurae of fourth seg-
ment extending inward well beyond those of segment 3 and very
abruptly turned upward, away from the body, into acute lobes.

Prostemmiulus gracilipes spec. nov.

One male (type) and six females from rain forest near Valle Nuevo,
about 6,000 feet elevation, Cordillera Central, Dominican Republic,
August 1938.

Diagnosis. The small size of the body with its interrupted light
median line, coupled with the form of the pleurae of the third and fourth
segments of the male, outwardly distinguish this species, although the
gonopods are more definitely characteristic.

Description. Body somewhat subulate, gradually tapering back-
ward from the middle; length 11 to 15 mm; number of segments 38 to

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

45. Color much as in P. sulcatus, the median spot on the prozonites is
broader and the lower spot on the sides is as large and distinct as that
at each pore.

Fig. 6. Prostemmiulus gracilipes. a, Second leg of male, anterior view; b,
Second leg of male, outer, lateral view; c, Gonopods, anterior view; d, Gonopods
posterior view.

Head with the antennae rather short and stout; the second joint
not greatly longer than the ensuing three subequal joints; the sixth
joint fully two-thirds as long as joint 5; anterior ocellus minute, dis-
tinctly less than half the diameter of the posterior ocellus which itself

28 bulletin: museum of comparative zoology

is unusually small, not equalling half the width of the antennal socket;
mandibulary stipe of the male narrower and less inflated than that of
the female, both acute in front.

First segment with setae scattered over the surface, those of the male
more numerous than of the females but possibly some have been lost
from the latter; side of segment of the female, in lateral view, sharply
bent under beneath the single margining stria; in the male the surface
bearing the secondary striae is not bent under but is continuous in con-
. tour with the surface bearing the primary stria.

Ensuing segments with striae quite well defined, first reaching the
median line of the dorsum on the anterior portion of the metazonite
of segment 14 or 15; median sulcus moderately broad and deep, a
short narrow nick in the posterior margin at its end; serrations of the
lower margins fine but distinct.

Preanal scale rounded-truncate behind.

Male with the inner margin of the pleurae of segments 2 and 3 much
longer than on other segments, rounded; the pleurae of segment 3 not
specially produced inward or elevated; each pleura of segment 4 in-
wardly produced into a parallel-sided lobe with only the margin of the
inner, rounded end slightly raised.

First male legs not notably stouter than the third or ensuing pairs ;
second legs as shown in figure 6, a and b, the outer joint unusually long
and slender.

Gonopods as shown in figure 6, c and d.

Prostemmiulus tridigitatus spec. nov.

Two males (one the type) and a female from Mt. Quita Espuela,
between 1,000 and 3,000 feet elevation, July 1938; in the same month
two males and two females collected between 3,000 and 4,000 feet
elevation, Mt. Diego de Ocampo, Northern Range, Dominican Re-
public.

Diagnosis. The gonopods, with the three finger-like processes at the
posterior apex of each inner armature, are the most distinctive char-
acter of this species but an outward peculiarity is the faint impression
of the dorsal striae and generally their failure to approach close to the
median sulcus of the segments.

Description. Body strongly subulate, narrowing sharply from the
middle to the very narrow last segment; the larger specimens all about
25 mm. long, the females 2.2 mm. wide and 2.5 mm. high, males not so
stout or so noticeably compressed; segments 43 to 47.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 29

Color in alcohol brown with light markings; specimens from type
locality with anterior two-thirds of first segment dark, the posterior
third and segments 2 and 3 and the anterior portion of the fourth
segment very light; in specimens from the other locality these seg-
ments are more similar in coloration to those succeeding, the first of
which have a white median fascia, wide on the prozonites and narrow
on the metazonites, but this becoming very broad on the mid-body
segments although even there it is widest on the prozonite and shows
through the metazonite of the preceding segment; there is a small white
spot near each pore and a still larger white spot farther down the side
below which the brown color begins to fade, the lower sides white.

Fig. 7. Prostemmiulus tridigitatus. a, Gonopods, anterior view; b, Gono-
pods, posterior view.

Head with the antennae long and slender; joint 2 longest; joints 3,
4, and 5 subequal in length; joint 6 two-thirds as long as joint 5; ocelli
very convex; the posterior one two-thirds the diameter of the antennal
socket; the anterior one half the size of the posterior; the female from
the type locality has, on one side of the head, a third ocellus the size of
the anterior one and mesad of the posterior ocellus ; mandibulary stipe
of the male rounded-acute in front ; a raised rim on all sides, with the
inner surface evenly inflated; stipe of female larger, relatively broader
and slightly more convex.

First segment with a few scattered hairs on the dorsum; the side of
the segment, below the single primary stria, bent under in both sexes,
hiding the subventral or secondary stria from lateral view.

On ensuing segments the oblique striae are very fine and weakly
impressed and do not extend above the pores until after segment 10 or

30 bulletin: museum of comparative zoology

even further back, and thereafter four striae, at most, are seen above
the pore on either side, the innermost seldom approaching close to the
median sulcus even toward the anterior limit of the metazonites ; dorsal
surface with coarse, longitudinal, slightly undulated aciculations;
lower posterior margin of segments with coarse serrations quite re-
markable in view of the weak striae.

Preanal scale rounded truncate behind.

Gonopods as shown in figure 7, a and b.

Second male legs much like those of P. scaurus, the outer joint
turned forward sharply; third legs normal.

Pleurae of third segment of male scarcely at all produced mesad, the
inner posterior corner slightly raised ; each pleura of the fourth segment
strongly produced inward and forward, the mesial half bent sharply
upward, away from the body, into an acute lobe.

Prostemmiulus sulcatus spec. nov.

The single male (type) collected between 1,000 and 3,000 feet ele-
vation, Mt. Quita Espuela, Dominican Republic, July 1938.

Diagnosis. The color pattern is distinctive if that of the single
specimen is typical of the species. The gonopods are of a form not to
be confused with any other species although showing relationship to
the Haitian P. hctcrops Loomis.

Description. Body subulate, tapering toward the back from in front
of the middle of the body; length 20 mm; number of segments 42.

Color generally dark, the usual light median fascia reduced to an
elongate spot on the prozonite of each segment but showing through
the translucent integument of the metazonite of the preceding seg-
ment, a larger white spot at each pore with a smaller, less definite spot
between it and the base of the legs.

Head with the eyes small, the anterior ocellus half as large as the
posterior one which is much smaller than the antennal socket; anten-
nae with joint 2 longest; joints 3, 4, and 5 subequal; joint 6 over two-
thirds as long as joint 5, mandibulary stipe surrounded by a raised
rim; the median portion strongly inflated; anterior margin very
obliquely truncated, the truncation almost continuous with the lower
margin.

First segment with setae sparsely scattered over the surface; a
single stria along the anterior margin each side with shorter, less pro-
nounced striae on the sharply bent under, sub ventral surface.

Oblique striae of ensuing segments well defined, first reaching the

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 31

middle of the dorsum at the anterior part of segment 11 or 12; median
sulcus broadly and deeply impressed on the posterior part of each
metazonite, the notch in the margin at its end shallow and very
conspicuous; serrations along the lower posterior margin of the seg-
ments weak.

Fig. 8. Prostemmiulus sulcatum, a, Gonopods, anterior view; b, Gonopods,
posterior view.

Preanal scale with a quite definite truncation at apex, not evenly
rounded.

Gonopods as shown in figure 8, a and b.

First male legs very little stouter than the normal third legs.

Third pleura not produced mesially but inwardly elevated, espe-
cially at the posterior corner; fourth pleura produced inward, ending
in a broadly rounded, slightly upturned, apex.

Prostemmiulus setosus spec. nov.

A single female from Valle Nuevo, southeast of Constanza, eleva-
tion about 7,000 feet, Cordillera Central, Dominican Republic,
August 1938.

Diagnosis. In this genus, founding of species on female types cannot
be justified unless a unique distinguishing character or a peculiar
combination of several characters is present. The scattered setae on
the dorsum of the above specimen is a character which fully warrants
erection of a new species as it is found in no other known member of

32 bulletin: museum of comparative zoology

the family, the usual arrangements of setae being a single series along
the posterior margin of each segment.

Description. Body mostly parallel-sided, the posterior third narrow-
ing slowly to the moderately broad last segment; not conspicuously
compressed laterally; length 17 mm; number of segments 46.

Color brown with an interrupted median fascia present as a large
white spot on the prozonite but showing through the transparent
posterior third of the overlapping metazonite; a tiny white spot at
each pore and a larger spot half way to the legs; the color below this
spot much lighter brown.

Head with a very short, faint, median sulcus on the crest of the ver-
tex; ocelli greatly differentiated in size, the posterior one half the
diameter of the antennal socket, the anterior one minute, only a
fourth as broad as the other ocellus; antennae with joint 2 much
longer than any other, the subequal joints 3 and 5 next in length with
joint 4 slightly shorter than either; antennae widest at apex of joint 5;
mandibulary stipe narrowly subtriangular, acute in front, the surface
ascending in a flat plane to an acute elevation along the dorsal third;
lower margining rim prominent.

First segment with a fine primary stria in front, below which are one
or two very fine, short, secondary striae; region above the lower part
of the primary stria subrugose with fine longitudinal wrinkles; surface
scattered with fine erect setae which are more numerous along the
front and back margin than on the central portion.

Ensuing segments with a posterior marginal row of 16 to 22 fine
setae and additional setae sparsely scattered over the remainder of the
dorsal surface of the metazonites; oblique striae well defined, not
reaching the moderately impressed median sulcus until the middle of
the body or beyond ; posterior half of metazonites coarsely marked with
numerous deep longitudinal aciculations; median notch of posterior
margin minute, very short and narrow, serrations of margin very
small but acute, confined to the area beneath the legs.

Preanal scale almost semicircularly rounded behind.

Prostemmiulus sp.

Eight females, ranging in length from 18 mm. to 45 mm. from El
Yunque, Puerto Rico, May 1938.

It is to be regretted that males were missing from this collection, for
without them identification is not possible. The specimens appear to
belong to a single species, in spite of the great variation in size, and

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 33

were only those of intermediate size present they would be considered
as P. compress-us (Karsch). However, since the largest specimens far
exceed any other West Indian Prostemmiulus in length they cannot be
assumed to be P. compressus until more is known about that species.

Scoliogmus genus nov.

Diagnosis. Related to Prostemmiulus but differing in the following
major particulars: the dorsal striae are not straight and oblique but
are strongly curved, the inner one or two showing a tendency to meet
or cross the median sulcus almost transversely; the stipes of the
gnathochilarium have a pronounced outer lobe or flange not found
in other genera of the family and the gonopods are generically distinct.

Description. Body of moderate size, cylindric, much like Paraiulus,
scarcely compressed laterally.

Head with two large, strongly convex ocelli, the anterior one a little
smaller than the one behind which is more convex; antennae long and
rather slender, the second joint longest ; stipes of the gnathochilarium,
in the male at least, with a prominent thickened lobe or flange along
the outer side.

First segment much as in Prostemmiulus.

Ensuing segments with pronounced striae, those on the dorsum
strongly sinuously curved except the uppermost one or two which are
more nearly curved in a simple segment of a circle and tend to meet
or cross the lightly impressed median sulcus at approximately a right
angle; pleural sutures closed as in Prostemmiulus.

Gonopods terminating in a long, curved arm on either side, which,
in repose, remains outside the body.

Anterior legs of the male lacking the strongly clavate setae char-
acteristic of these legs in Prostemmiulus.

Type. S. teres spec. nov.

Scoliogmus teres spec. nov.

A single male from the Maricao Forest, elevation about 2,500 feet,
western Puerto Rico, June 2 and 3, 1938.

Description. Body cylindric, very indefinitely compressed laterally,
except on the caudal segments, the proportions much like Paraiulus,
the posterior end less tapering than usual in Prostemmiulus; length 33
mm.; number of segments 49.

Color in alcohol dark brown with yellow and white markings; median

34 bulletin: museum of comparative zoology

line yellow, continuous from segment 2 to the apex of the last segment ;
on all but a few of the first and last segments the line is much broader
at the anterior margin of each metazonite; prozonites white, in part
showing through the transparent posterior portion of the metazonites;
in front of each pore and slightly above it is a tiny white spot ; pore in
a large area maculate with light spots ; below the pore the color lightens,
becoming almost white ventrally.

Head with the median furrow long, faintly impressed ; antennae long
and moderately slender; joint 2 longest and as slender as joint 3;
joints 3, 4 and 5 decreasing slightly in length; joint 5 over twice as
broad at apex as at base but still not as broad as joint 6 which is over
two-thirds as long; ocelli large and prominent, the posterior one nearly
as broad as the antennal socket and almost hemispheric-ally convex;
anterior ocellus considerably more than half as broad as the posterior
but not quite as convex; mandibulary stipe broad, rounded in front,
strongly inflated, surrounded by a raised rim thickest along the lower
margin ; gnathochilarium peculiar in having the stipes expanded on the
outer side by the presence of a large thickened lobe or flange as shown
in figure 9, a.

First segment (Fig. 9, b) not sharply turned under at the lower angle,
showing two secondary striae in lateral view, with a long primary stria
above them and, considerably above this, another stria proceeding
from the posterior margin half way to the anterior margin.

On ensuing segments the dorsal striae first reach the median sulcus
at segment 7 or 8, the striae well defined, not straight and oblique as in
Prostemmiulus but strongly bent, undulated or merely simply curved,
the innermost one or two meeting or crossing the fine, lightly impressed
median sulcus at nearly a right angle (Fig. 9, c) ; intervals between the
striae with fine longitudinal aciculations; median notch of the posterior
margin small, broad, and shallow, inconspicuous; serrations of the
lower posterior margins very fine, almost obsolete; caudal segments
gradually narrowing and slightly compressed laterally, the last seg-
ment wider than in most species of Prosicmmiulus.

Preanal scale large, broadly rounded behind.

Gonopods as shown in figure 9, d, c, and /, the finely striate, curved,
apical arm of each gonopod protruding outside the body and resting
against the lower side of segment 7.

First male legs with the two basal joints thin but very broad, much
broader than long, the next joint longer and thicker than the corre-
sponding joint of the midbody legs, the three outer joints little heavier
than those of the third or ensuing legs but with a comb of fine setae

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

Fig. 9. Scoliogmus teres, a, Gnathochilarium; b, Segment 1, ventrolateral
view to show the two sub ventral striae below the long lateral stria; c, Three
segments from middle of body, dorsal view; d, Gonopods; anterior view; e,
Gonopods, posterior view; /, Gonopods, anterolateral view from slightly above;
g,h,& i, Second male legs, anterior, posterior, and lateral views, respectively; ;*,
Basal joints of legs and sternal plate of anterior legs of segment 13, male.

36 bulletin: museum of comparative zoology

beneath the last joint; none of the pregenital legs with the clavate
setae usually found in the males of Prostemmiulus.

Second male legs as shown in figure 9, g, h, and i.

Third legs slightly smaller than the fourth and ensuing pairs; the
sternal plate and basal joints of the anterior pair of legs from a mid-
body segment as shown in figure 9, j.

Male with the pleura on either side of segment 3 not at all produced
inward, in fact not reaching inward as far as the pleura of segment 2,
the inner margin broadly rounded and not elevated; pleura of segment
4 produced inward far beyond that of segment 3 into a broad, thick-
ened, inwardly rounded lobe which, if anything, is bent toward the
body rather than away from it as is common in Prostemmiulus.

EPINANNOLENIDAE

Epinannolene curta spec. nov.

Four males, one the type, and three females from El Yunque,
Puerto Rico, May 1938.

Diagnosis. From the shape of the gonopods it appears that this
species is much more closely associated with the Cuban E. biseriatus
Loomis and its Costa Rican and Cocos Island relatives than with any
of the species on the intervening island of Hispaniola. It is distin-
guished from E. biseriatus in having three series of ocelli, and definite
pits throughout the transverse sulcus of the segments.

Description. Body long and slender, the females longer and stouter
than the males; length of largest male 19 mm, largest female 23 mm;
number of segments 46 to 57.

Color in alcohol rather light brown with the posterior third or half
of the metazonites transparent, allowing the color of the prozonites to
show through; on the caudal segments the transparent band is wider
than on the segments farther forward; below and in front of each pore
the color is darker brown than elsewhere.

Head with 15 to 17 ocelli in three series as shown in figure 10, a;
clypeal fovea 2-2, labral setae 7-7 or 8-8; mandibulary stipe of male
narrower and a little less convex than that of the female and with a
higher raised margining rim, anterior end narrowly rounded in both
sexes.

First segment as shown in figure 10, a, with a fine stria along the
front margin below the eye; behind this is a longer stria, its lower half
broader and much more deeply impressed than the upper half, the

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

stria variously curved in different specimens or even on the two sides
of the same one; above the long stria one or two shorter but usually
strongly impressed, curved, striae proceed forward from the posterior
margin.

Ensuing segments with the constriction strongly impressed across
the dorsum as well as on the sides, its bottom occupied by a row of
pits which are small but distinct on the dorsum and increase ma-
terially in size on the ventral half of the body; prozonites and meta-
zonites equally convex, the former brilliantly shining as are the
metazonites above the fine lateral striae, each of which originates from

Fig. 10. Epinanriolene curta. a, Head and first segment, lateral view; b,
Left hand gonopod, anterior view.

the bottom margin of a pit in the constriction and extends back in a
downwardly bowed curve; the striae extend fairly well up on the sides
of segments 2 and 3 but reach their highest point on segment 4, 5, or
6, after which they suddenly decrease so that behind segment 10 or
1 1 only one or two very fine striae may be seen close to the base of the
legs ; at the highest point reached by the striae they are well below the
line of the pores which begin at segment 5.

Last segment large and hood-like, the apical portion not in the least
produced, the posterior margin continuing in the same direct line
throughout its length.

Preanal scale with posterior margin much less rounded than the
anterior margin or very faintly angled at middle; lateral processes of
moderate size.

bulletin: museum of comparative zoology

Gonopods shown in figure 10, b, quite closely resembling those of
E. biseriatus but several differences are apparent.

Ventral margin of segment 7 raised into a lobe on each side much
like that shown in Brolemann's figure for E. pittieri1, the lobes slightly
narrower in relation to their height.

SPIROBOLIDAE

Rhinocricus parous Karsch
Rhinocricus parens Karsch, Zeits. Naturwiss., Ser. 3, Vol. 6, p. 68, 1881.

One female from Maricao Forest, western Puerto Rico, at about
2,500 feet elevation, June 2 and 3, 1938.

Although the size of this specimen is considerably less than given
for the species, it being 64 mm. long, 9 mm. in diameter and having 44
segments, the position of the pores and the scobination of the segments

Fig. 11. Rhinocricus parens.
segment from middle of body.

Repugnatorial pore and sutures on side of

leaves no doubts as to the correctness of identification. The repugna-
torial pores all are well removed below the lateral suture and in front
of the transverse one; the sutures, clearly indicated by light colored
lines in the dark integument, are as indicated in the accompanying
figure 11. Scobinae are present on segments 8 to 12 only and are much

1 Ann. Soc. Ent. Fr., Vol. 72, p. 138, figs. 6, 7, 1903.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 39

broader in proportion to the size of the body than in any other species
with which I am familiar, the broadest measuring over 3 mm. each.
On segments 7 to 11 the posterior margin, above each scobina of the
succeeding segment, is deeply emarginate and the margin is much
thicker than elsewhere.

Seventh joint of the antennae with numerous sense cones.

Rhinocricus hispaniolus spec. nov.

The male type and five paratype males and females from Jarabacoa,
Dominican Republic, August 2, 1938. Additional specimens also
collected in 1938 in the following Dominican localities — Foothills
north of Loma Rucilla, 5,000 to 8,000 feet elevation, Cordillera
Central, June; Mt. Quita Espuela, 1,000 to 3,000 feet elevation, July;
Loma Vieja, southwest of Constanza at about 6,000 feet elevation,
Aug. 7-9; Valle Nueva, near Constanza, between 6,000 and 8,000
feet elevation, Aug.; Constanza to Valle Nueva, between 3,000 and
' 7,000 feet elevation, Aug.

Diagnosis. The unusual gonopods with their curiously formed
median plate, are quite different from those of other members of the
genus, although Cuban species, such as R. sagittatus Loomis and
R. clypcatus Loomis, show strong tendencies toward similar develop-
ment. The median plate is quite similar in structure to that in Leio-
cricus diversipes Loomis but other characters exclude the species from
that genus.

Description. Body from 40 to 50 mm. long and 3.5 to 4.5 mm. in
diameter, the males more slender than the females; number of seg-
ments 47 to 50; color in alcohol dark brown, almost black, with lighter
mottlings on the sides, the posterior margin of each segment colorless-
translucent; head and median area of first segment lighter brown.

Head with 19 to 25 flat, inconspicuous, ocelli in five series forming
a rather small rounded group; median sulcus more or less impressed
from the back of the vertex to the frontof the clypeus; antennae short
and stout, the sixth joint densely pubescent in contrast to joint 5 which
has only a few hairs at the distal end; sense cones four; cardo of man-
dibles broad and flat, with a thick raised rim except along the back
margin, the anterior margin squarely truncated, the lower anterior
corner a right angle or frequently produced into a distinct tooth;
gnathochilarium as shown in figure 12, a, the lower half of the mentum
transversely striate, the lower half of the stipes with several vertical
wrinkles.

bulletin: museum of comparative zoology

First segment broadly and evenly rounded on the sides, the raised
rim short, present only along the lower anterior-lateral margin.

Second segment with a distinct angular shoulder below the limits of
segment 1, the ventral surface concave and coarsely striate.

Fig. 12. Rhinocricus hispaniolus. a, Gnathochilarium; b, Gonopods, an-
terior view; c, Gonopods, posterior view; d, Apical portion of inner gonopod,
same scale as b and c.

Principal body segments have mid-belt a third longer than the other
two belts; the fore-belt usually separated from the mid-belt by a
lightly impressed sulcus in front of which the surface is very faintly
striate with fine lines; seobinae lacking; beginning on segment 2 the
mid- and hind-belts are separated by a strong and abruptly impressed
sulcus in full evidence to the penultimate segment on which it is
variably impressed, the sulcus usually straight in passing behind the
small pores which are somewhat conspicuous from being at the

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 41

bottom of a distinct pit; lateral sulcus seldom in evidence at the line
of the pores; surface of mid- and hind-belt shining and apparently
smooth but close inspection shows tiny fine aciculations; ventral
striae entirely beneath the legs in ventral view, coarser and more
extensive on the hind-belt than on preceding belts; last segment with
apex somewhat produced but not surpassing the anal valves.

Anal valves not strongly convex, the margins appearing a little
raised and sometimes set off by a distinct furrow; preanal scale
triangular, the posterior margins converging in straight lines to the
acutely rounded apex.

Gonopods as shown in figure 12, b, c, and d.

Anterior male legs stouter than those following the gonopods ; joint
2 of the second legs enlarged, inwardly flattened, the posterior corner
produced into a small rounded lobe; coxae of legs 3 to 7 inclusive
produced into thickened conical lobes increasing in size to the seventh
legs.

Ventral surface of seventh segment of male slightly raised into a
thick transverse crest not excavated in front and scarcely concave for
the accommodation of the gonopods.

Rhinocricus arboreus (Saussure)

Spirobolus arboreus Saussure, Linnaea Ent., Vol. 13, p. 331, 1859.

Three females Maricao Forest, western Puerto Rico, June 1938 at
2,500 feet.

Alcimobolus angustipes Loomis
Bull. Mus. Comp. Zool, Vol. 80, No. 1, pp. 57-58, 1936.

Specimens collected in Dominican Republic, July 1938, at Sanchez;
Villa Altagracia; and Mt. Diego de Ocampo, Northern range at an
elevation between 3,000 and 4,000 feet. Additional specimens from
Mt. Is. de Torres, Puerto Plata, in September 1938. In the U. S.
National Museum collection of millipeds are three males and a female
from Rio San Juan, March 1928.

Included in the above are specimens greatly increasing the size
given in the original description, the largest female being 110 mm.
long and 13 mm. in diameter, the largest male 85 mm. long and 9 mm.
in diameter. In all the males the tips of the gonopods project outside
the body cavity, approximating the tips of the coxae of adjacent legs.

42 bulletin: museum of comparative zoology

Trigoniulus lumbricinus (Gerst.)
Spirobolus lumbricinus Gerstaecker, Gliederthier-fauna Sansibar, p. 516, 1873.
Sanchez and vicinity, Dominican Republic, July 1938.

MICROSPIROBOLUS MARMORATUS Silvestri
Bull. Amer. Mus. Nat. Hist., Vol. 24, pp. 571-572, 1908.

A single female from about 2,500 feet elevation, Maricao Forest,
western Puerto Rico, June 2 and 3, 1938, collected with specimens of
M. insularis Silvestri.

This specimen is referred with some doubt to the present species,
although the last segment is typical, and the color, while generally
darker, shows light yellow on the dorsum in spots and may be ascribed
to age. If this specimen is correctly assigned, it is older and larger
than the specimens Silvestri examined, being over 30 mm. long and
with 50 instead of 37 or 38 segments. The segments are without any
indication of transverse constrictions.

Microspirobolus insularis Silvestri
Bull. Amer. Mus. Nat. Hist., Vol. 24, pp. 572-573, 1908.

Two males and several females from Maricao Forest, about 2,500
feet elevation, western Puerto Rico, June 2 and 3, 1938.

A male nearly 40 mm. long exceeds by almost 10 mm. the length
given by Silvestri, but the other male is smaller and the gonopods of
both agree with each other and with Silvestri's figure.

Microspirobolus sigillatus Loomis

Smiths. Misc. Coll., Vol. 89, No. 14, pp. 20-21, 1934.

Three males and three females from Mt. Diego de Ocampo, Northern
Range, 3,000 to 4,000 feet elevation, Dominican Republic, July 1938;
collected with specimens of M. signatus.

Microspirobolus signatus spec. nov.

Two males (1 the type) and two females from Mt. Diego de Ocampo,
Northern Range, between 3,000 and 4,000 feet elevation, July 1938,
and two other pairs from Villa Altagracia, July 1938, Dominican
Republic.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

Diagnosis. The color pattern; limitation of the sulcus in the con-
striction to the dorsum and upper sides of the body; and the slight
groove of the anal valves are outward characters distinguishing this
species, but final identification must rest, as with many other members
of the genus, on the peculiarities of the gonopods.

Description. Length of largest specimen, a female, 28 mm., diam-
eter 2.5 mm.; males slightly shorter and relatively more slender than
females, number of segments 40 to 44.

Fig. 13. Microspirobolus signatus. a, Gonopods, anterior view; b, Gonopods,
posterior view; c, Inner gonopod, posterior view.

Head rather dark, a little lighter at clypeus ; first segment encircled
by white band twice as wide along front margin as elsewhere; central
area dark brown or black; ensuing segments with a more or less con-
tinuous dark median line much the widest at the front of each segment
and narrowing to a point at or near the back margin; laterad of this
dark triangle, in front of the constriction, the color is dark with
variable light mottlings but the constriction itself is narrowly white
and behind it the surface is light reddish, becoming even lighter on
the posterior segments; last segment with a transverse white spot in
front extending from the dark median line to the ventral surface, the
remainder of the segment solidly dark, contrasting sharply with the

44 bulletin: museum of comparative zoology

lighter foregoing segments; anal valves dark above, lighter below and
along the margins; preanal scale light.

Vertex of head with fine median furrow; ocelli strongly convex,
numbering 28 to 37 in five or six series forming a nearly circular group.

First segment with the sides converging to the narrowly rounded
lateral limits; margin with a strong narrow rim extending from behind
the eye around to the posterior margin.

Beginning with segment 2 all but the last half dozen segments have a
shallow constriction marked by a fine sulcus crossing the dorsum and
reaching a short distance below the pore on each side; sides and dorsum
dully shining, with numerous length-wise aciculations; ventral striae
fine, extending farther up the side in front of the constriction than
behind it but not closely approaching the small pore.

Last segment with a slightly produced, rounded apex not surpassing
the anal valves which are evenly inflated but meet in a narrower,
shallower groove than is usual; preanal scale large, the apex evenly
rounded.

Gonopods as shown in figure 13, a, b, and c.

Legs 3, 4, and 5 of the male with the coxae apically depressed,
probably indicating the presence of an inflated pad in life; a similar
depression on the ventral face of the second joint of the legs beginning
with the third pair and extending to the caudal pair.

Ventral crest of segment 7 of the male high and short and sharply
rolled back, its anterior face more deeply and narrowly concaved than
usual.

MlCROSPIROBOLUS TENUIPES Spec. IIOV.

Thirteen specimens of both sexes, including the male type, from be-
tween 5,000 and 8,000 feet elevation, Loma Rucilla and mountains
north, Cordillera Central, Dominican Republic, June 193S.

Diagnosis. The shape of the very thin, compressed gonopods offers
the most satisfactory character for distinguishing this species from
other members of the genus.

Description. Largest male, the type, 30 mm. long and 2.3 mm. in
diameter, the females stouter, one of the same length being almost
3 mm. in diameter; number of segments 40 to 46.

Color in alcohol dark slate gray, at times with a slight tinge of deep
red ; a broad median spot of light color on the forebelt of each segment
showing somewhat through the foregoing segment, especially its
narrow, colorless, translucent posterior border; immediately behind
the transverse constriction, a short distance on either side of the

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 45

middle of the dorsum, is a small elongate light spot either transverse or
slightly inclined backward from the outer end; in younger specimens
the light colored areas are larger but with less distinct limits.

Head with median furrow of vertex short and faint; eyes the size of
the antennal socket or smaller, the ocelli relatively few, small and not
distinctly elevated above surface of head, numbering 21 to 23 in five
series.

Fig. 14. Microspirobolus tenuipes. a, Gonopods, anterior view; b, Gono-
pods, posterior view.

First segment narrowing on each side to the rounded-truncate lateral
limits and with a raised rim extending from behind the eye around to
the posterior margin, the front margin below the eye straight or only
slightly emarginate. Ensuing segments with a strong sulcus marking
the transverse constriction from segment 2 to the last five segments
where the constriction vanishes; pores of moderate size surrounded
by a small impressed ring; dorsal surface of segments smooth and
shining; ventral striae fine and confined to the lower surfaces except
on a few of the anterior segments where short striae immediately be-
hind the transverse sulcus exceed the tips of the legs but do not ap-
proach close to the line of pores.

Last segment produced to an acute point not surpassing the valves,
the thick margins of which meet in a deep groove; preanal scale large,
subtriangular, the sides straight to the rounded-acute apex.

Gonopods very thin, their anterior and posterior aspect shown in
figure 14, a and b.

Ventral surface of segment 7 of the male raised at middle into a very
thin, backwardly curved, lip-like ridge, broad and low on either side.

46 bulletin: museum of comparative zoology

Males with the second joint of all but the first two pairs of legs
with a depression along the ventral side indicating the presence of an
inflated pad in the living animals, the legs normal in other particulars.

Microspirobolus pullus spec. nov.

Four males (1 the type) and a female from between 1,000 and 3,000
feet elevation, Mt. Quita Espuela, Dominican Republic, July 1938.

Diagnosis. The dark color, combined with the medium body size,
and the shape of the preanal scale, are external characters distinguish-
ing this species but, as usual, the gonopods present the best differences.

Description. Largest specimen, a female, 26 mm. long and 2.5 mm.
in diameter, the largest male the same length but only 2 mm. in
diameter; number of segments 39 to 43.

Color in alcohol generally very dark brown, almost black; head
rather light brown; the eyes black; the first segment with a broad an-
terior band of light brown, the posterior margin and that of the other
segments narrowly light translucent amber; last segment, preanal
scale and anal valves dark; legs and antennae light reddish.

Vertex and front of head evenly convex and shining, entirely smooth
or with only a fine short sulcus on the vertex; eyes composed of about
26 ocelli in a four-sided or sub-triangular group of five series, the
individual ocelli convexly raised.

First segment with lateral limits short, obliquely truncated, the
anterior corner forming a right angle or less, posterior corner much
more obtuse; front border slightly emarginate below the eye and with
a broad rim extending around to the posterior corner; just above the
posterior corner one or two fine short striae usually proceed forward
from the back margin.

Beginning with segment 2 the segments of the anterior half of the
body have a strong constriction marked by a sharply impressed sulcus
in and behind which rudiments of the ventral striae, decreasing in
length, reach to the pores or slightly above them ; on the posterior half
of the body the constriction is evident on all but the last half dozen
segments but is not marked on the dorsum by an impressed sulcus and
the ventral striae are restricted to the lower sides of the body, adja-
cent to the legs; surface of segments smooth and shining; pores of
medium size.

Last segment with apex slightly produced, sub-angular, but not
surpassing the anal valves. Valves strongly shining and moderately
convex, not margined. Preanal scale short, three times as broad as

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

long; in four specimens, including the type, it is shaped as shown in
figure 15, a; in the other specimen it is more definitely triangular.
Gonopods as shown in figure 15, b, c, and d.

Fig. 15. Microspirobolus pullus. a, Preanal scale; b, Gonopods, anterior
view; c, Gonopods, posterior view; d, Inner gonopod, posterior view.

Male legs 3 to 7 with coxae slightly swollen at apex but without
distinct lobes; other joints normal.

Ventral surface of seventh segment of male raised into a transverse
ridge, high and thin at middle, the sides lower and thicker.

Microspirobolus instratus spec. nov.

One male (type) and four females collected with M. pullus between
1,000 and 3,000 feet elevation, Mt. Quita Espuela, Dominican Re-
public, July 1938.

Diagnosis. This is the largest member of the genus thus far dis-
covered in the West Indies. Its color pattern is distinctive, as also are
its gonopods.

Description. Largest female 46 mm. long, 3.5 mm. in diameter and
with 53 segments; male 34 mm. long, 2.3 mm. in diameter and with
46 segments, thus the males are slightly more slender than the females.

bulletin: museum of comparative zoology

In alcohol the head is dark brown with the elypeal region a little
lighter; segment 1 dark brown, nearly black, surrounded by a narrow
light translucent border; ensuing segments with forebelt nearly white
at the front of the dorsum where it is covered by the preceding seg-
ment, the hindbelt narrowly translucent white behind on the dorsum,
the light color of both belts broadening in descending and confluent
at the base of the legs, the intervening area on the dorsum and sides
dark brown, nearly black.

Fig. 16. Microspirobolus instratus. a, Gonopods, anterior view; b, Gono-
pods, posterior view; c, Inner gonopod, posterior view.

Head with a short, deeply impressed sulcus on the vertex and a
smaller one on the front below the line of the antennae; eyes large,
composed of definitely convex ocelli in five or six series, those of the
type containing ocelli as follows : 4, 6, 7, 7, 7-5, 6, 7, 6, 5 for the two
sides of the head, the largest female having ocelli 2, 6, 7, 7, 7, 6-3, 6,
7, 8, 7, 6.

First segment with sides converging to the rather narrow and
sharply rounded lateral margin; front margin slightly emarginate

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 49

below the eye and bordered by a fine rim around to near the back
margin; posterior margin with several fine striae proceeding forward a
short distance just above the lateral limits.

Ensuing segments with a strong constriction marked by a sharply
impressed sulcus crossing the dorsum of all but a few of the last seg-
ments; ventral striae much as in M. pullus, reaching the line of the
pores or above on the anterior segments; pores small.

Last segment slightly produced into a rather acute apex which does
not surpass the valves; preanal scale quite large, sub-elliptic, the
posterior margin more broadly rounded than angular.

Gonopods as shown in figure 16, a, b, and c.

Male legs 3, 4, and 5 with the coxal joints slightly inflated at apex,
the next joint slightly inflated along the ventral side, the legs normal
in other particulars.

Ventral surface of seventh male segment raised in the customary
ridge, the median portion rather high and thin, the sides lower and
broader.

STRONGYLOSOMIDAE

Orthomorpha coarctata (Saussure)

Polydesmus coarctatus Saussure, Mem. Myr. Mex., p. 297, 1860.

Specimens of this tropicopolitan species collected in the following
localities of the Dominican Republic in 1938: Villa Altagracia; Santi-
ago; Sanchez and vincinity; Puerto Plata; Monte Cristi.

CHELODESMIDAE

ACHROMOPORUS HETEROMUS spec. nOV.

Three males (1 the type) and a female from Sanchez and vicinity,
Dominican Republic, July 1938.

Diagnosis. The gonopods resemble those of A. enneryensis Loomis
but are more slender; the dark color of the dorsum is more extensive
than in the four other species ; the difference in size of the poriferous
and non-poriferous keels, especially in the males, is peculiar and sug-
gested the specific name.

Description. Length 23 to 28 mm., the females more convex and
robust than the males and with lateral keels smaller; males with sides
scarcely narrowing after the third or fourth segment, almost parallel.

Head with vertex and front brown to the clypeal region; first seg-

bulletin: museum of comparative zoology

ment dark brown with a white spot at middle along the front margin,
a white band along the posterior margin, including the lateral angles,
widest at the middle where it approaches, but does not join, the
anterior spot; on the ensuing three segments the keels are dark brown
except for the narrowly white outer and posterior margin including
the posterior corner, the remainder of the dorsum white; on the other
non-poriferous metazonites the keel is dark brown except for an area
at the posterior corner smaller than that on segments 2, 3, and 4;
poriferous metazonites wholly white above; sides below the keels of all
segments dark brown; last segment brown on the sides; anal valves
brown, the scale white; outer joints of legs and antennae pink.

Fig. 17. Achromoporus heteromus. a, Segments 10 to 12, left hand half
from above; b, Gonopods.

Surface of segments smooth and shining, the keels, especially the
non-poriferous ones of the males, sometimes with several small faint
tubercles, the one nearest the posterior corner most evident.

Poriferous keels of male broader than those lacking pores, as shown
in figure 17, a; the same condition exists in the female but with the
much smaller keels it is less striking; in the other species this condition
is lacking or only faintly evident.

Preanal scale triangular, much as in A. cnncrycnsis, the tip not
noticeably prolonged.

Gonopods as shown in figure 17, b.

Coxae of second legs not elevated in the female but with a small
conic corner in the male; third male legs with the sternum depressed

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 51

at middle but without special tubercles either side; male legs 3 to 7
with third joint indistinctly swollen on the under side as in A. ennery-
ensis.

Lasiomazus genus nov.

Diagnosis. Considering the gonopods, the position of this genus
appears closest to the Haitian Aehromoporus Loomis rather than to
any other American genus but even so the relationship is remote; the
outer gonopods are more sharply bent and their two branches more
diverse in size and shape, and the inner gonopods are unusually ex-
panded; none of the legs or sterna has specialized swellings but both
are remarkable in being abundantly decked with curved hairs.

Description. Body over 25 mm. in length, broadest at segments
1 to 3; dorsum smooth and shining, not greatly convex, the rather
narrow lateral keels entirely above the middle of the body; pore
formula normal.

Head deeply sulcate on vertex with two closely placed erect setae
on each side of the furrow in front; surface in front of the antennae
sparsely hispid ; antennae moderately long and slender, joint 2 longest
and almost glabrous; joints 3 to 6 nearly as long, subequal, much
more pubescent, the density of the pubescence increasing on each
succeeding joint.

First segment semicircular, with hind margin somewhat emar-
ginate at middle, posterior corners slightly produced backward.

Segments 3 to 5 with a tiny tooth on each broadly rounded anterior
corner; posterior corners of these and succeeding segments not cau-
dally produced until after segment 15, the corners of segment 18 most
produced, those of segment 19 very much smaller and scarcely exceed-
ing the posterior margin ; lateral keels narrow, rather thick, continuous
with the dorsum and projecting from high above the middle of the
body, the outer margin with a sharply defined rim which is slightly
thickened around the pores, the pores opening obliquely outward and
upward.

Last segment quite long and narrow, the slender apex scarcely
deflexed.

Anal valves with rather thin margins strongly elevated; preanal
scale large, triangular or angularly rounded behind.

Gonopods as shown in figure 18.

Sterna throughout body broad and densely beset with caudally
curved hairs; similar hairs are present on the ventral surface of the
joints of the legs; the legs and sterna in front of the gonopods slightly

bulletin: museum of comparative zoology

stouter and with hairs longer and more dense than elsewhere, other-
wise unmodified.

Type. L. concolor spec. nov.

Lasiomazus concolor spec. nov.

Male type and another male from Loma Vieja, near Constanza,
about 6,000 feet elevation, Cordillera Central, Dominican Republic,
Aug. 7 to 9, 1938.

Characters not given in the generic description are as follows:
Length 26 to 28 mm., width 3.1 to 3.4 mm.; body widest at segments
1 to 3, narrowing gradually thereafter; surface of segments smooth and

Fig. 18. Lasiomazus concolor. Gonopods.

shining, some of those at middle of body with a very shallow, indefinite
transverse median depression; non-poriferous segments 8, 11 and 14
are slightly narrower than the adjacent poriferous ones; body colorless
in alcohol except the light red antennae.

Legs throughout body with joints 1 to 4 nearly glabrous above,
densely pubescent or hispid below; joints 5 and 6 completely hispid;
the broad, hispid, sterna of each segment separated by a short glabrous
space; sternum of the seventh male legs strongly depressed in contrast
to the sternum of the sixth and foregoing legs; sternum of the pair of
legs following the gonopods short and vertical.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 53

Hypselodesmus genus nov.

Diagnosis. The large, thin, ascending lateral keels of the segments,
with a tooth at the anterior corner of all but those at the ends of the
body; the indefinite but obvious transverse depression of the segments
followed by a submarginal series of tubercles; and the structure of the
gonopods show the remote association of this genus with other .mem-
bers of the family in the West Indies.

Description. Body of moderate size; the sides nearly parallel; color
of the poriferous segments somewhat different from those lacking
pores.

Head with a strong sulcus on the vertex, on each side of which
are two erect, closely placed setae; antennae of moderate length,
joints 2 to 6 inclusive of equal length with the inner joints almost as
abundantly pubescent as joint 6; surface in front of the antennae with
a few scattered, erect setae.

First segment short, subelliptic, the median portion of the posterior
border slightly emarginate, the outer portion each side extending
obliquely outward and forward to the acute corner.

Succeeding segments with the dorsum only slightly convex and,
from segment 3 or 4 caudad, transversely crossed at middle by a
broad, shallow, indefinite but obvious depression, behind which, close
to the posterior margin, is a series of eight to ten small faint tubercles
most distinct on the posterior half of the body; in front of the depres-
sion is a series of four to six less apparent tubercles ; lateral keels rising
from far above the center of the body and strongly ascending so that
their outer margins are higher than the middle of the dorsum; the
keels are thin, especially those without pores, and are unusually large,
each extending outward a distance almost equal to the width of the
dorsum itself, the raised rim along the three free margins strong; pores
opening obliquely outward from the bottom of an elongate oval de-
pression in the additionally thickened rim of the usual segments ; from
segment 2 to segment 15 or 16 a small acute tooth is present at the
broadly rounded anterior corner of each keel; anterior segments with
the posterior corners not surpassing the back margin but at segment
6 or 7 the corners begin to be acutely produced and this increases to
segment 17 where the corners are very large and conspicuous, the
corners on segments 18 and 19 reduced, those of segment 19 very
small, hardly a third as large as those of segment 18; segments 3 and
4 with a fine bowed ridge low on each side just above the base of the
legs, faint indications of a similar ridge present on segments 2 and 5 also.

bulletin: museum of comparative zoology

Last segment quite long, with the produced apex slightly deflexed.

Anal valves rugose, the rather thick raised margins smooth and
shining; preanal scale a little broader than long, almost evenly rounded
behind from side to side.

Gonopods quite simple, each consisting of a distally slender sig-
moid flexure curving out and around behind a short, stout, subcylindri-
cal inner joint.

Second male legs with large seminal tubercles on the coxae; other
legs and sterna without special modifications, the legs long and slender,
far exceeding the sides of the body, joint 3 longest but almost equalled
by the last joint; each sternum with six to ten erect setae on either side
of the middle.

Type. //. bicolor spec. nov.

Hypselodesmus bicolor spec. nov.

Three mature males, including the type, and two immature speci-
mens from between 1,000 and 3,000 feet elevation, Mt. Quita Espuela,
Dominican Republic, July 1938.

Fig. 19. Hypselodesmus bicolor. a, Segments 10 and 11, dorsal view; b,
Segments 17 to 20, dorsal view; c, Gonopods.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 55

Description. Length 29 mm, width 4 mm; body widest at segments
1 to 3.

Color dark brown except at the posterior corner of each keel which
is white, the area of white being much larger and more intense on the
poriferous segments than on those without pores.

Lateral keels large, as shown in figure 19, a, those at the posterior
end of the body as shown in figure 19, b.

Gonopods as shown in figure 19, c.

Seminal duct opening from the very obliquely truncated inner face
of a long cylindrical erect process rising from each coxa of the second
legs.

Other characters are given in the generic description.

Cyrtaphe domingensis spec. nov.

Many specimens of both sexes, including the male type, from be-
tween 3,000 and 4,000 feet elevation, Mt. Diego de Ocampo, Northern
Range, July 1938, and other specimens from Mt. Is. de Torres, Puerto
Plata, Dominican Republic, Sept. 1938.

Diagnosis. The shape of the gonopods definitely associates this
species with C. continuata Loomis but the coloration of the body is
different and there is less similarity in pattern between the poriferous
and nonporiferous segments. The sterna between the third and fourth
pair of legs, in the male, are without special swellings.

Description. Length 30 to 33 mm, females with the body thicker
and the dorsum more convex than in the males and with the posterior
corner of the keels less acute.

In alcohol the head is dark brown except in the clypeal region which
is yellowish; first segment encircled by a white margin most extensive
at the lateral angles but also somewhat broadened at the middle of the
front and back margin; on ensuing segments the prozonite is brown
with a large white spot at middle; metazonites on the non-poriferous
segments dark brown with only the outer margin of the lateral keels
white; poriferous metazonites of fully colored specimens lighter brown
with the keels wholly white; in specimens not in full color segments 2,
3, and 4 have a broad continuous median band and ensuing poriferous
metazonites may be white with only a dilute brown spot on each side
of the dorsum in front; last segment brown except at the middle in
front and at the tip.

First segment evenly rounded in front from side to side, the pos-
terior margin medianly emarginate, lateral angles acute.

bulletin: museum of comparative zoology

Segments 3 and 4 sometimes with a minute inconspicuous tooth on
the rounded anterior corner of the keels.

Surface of segments smooth and shining except that the keels from
segment 2 or 3 to segment 17 or 18 usually have two or three small,
low, indistinct tubercles near the base.

Fig. 20. Cyrtaphe domingensis. Gonopods.

Anal valves resembling those of C. alternata Loomis.

Gonopods as shown in figure 20.

Second male legs with the inner corner of the coxae rounded, not
produced. Third sternum little narrower than in the female, lacking
definite swellings as also does the fourth sternum; in general the
sterna of the males are considerably more hispid than those of the
females.

Ricodesmus stejnegeri Chamberlin

Three males, collected at about 2,500 feet elevation Maricao
Forest, western Puerto Rico, June 2 and 3, 1938; many specimens of
both sexes from El Yunque, Puerto Rico, May 1938.

A large female measure 30 mm long.

Podiscodesmus genus nov.

Diagnosis. This genus seems more closely affiliated to Antillodcsmus
Chamberlin than to other known members of the West Indian fauna

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 57

but the small gonopods, while of somewhat similar form, distinguish it
as also does the ridge on the side of the anterior segments immediately
above the base of the legs.

Description. Body of intermediate size, with the dorsum strongly
convex and the lateral keels projecting but a short distance from the
sides; body widest at segments 1 and 2, the next three or four segments
gradually narrowing, followed by poriferous segments which are uni-
formly wider than those lacking pores ; poriferous segments no different
in color from the others; segments smooth above except for slight
rugulosity behind near the base of the keel.

Head with a pronounced median sulcus ; antennae moderately long,
joints 2 to 6 subequal in length with pubescence gradually increasing
to joint 6.

First segment of semi-circular form as common in the family, the
posterior border emarginate along the middle.

Ensuing segments with lateral keels projecting from above the
middle of body, not following the descent of the dorsum but tending to
rise to the horizontal; anterior segments lacking tooth at the front
corner of the keels; posterior corner of the keels, to near the back end
of the body, not produced backward as they are so distinctly in Antil-
lodesmus, but those lacking pores sharper than where pores are present,
the callus rounding the angle; posterior corners of segments 17 to 19
a little produced, those of segment 18 strongest, the keel on each side
of segment 19 reduced to a small tooth projecting out and back from
near the hind margin but not exceeding it; pore formula normal;
lower sides of segments 3 to 11 with a sharply denned elevation im-
mediately above the base of the legs, beginning as a sharp tooth on
segment 3 and thereafter developing into a downwardly bowed ridge
which later is reduced to a small tubercle near segment 1 1 before dis-
appearing. In AntUlodesmus this ridge is replaced by an indefinite,
broad, low swelling with a few tiny granules scattered on the surface.

Anal valves rugulose, the raised margins smooth and shining; pre-
anal scale quite large; rounded, rather than angular, behind.

Gonopods unusually small, straight and slender, their tips scarcely
reaching the posterior edge of the sternum of the seventh legs, the
basal joint not projecting outside the segmental aperture which is
narrower than adjacent sterna.

Legs with joint 3 surpassing the others in length; a single long setae
on the under side of joints 1 and 2, lacking from joints 3 and 4; sterna
broad, low, glabrous; pregenital legs and sterna like the postgenital
ones.

Type. P. carinatus spec, now

bulletin: museum of comparative zoology

PODISCODESMUS CARINATUS spec. nOV.

Two males, one the type, from Sanchez and vicinity, Dominican
Republic, July 1938.

Length 25 mm, width 3.5.

In alcohol the head is dark brown gradually whitening at the labral
region; antennae with basal joint white, the others gradually deepen-
ing in color to dark brown; legs entirely white; first segment reddish
brown, the back margin lighter brown, widest at middle, lateral angles
white; lateral keels of segments 2 to 4 white, thereafter the light color

Fig. 21. Podiscodesmus carinatus. Gonopods.

gradually becomes more restricted to the posterior corner of the keels,
the remainder of the dorsal surface reddish brown with a broad area of
light brown (possibly white in life) along the middle of the posterior
margin; last segment light reddish brown in front; white at the apex.

Head with a median sulcus at the bottom of a more general depres-
sion of the surface ; front below the antennae sparsely beset with erect
setae.

On either side of segment 3 just above the base of the legs an acute
tooth projects caudo-laterad, its free apex equalling the posterior mar-
gin ; on the next three segments this tooth is replaced by a downwardly
bowed longitudinal ridge, largest on segment 6 after which it gradually
lessens in size, becomes a small tubercle and finally vanishes at seg-
ment 11 or 12; pores opening obliquely outward from a large, elongate
callus or thickening of the marginal rim.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 59

Last segment quite long, the slightly deflexed apex rather suddenly
constricted.

Gonopods as shown in figure 21.

Craterodesmus genus nov.

Diagnosis. This is an unusually stout-bodied genus with no close
relatives known in the West Indian fauna. The swollen preanal scale
probably is a generic character and of diagnostic value but most im-
portant are the gonopods which are notably different from other recog-
nized genera.

Description. Size large, body stout and convex, much like Fontaria
but the lateral keels are narrower and on all but the most anterior and
posterior segments are well separated from each other; surface shining,

Fig. 22. Craterodesmus ovatus. Gonopods.

almost smooth except for a few fine wrinkles and two transverse rows
of what appear to be folicles on the keels and adjacent dorsum;
poriferous metazonites almost colorless, the non-poriferous metazonites
mostly dark brown with lighter margins.

Head with a strongly impressed furrow across the vertex and a pair
of erect setae on each side opposite its middle; surface in front of the
antennae sparsely hispid; antennae slender and moderately long,
joint 2 longest, the next four joints of subequal length; joint 6 slightly
broader distally than the other joints.

60 bulletin: museum of comparative zoology

First segment narrowly transversely elliptic with the back margin
broadly concave at middle; lateral angles acute with front and back
margins bordered by a raised rim for over half way to the middle of the
dorsum.

Ensuing several segments with lateral keels overlapping slightly,
followed by segments where the keels are definitely separated but on
the last few segments some overlapping occurs; anterior corner of seg-
ments without a tooth, the posterior corners not produced backward
except moderately so on the posterior segments; segment 19 with the
lateral keels scarcely indicated, the posterior angles tiny; pores in
normal arrangement, opening almost vertically from a gradually much
expanded and flattened posterior part of the margining rim; anterior
spiracle on each segment preceded by an especially high, thin crest, in
outline almost an equilateral triangle.

Last segment broad at base, short, rapidly narrowing to the short
apex which is not deflexed.

Preanal scale large, inflated behind in both sexes.

Sterna broad and low, separated by an impressed line only near the
sides.

Type. C. oratus spec, now

Craterodesmus ovatus spec, now

Two males, one the type, and three females from between 3,000 and
4,000 feet elevation, Mt. Diego de Ocampo, Northern Range, and five
females from between 1,000 and 3,000 feet elevation, Mt. Quita Es-
puela, Dominican Republic, July, 1938.

Description. Largest male 35 mm long and 6 mm wide; largest
female 42 mm long and 8 mm wide; the males a little less convex than
the females.

Head dark brown on the vertex and in a narrow median line ex-
tending between the antennae to the labrum where it broadens
slightly; sides of head, antennae, legs and ventral surface of body
uncolored ; first segment margined with white, broadening in front and
back at the middle, inner area moderately dark brown; next three
segments with a large dark brown spot on either side partly on the
dorsum and partly on the keel, the two spots joined by a lighter band
of brown suffusing the junction of the zonites, remainder of segment
white; on ensuing prozonites there is a small light brown spot below
the line of the keels and a larger darker spot above it; poriferous meta-
zonites almost colorless; those without pores colorless along the back

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 61

margin and up the outer margin of the keel, the posterior band widest
at the median line; remainder of zonite dark brown on the sides,
lighter toward the middle ; last segment with apex and a spot on each
side in front light brown.

In addition to the structural characters given in the generic descrip-
tion it may be noted that the anal valves are slightly coriaceous, with
thick, high, shining margins; preanal scale transversely oval in outline,
the surface low in front adjacent to the last segment but thereafter
suddenly inflated, the swollen posterior face of which bears the cus-
tomary two setae widely separated.

Near the legs the anterior spiracle is definitely larger than the pos-
terior one and is immediately preceded by a short, thin, high, angular
crest absent from the posterior spiracle.

Gonopods large and conspicuous, shaped as shown in figure 22.

Third joint of legs immediately before and after the gonopods with
a slight bend or umbo beneath; sterna of male legs 5 to 7 broadly de-
pressed at middle to the level of the posterior margin of the segment,
apparently for the reception of the gonopods.

Biaporus genus nov.

Diagnosis. Instantly recognized by the pore formula which is unique
in the family if not also in the entire order.

Description. Body of moderate length, rather slender; females
parellel-sided, males similar to females in width to segments 15, 16,
and 17 which are definitely wider than the foregoing segments and
more depressed ; lateral keels narrow as compared to most other mem-
bers of the family, thick at base and projecting only a little way from
high above the middle of the body and continuous in contour with the
smooth, quite convex dorsum.

Head with furrow of vertex long and strongly impressed ; one or two
erect setae on either side of it in front; surface below antennae rather
sparsely hispid; antennae of moderate length, the second joint longest,
sparsely pubescent, the pubescence increasing on succeeding joints.

First segment smaller than in many genera of the family; semi-
circular; the back margin almost straight across, only faintly emargi-
nate at middle.

Segments 2 to 5 with anterior corners of the keels abruptly rounded
and each with a tiny tooth; succeeding keels more broadly rounded at
the anterior corner and toothless ; posterior corners of keels almost right
angles, first produced backward on segment 16 but even on segment 18,

bulletin: museum op comparative zoology

where the angles are more produced, they are short and not very acute,
those of segment 19 much smaller but relatively more acute, the whole
segment much narrower than segment 18; last segment short, with a
series of four erect setae across middle and two others at base of the
apex which is short and slightly deflexed; pores small, opening ob-
liquely outward from a depression in the somewhat expanded rim of
the keel close to the posterior angle on segments 5, 7 to 13, and 15 to
19, thus segments 6 and 14 are poreless.

Anal valves with rather thin and not greatly elevated margins,
preanal scale large, subtriangular, the apex slightly produced as in
several species of Achromoporus Loomis; two setae in the margin ad-
jacent to the apex.

Legs with joint 3 definitely longer than any other joint.

Gonopods as shown in figure 23.

Type. B. montanus spec. nov.

Biaporus montanus spec. nov.

A male (type) and a female from Pico del Yaque, Loma Rucilla,
8,000 to 10,000 feet elevation, Cordillera Central, Dominican Repub-
lic, June 1938; another female from the mountains north of Loma
Rucilla between 5,000 and 8,000 feet elevation the same month.

Fig. 23. Biaporus montanus. Gonopods.

The following characters were not mentioned in the generic descrip-
tion. Length 22 to 24 mm, width 2.7 to 3 mm; color apparently white
or nearly so in life, very light brown in alcohol; surface of body smooth

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 63

and shining; anal valves slightly coriaceous with the raised margins
smooth and shining; preanal scale smooth; male legs in front of the
gonopods stouter than those following and with the upper surface of
the second joints more swollen, the sterna and ventral surface of these
legs sparsely beset with long erect hairs, the other legs and sterna and
all those of the females with almost no setae, those present being much
smaller than the anterior ones of the male.

Synecheporus genus nov.

Diagnosis. Insofar as I am aware no other member of the Chelodes-
midae has as continuous a pore formula as the present genus, in fact
one of the most constant characters of the family is the normal, dis-
continuous pore formula. Hence, it is remarkable to find two genera,
the present one and the foregoing in the same region, both exhibiting
departures from the usual formula.

Description. Body of moderate size, over 20 mm long; widest at
segments 16, 17, and 18; males similar to females in size and shape;
dorsum smooth and shining, only slightly convex, the lateral keels
projecting a short way from the sides well above the middle of the
body; a few segments at the anterior end of body with a small tooth
at the front corner of the keels.

Head with long, deep, median furrow on the vertex on either side of
which, in front, are five to ten erect setae; surface below the antennae
evenly convex and scattered with many erect setae; antennae of
moderate length; joint 2 slightly longer than any of the four subequal
joints ensuing, first joint glabrous, those thereafter increasingly
pubescent ; antennae of male a little stouter than those of the female.

First segment semi-circular with the back margin slightly emarginate
at middle.

Segments 2, 3, and 4 shorter than those that follow; from segment
2 to the middle of the body the posterior corner of the keels is nearly a
right angle but thereafter it is slightly produced, becoming strongly so
on segments 17 and 18, particularly the latter, where each corner is
broadly triangularly produced; segment 19 with the posterior corners
greatly reduced in size; segments 16 to 18 wider than any others and
with the dorsum flatter; pores small and opening almost straight up-
ward from a definite depression in the expanded raised rim of segments
5 and 7 to 19 inclusive.

Last segment with a dorsal row of four setae across the middle and
two other setae near the apex which is produced backward but not
deflexed.

04 bulletin: museum of comparative zoology

Anal valves with rather thin raised margins; preanal scale sub-
triangular, its posterior margin suddenly thickened.

Legs with the third joint definitely longer than any other.
Gonopods as shown in figure 24.
Type. S. platyurus spec. nov.

Synecheporus platyurus spec. nov.

Six males (one the type) and four females from Pico del Yaque,
Loma Rucilla, between 8?000 and 10,000 feet elevation, Dominican
Republic, June 1938; other males and females from Loma Rucilla and
mountains north, 5,000 to 8,000 feet elevation, the same month.

Characters not given in the generic description are as follows : Body
from 21 to 25 mm long and up to 3.5 mm wide. Color probably white
in life, rather light brown in alcohol, apparently stained.

Fig. 24. Synecheporus platyurus. Gonopods.

Beginning at segment 2 and extending to segment 5, 6, 7, or even
to segment 8, a small tooth is found on the rounded anterior corner of
each keel; from segment 5 to 19 inclusive only segment 6 is without
pores; preanal scale subtriangular with the apex produced slightly as
that in Achromoporus Loomis; the entire back margin suddenly
thickened, lowest and narrowest at the sides, highest and broadest at
the apex, near each side of which an erect seta projects from the
thickened margin.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 65

Male legs 3 to 7 with the second joint swollen next to the body, the
other joints also a little stouter than on ensuing legs ; sterna and ven-
tral surfaces of the legs in front of the gonopods beset with long hairs;
all sterna and legs of the females and those of the male, following the
gonopods, with a few scattered hairs.

POLYDESMIDAE

Cryptogonodesmus Silvestri

Cryptogonodesmus Silvestri, Anal. Mus. Nac. Buenos Aires, Vol. 6, pp. 59-60>

1898.
Chilaphrodesmus Loomis, Smith, Misc. Coll., Vol. 89, No. 14, pp. 42-43, 1934.

There appears to be no doubt as to the correctness of the above
synonymy.

Cryptogonodesmus rubellus (Loomis)

Chilaphrodesmus rubellus Loomis, Smiths. Misc. Coll., Vol. 89, No. 14, pp.
42-44, 1934.

A female from between 5,000 and 8,000 feet elevation, Loma
Rucilla and mountains north, Cordillera Central, Dominican Republic,
June 1938.

Cryptogonodesmus darlingtoni spec. nov.

One male (type) and three females from Loma Vieja, south of Con-
stanza, elevation about 6,000 feet, Cordillera Central, Dominican Re-
public, August 1938.

Diagnosis. A larger species than C. rubellus and with the outer mar-
gin of the keels smoother, their posterior corners less acutely produced.
The gonopods further distinguish it from rubellus as well as from the
South American species.

Description. Length 7.5 mm; color in alcohol light brown; with
sufficient magnification the entire dorsal surface of the metazonites
and the exposed portion of the prozonites is seen to be densely
covered with fine smooth granules of uniform size; dorsum of the seg-
ments with the quadrate areas separately elevated and easily dis-
tinguishable; dorsal setae of segments 2 to 19 in three transverse series,

bulletin: museum of comparative zoology

four in the front series, six in each of the following series, those of the
last series projecting backward from the straight and smooth posterior
margin; outer margin of the lateral keels not dentate or serrate as in
the other species, smooth or at most slightly undulate at the marginal
setae of which there are three on the nonporiferous segments and four
on those with pores.

Antennae much like those of C. rubellus but possibly a little more
slender.

Fig. 25. Cryptogonodesrnus darlingtoni.
b, Gonopod, posterior view.

a, Gonopod, oblique lateral view;

First segment nearly semi-circular in outline, the front margin
broadly rounded, the back margin almost straight, faintly bisinuate;
surface with setae as in C. rubellus, an anterior row of ten, a median
row of four, and a posterior row of six setae.

Second segment with lateral keels directed farther forward than in
the other species, the outer margin rounded and without definite an-
terior or posterior corners; keels of segments 3 and 4 also somewhat
carried forward, rounded at the anterior corner but with the posterior
corner marked by the small tooth bearing the last of the three marginal
setae; the other nonporiferous segments have similar posterior corners
on the keels which are much less conspicuous than those of C. rubellus,
and there is only a tiny sinus or emargination at the base of each keel

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 67

behind; poriferous segments with the posterior corners blunter than
those of C. rubcllus and with the pores opening obliquely backward
from between the last two marginal setae much as in C. brevicomis
Carl, the pores scarcely dorsal.

Segments 18 and 19 with the poriferous corners much more slender
and acute than those of preceding segments, moderately produced be-
yond the straight posterior margin.

Gonopods as shown in figure, 25, a and b.

Legs and sterna on both sides of the gonopods without special modifi-
cations.

CHYTODESMIDAE

Key to the West Indian Genera of Chytodesmidae

Pore formula irregular, the pores present on segments 5, 10, 13, 16, 17, 18'

and 19 Henicomus gen. nov

Pore formula normal, the pores present on segments 5, 7, 9, 10, 12, 13, 15, 16'

17, 18, and 19

First segment with the posterior margin coarsely scalloped. Lobodesmus Loomis
First segment with the posterior margin simple or very indistinctly scalloped . .

Segments with slender tubercles bent toward rear : Cyphotylus Loomis

Tubercles usually low, often indistinct, never raised and bent backward

First segment elliptical or oval in outline Coccoelasma Loomis

First segment with front margin rounded but hind margin definitely angled . . .

Body very strongly arched, the lateral keels sharply descending

Iomoides Loomis

Body slightly arched at most, lateral keels nearly horizontal

All margins of the lateral keels with conspicuous lobes between deep incisions

Iomus Cook

Keels with not more than one margin having strong lobes separated by deep

incisions .

Posterior margin of lateral keels with one or two large lobes bounded by deep

incisions Melanodesmus gen. nov.

Outer and usually the posterior margin of the keels with small scallop-like

lobes none of which are separated by deep incisions

Poriferous keels of segments 7, 9, 10, 12 and 13, with three scallop-like lobes

on the outer margin Tridesmus Cook

Poriferous keels of above segments with four lobes instead of three on the

outer margin of each segment Docodesmus Cook

bulletin: museum of comparative zoology

Docodesmus alifer spec. nov.

Female type and another female from Pico del Yaque, Loma
Rueilla, 8,000 to 10,000 feet elevation, Dominican Republic, June
1938.

Diagnosis. This is a most unusual species in that the lateral keels,
instead of descending or being held horizontally, are strikingly ele-
vated, with their outer margins as high or higher than the middle of
the dorsum. Also the dorsum is less sculptured than in any other
known species.

Fig. 26. Docodesmus alifer. a, Segment 1, dorsal view; b, Segment 4, pos-
terior view; c, Segments 18 to 20, dorsal view.

Description. Length 16 to 17 mm, width 3 mm. Both specimens
light brown, probably white in life from which it may be inferred that
maturity had just been reached before capture, as brown to almost
black is the usual color of old specimens in the genus.

Head with the antennae slightly longer and more slender than in
other large species, joint 5 considerably the longest, exceeding joints
6 and 7 together; surface in front and to the side of each antenna
raised into a broad low ridge behind which the basal joints of the an-
tenna lie when at rest.

First segment quite long as shown in figure 26 a, its latero-posterior
margins slightly scalloped, very oblique, longer than the median por-
tion which is almost straight and bordered by a raised rim; anterior

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 69

expanded margin raised much higher than the central region, its
quadrate areas much longer than broad, especially the outer ones;
central region nearly flat, almost smooth, the usual tubercles very
faintly indicated by tiny low swellings.

On ensuing segments the lateral keels are thin and extend obliquely
upward far from the sides of the body, the outer margins as high or
even higher than the dorsum itself, as shown in figure 26 b; inner area
of each keel depressed below its margins ; non-poriferous segments and
segment 5 with 3-lobed keels, the other poriferous segments with
4-lobed keels, none of the outer margins strongly scalloped as are the
posterior margins of the keels and the dorsum; the sulci separating
the marginal areas of the keels and dorsum well impressed, constitut-
ing the most conspicuous sculpturing of the segment, as the usual
transverse areas of the dorsum are but faintly indicated, if at all, and
the central tubercle of each area is almost obliterated; the entire sur-
face dully shining and less sculptured than in any other known species;
raised rim at the front of each segment low, thin and inconspicuous on
the dorsum as well as on the keels.

Penultimate segment with keels slightly raised, unusually large as
compared with those of other species, produced backward and sharply
inward, the sinus between them wide in front but narrow between the
blunt tips of the keels as shown in figure 26 c.

Last segment small, much exceeded by the keels of the penultimate
segment ; without dorsal tubercles but with two small apical lobes and
another small lobe on either side.

Ventral ridge of the third segment narrow and high, rising to a
median point, thus it is triangular in outline and is inclined toward the
rear.

Docodesmus griseus spec. nov.

A dozen specimens, including the male type, collected at Sanchez
and vicinity, Dominican Republic, July K)3S.

Diagnosis. The smaller size, lack of secondary tubercles on the dor-
sum, the greater accentuation of the primary ones, and the more acute
keels at the posterior end of the body distinguish this species from D.
haitiensis Chamberlin. The males have the keels of segments 2, 3,
and 4 distinctly lifted above the horizontal and the coxae of the fourth
legs have hispid swellings not found in other species.

Description. Body up to 14 mm. long and 3 mm broad; color nearly
white to light grayish brown in alcohol.

bulletin: museum of comparative zoology

Head with antennae white, quite long and slender, joint 5 longest
and broadest; surface above the antennae finely granular, median
channel faint; surface between and below the antennae smooth, shin-
ing, and finely hispid, the clypeal region only slightly inflated but
laterad of each antenna a broadly swollen ridge is present.

In general the shape of the body and of the individual segments,
their position and sculpturing resemble D. haitiensis but the following
differences are noted. In the males the keels of segments 2, 3, and 4

Fig. 27. Docodesmus griseus. Right hand gonopods with sternum and coxal
joints of ensuing legs.

are obliquely raised a little above the horizontal; the posterior seg-
ments of both sexes have the keels produced farther backward and
more acute; quadrate areas of the dorsum with the central tubercle
larger, no additional smaller tubercles present; raised ridge across the
front of each metazonite as high or higher than in D. haitiensis and
more irregular at apex.

Anal valves notably flattened, especially near the scale, the margins
lower and broader than usual; surface, and that of the scale, smooth
and shining.

Gonopods as shown in figure 27.

Males with each coxal joint of the fourth legs bearing a broad, low
swelling covered with fine erect setae.

Females with the ventral crest of the third segment higher than in
most other species, thin, rising gradually from each side to the broadly
rounded median portion; surface behind the crest low and nearly flat.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

Docodesmus haitiensis Chamberlin

Docodesmus haitiensis Chamberlin, Bull. Mus. Comp. Zool., Vol. 62, No. 5,
p. 216, 1918.

Two males and two females from Mt. Diego de Ocampo, Northern
Range, elevation 3,000 to 4,000 feet, Dominican Republic, July 1938.

Fig. 28. Docodesmus haitiensis. a, Left hand gonopods, posterior view; b,
Gonopod, lateral view.

While these specimens differ very slightly from typical haitiensis the
differences appear too insignificant to justify recognition as even a
variety.

A drawing of the gonopods of this species from a specimen collected
at Trouin, Haiti, is shown in figure 28.

Docodesmus angustus spec. nov.

Four males, one the type, and eight females from Valle Nuevo,
southeast of Constanza, elevation about 7,000 feet, Dominican Repub-
lic, August 1938; one male from rain forest near Valle Nuevo at about
6,000 feet elevation, August 1938; several males and females from
Loma Vieja, south of Constanza at about 6,000 feet elevation August
1938; a male and female from between 5,000 and 8,000 feet elevation,
Loma Rucilla and mountains north, June 1938.

Diagnosis. The wide second segment, gradually attenuated pos-

bulletin: museum of comparative zoology

terior end of body, and conspicuous lobes on outer margins of keels,
are noteworthy characters supplementing those shown by the gonopods.
Description. Body quite slender for its size, the largest female being
18 mm long and 3.5 mm wide, the largest male 15 mm long and 2.7
mm wide; sides of body parallel from segment 2 to near the posterior
end of body which narrows more gradually than in other known species ;
color in alcohol rather dark brown with the lateral keels a little lighter.

•s

' y^-

o | r>

° •

•- .-■

-^r/. /

- ' '

1 .,

/ O .

" ■"

■ ?' '. -

' — i .A , j. ,.

■""' "-■-

v 1 U

Fig. 29. Docodesmus a?igustus. a, Segments 1 to 3, dorsal view; b, Segment
10, dorsal view; c, Segments 18 to 20, dorsal view; d, Left hand gonopods, pos-
terior view.

Head with antennae short and stout as in most species, the outer
sides of the head in front of them swollen; vertex minutely granulose,
the median furrow fine and short, stopping considerably above the
antennae between and below which the surface is finely hispid.

First segment short and conspicuously narrower than segment 2 as
shown in figure 29 a, the front margin evenly rounded without indica-
tion of scallops, but with a fine raised rim ; dorsal sculpture of this and
succeeding segments quite similar to, but not as strong as, that of the

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS <d

much smaller D. parvior, shown in Bui. Mus. Comp. Zool., Vol. 80,
No. 1, pi. 3, 1936.

All segments, except three or four at each end of body, have the
outer margin of the lateral keels more definitely lobed than in other
species as shown in figure 29 b., the raised rim across the anterior
border of each segment is fairly well developed on anterior segments
but thereafter gradually diminishes and on the posterior segments is
limited to the lateral keels or is lacking; posterior segments narrowing
gradually as shown in figure 29 c, in which the difference in size of the
produced keels of segments 18 and 19 may be observed and also the
considerable exposure of the last segment which projects far behind
the keels of the penultimate segment.

Gonopods as shown in figure 29, d.

In the female the anterior ventral crest of segment 3 is raised as in
other species but the entire surface behind it is raised into a thickened
ridge almost as high as the anterior crest, a condition observed in no
other species of this genus.

Melanodesmus genus nov.

Diagnosis. Apparently most closely related to Docodcsmus but
differing in having the entire dorsum, including the tubercles, evenly
finely granulose; and the posterior margin of each segment has two
deep incisions at the base of the lateral keel, the intermediate margin
a conspicuous lobe.

Description. Body of the shape and proportions of Docodcsmus;
black; the entire dorsum, including the large tubercles, finely and
evenly granulose.

Head with the vertex finely granular, without a definite median
furrow; surface between the antennae hispid, the clypeal region smooth
and shining and somewhat inflated; antennae quite short and stout,
quite densely pubescent.

First segment noticeably narrower than those that follow; shaped as
in Docodesmus but the marginal areas less distinct and the margin
more continuous ; posterior margin with a small incision at the base of
the expanded margin; central area with ten tubercles.

Ensuing segments with keels projecting outward almost horizontally
from well above the middle of the body, the outer lobation of the keels
as in Docodcsmus but the front margin thicker; the posterior margin
thin and with two wide and deep incisions having a prominent lobe
between them at the base of the keel; dorsum with quadrangular

bulletin: museum of comparative zoology

areas faintly perceptible or not at all; four longitudinal rows of large
granule-bedecked tubercles present, three tubercles in each row except
on the anterior segments, the third tubercle in each row broader and
less distinct than the others and occupying a small lobe of the pos-
terior margin; front margin of each metazonite with an irregularly
raised rim extending across the dorsum and half way or more across
each keel ; posterior end of body narrowing suddenly and with the large
tubercles decreasing in size; keels of segment 19 considerably smaller
than those of segment 18, with the last segment exposed in the sinus
between them.

Anal valves moderately inflated, shining, almost smooth; raised
margins thin.

Preanal scale triangular, with a setose conic tubercle rising from the
surface on either side near the apex.

Third segment of the female with a high thin crest along the anterior
margin behind the second legs.

Type. M . granulosus spec. nov.

Melanodesmus granulosus spec. nov.

A single female collected between 5,000 and 8,000 feet elevation,
Loma Rucilla and mountains north, Cordillera Central, Dominican
Republic, June 1938.

Fig. 30. Melanodesmus granulosus, a, Antenna; b, Segments 1 to 3, dorsal
view; c, Segments 10 and 11, dorsal view; d, Segments 18 to 20 in outline,
dorsal view.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 75

Length 8.3 mm, width 2 mm; dorsum entirely dull black, the front
of head, antennae, legs, sterna, hispid apex of last segment, anal valves
and preanal scale white or colorless, remaining ventral surfaces black.

Head with hispid antennae as shown in figure 30, a, joints 6 and 7
together longer than either joint 2 or 5 which are subequal in length
with the latter exceeding any of the other joints in width.

First segment distinctly narrower than segment 2 as seen in figure
30, b, almost semi-circular with the expanded front margin not ap-
preciably scalloped, a few short indistinct lines giving faint indication
of the usual quadrate areas ; posterior margin with a small incision near
the middle of each oblique section.

Segments 2 and 3 with dorsal tubercles as shown in figure 30, b, but
on ensuing segments they are as shown in figure 30, c, except that on
the caudal segments the tubercles decrease in size; the segments at the
posterior end of the body shown in outline in figure 30, d.

Third segment of the female with the very thin ventral crest slightly
higher at each side than at the middle.

IOMOIDES PARALLELA Spec. nOV.

Male type and two females from Sanchez and vicinity, Dominican
Republic,' July 1938.

Diagnosis. Similar to I. glabra Loomis in the hairless dorsum but
differing in the larger size; greater development of the dorsal sculpture,
with the four rows of tubercles parallel to each other; and in the form
of the gonopods.

Description. Body longer and relatively broader than /. glabra,
from 8.5 to 9 mm long and 2 to 2.2 mm wide, the dorsum without
setae; all the tubercles larger and more prominent than in that species;
color in alcohol black.

First segment with the anterior margin a little more rounded and
scalloped than in the other species; disk with an anterior row of six
tubercles, the outer one on each side very small, the next double its
size and the inner one double the size of the second and much higher;
in the posterior row of four tubercles the outer one is slightly smaller
than the second tubercle of the first row, and the inner tubercle is as
small or even smaller than the outer one of that row.

On ensuing segments the four rows of tubercles are strongly ele-
vated, especially the two inner rows; all rows parallel instead of ex-
tending obliquely mesad from front to back; on segments 16 to 18,
quite in contrast to the condition in I. glabra, the posterior tubercle

76 bulletin: museum of comparative zoology

most developed and projecting caudad especially on segment 18; on
segment 19 the anterior tubercle of each inner row is very small and
hidden beneath the projecting tubercle of the preceding segment,
second tubercle slightly larger, the last tubercle almost as large as that
of the foregoing segment and projecting straight back contiguous to
the last tubercle of the opposite row; segment 19 broader than in I.

Fig. 31. Iomoides parallela. Left hand gonopod, mesoposterior view.

glabra, the posterior corner of the keels only slightly exceeded by the
tips of the median tubercles; lateral keels of all segments definitely
thicker than in /. glabra, the sinuses between the lobes of the anterior
and posterior margins more open.

Gonopods as shown in figure 31.

Females with the ventral crest of segment 3 broader and lower than
in I. glabra.

Iomoides conjuncta spec. nov.

A single male collected at Villa Altagracia, Dominican Republic,
July 1938.

Diagnosis. Intermediate between /. hispida Loomis and /. ■parallela,
but lacking the long seta on each dorsal tubercle as in the former species
and with much coarser dorsal sculpture; from I. parallela it differs in
the hispidulous dorsum and the oblique rows of tubercles; the coales-
cence of the three tubercles in each inner row, forming simple crests on
segments 2 to 5, does not occur in the other three species.

Description. Color black as in the other species ; the body somewhat
broader, being 8 mm long and 2.4 mm wide; the dorsal surface densely
hispidulous but lacking the long seta on each tubercle as found in
/. hispida.

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS 77

First segment with the ten rectangular areas of the front margin
long, especially the outer three on each side which are over twice as
long as wide ; inner surface strongly convex with an anterior row of six
tubercles and a posterior row of four tubercles, the tubercle at each
end of the front row and the inner pair of the back row are very minute ;
the second tubercle in the front row and the outer one in the back row
are somewhat larger but less than half as large or as high as the inner
pair of sharply conical tubercles of the front row.

Fig. 32. Iomoides conjuncta. Right hand gonopod, mesoposterior view.

On segments 2 to 5 the tubercles of each inner row are coalesced and
elevated into a simple ridge definitely higher than the corresponding
rows of tubercles on mid-body segments, the two ridges on segment 2
are short but increase in length on succeeding segments ; on segment 5
the individual tubercles composing each ridge are faintly evident and
on segments 6 and 7 become more separated and thereafter are com-
pletely separated and lower; on segment 17 the last tubercle of each
inner row is suddenly enlarged and produced backward far beyond
the posterior margin; segment 18 with these tubercles almost as large
and greatly produced but the two foregoing tubercles in each row are
reduced to insignificance; from segment 2 to segment 18 the outer and
inner rows of tubercles are oblique and converge caudally toward the
median line, except those rows on the anterior segments which are de-
veloped into parallel crests; the tubercles of each outer row are of uni-
form size throughout the body, arranged in a curved oblique line, the
middle tubercle farthest laterad, followed by the anterior one, with the
posterior tubercle nearest the dorsum; segment 19 without dorsal
tubercles except the posterior one of each median row, the two tubercles
strongly projecting backward, touching along the inner side and to-

78 bulletin: museum of comparative zoology

gether filling the sinus between the lateral keels, which they slightly
exceed; on segments 2 to 19 each outer lobe of the lateral keels has a
single seta in the margin slightly larger than those of the dorsal surface
but smaller than the corresponding setae in I. hispida.

Gonopods as shown in figure 32.

Male with the second joint of legs 3 and 4 thicker than on the ad-
jacent legs.

Iomoides sp.

A 19-segmented male from the rain forest near Valle Nuevo, Cordil-
lera Central, elevation about 6,000 feet, Dominican Republic, August
1938.

The dorsal vestiture is like that of /. conjuncta although the sculp-
ture, which resembles that of I. hispida, precludes its inclusion in the
former species, but because of the immaturity of the specimen, a new
name is not considered justified.

Henicomus genus nov.

Diagnosis. Outstanding feature of this genus is the pore formula
which not only is unique in the family, where a normal formula is the
rule, but it is not known to be duplicated elsewhere in the order Mero-
cheta. The sequence of three-and four-lobed lateral keels is another
curious and unique character. General form and sculpture suggest
closest relationship with Docodesmus but the dorsum is more convex,
with strongly descending lateral keels, and the anterior and posterior
sterna of each segment are separately elevated.

Description. Body small, about six times as long as broad; dorsum
strongly convex with lateral keels sharply descending to opposite the
middle of the body or lower; sculpture resembling the type common in
Docodesmus but not as distinct.

Head and antennae much as in Docodesmus.

First segment with the expanded front margin divided into 12
quadrate areas of which the outermost on each side is much narrower
than any of the others; median area strongly convex, not divided into
geometric areas but with ten tiny tubercles arranged as in Docodesmus.

Ensuing segments with faintly set-off quadrate areas each usually
containing an indistinct tubercle; dorsum of these segments high and
strongly convex, the lateral keels not projecting as far from the sides
of the body as in Docodesmus and much more deflexed, the outer mar-
gin reaching opposite the middle of the body or below it; the outer

LOOMIS: PUERTO RICAN AND DOMINICAN MILLIPEDS

margin of segments 2, 3, 4, 5, 6, 8, 10, 11, 13, 14, and 16, three-lobed ;
the margin of segments 7, 9, 12 and 15 distinctly four-lobed; the mar-
gin of segments 17, 18 and 19 indistinctly and apparently indiscrimi-
nately three- or four-lobed; pores opening from the dorsal surface of
the posterior lobe of the keels on segments 5; 10, 13, 16, 17, 18 and 19.

Last segment with two dorsal sub-median tubercles and a smaller
one further forward near each side; apex slightly deflexed.

Anal valves but little convex, the thin margins only slightly raised;
preanal scale rounded-triangular behind, with a large conic setiferous
tubercle on either side surpassing the posterior margin.

Sterna definitely elevated, narrow, more so than in Docodesmus,
and with a longitudinal median furrow; on segments having two pairs
of legs the sterna are separated by a deep transverse channel.

Second legs of female followed by a raised transverse ridge.

Type. H. septiporus spec, now

Henicomus septiporus spec, now

A single female collected at about 6,000 feet elevation, Loma Vieja,
Cordillera Central, south of Constanza, Dominican Republic, August
1938.

Fig. 33. Henicomus septiporus. a, Segments 1 to 5, dorsal view; b, Right
hand half of segments 11 to 13, dorsal view.

80 bulletin: museum of comparative zoology

Length 9 mm, width 1.5 mm; color in alcohol cinnamon brown.
Segments 1 to 5 shown in figure 33, a, lateral carinae not projecting
far enough to hide the distal half of the last joint of the legs; pores
opening from the center of a broad, shallow, crater-like depression in a
special swollen area which occupies the whole of the last lobe of the
keel; on segments 5, 10, and 13 the lobe is rounded behind as shown in
figure 33, b, but on segments 16 to 19 the lobes are produced backward
into sharp angles, those of segment 19 not as long as on segment 18,
widely separated, the last segment visible between them and extending
a considerable distance beyond; segment 19 with six prominent scal-
lops occupying the posterior margin between the poriferous keels;
similar scallops, decreasing in size, are present on segments 18 and 17;
figure 33, b, also shows nonporiferous three- and four-lobed keels.

Crest following the second legs of the female high, thin, and ex-
tending opposite the middle of the second joint of the leg on either side.

COMODESMIDAE

Inodesmus caraibicus (Silvestri)

Lasiodesmus caraibicus Silvestri, Bull. Amer. Mus. Nat. Hist., Vol. 24, pp.
575-576, 1908.

A female from El Yunque, Puerto Rico, May 1938.

The length, not given by Silvestri, is 7 mm, the width about 0.75
mm. In other particulars the species has been well described and illus-
trated.

_■>

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE

Vol. LXXXVIII, No. 3

BIRDS OF LOWER AMAZONIA

By Ludlow Griscom
and James C. Greenway, Jr.

CAMBRIDGE, MASS., U. S. A.
PRINTED FOR THE MUSEUM
June, 1941

noAJ K^j^^^1

No. 3. — Birds of Lower Amazonia1
By Ludlow Griscom and James C. Greenway, Jr.

TABLE OF CONTENTS

Page

Introduction 85

Principal Collections 85

Geographical Notes 86

Area included 86

Collecting Localities 87

Collecting Localities of A.M. Olalla 90

Ecological and Distributional Notes 93

Migrants or Winter Visitants '..... 96

Systematic List 97

Tinamidae • 97

Colymbidae 102

Phalacrocoracidae 102

Anhingidae 103

Pelecanidae 103

Ardeidae 104

Cochleariidae 107

Ciconiidae 107

Threskiornithidae 107

Phoenicopteridae .• 108

Anhimidae 108

Anatidae 108

Cathartidae HO

Accipitridae 1 10

Falconidae 115

Cracidae 118

Phasianidae 122

Opisthocomidae 122

Aramidae 1 22

Psophiidae ? 123

Rallidae 124

Heliornithidae 127

Eurypygidae 127

Jacanidae 128

Haematopodidae 128

Charadriidae 128

Scolopacidae 129

Recurvirostridae 132

1 Published with the aid of a special gift from Mr. George R. Agassiz.

84 bulletin: museum of comparative zoology

Page

Laridae 132

Rynchopidae 134

Columbidae 134

Psittacidae 139

Cuculidae 149

Tytonidae 158

Strigidae 158

Nyctibiidae 160

Caprimulgidae 161

Micropodidae 166

Trochilidae 166

Trogonidae 179

Alcedinidae 183

Momotidae 184

Galbulidae 186

Bucconidae 189

Capitonidae 194

Ramphastidae 194

Picidae 199

Dendrocolaptidae 208

Furnariidae 219

Formicariidae 229

Conopophagidae 253

Rhinocryptidae 254

Cotingidae 255

Pipridae 263

Tyrannidae 270

Oxyruncidae 296

Hirundinidae 297

Troglodytidae 300

Mimidae 303

Turdidae 304

Sylviidae 305

Motacillidae 306

Cyclarhidae 307

Vireolaniidae 307

Vireonidae 307

Coerebidae 3] 2

Compsothlypidae 315

Icteridae 316

Tersinidae 320

Thraupidae 320

Fringillidae 332

Bibliography 340

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 85

INTRODUCTION

For several years prior to 1932, A.M. Olalla and various assistants
sent enormous collections of birds from the Amazon River Valley to
the American Museum of Natural History in New York. The area
covered included every important tributary river except the Purus.
With headquarters at Obidos, A. M. Olalla continued collecting in
1932 and 1933, and this collection, totalling 4023 skins, was purchased
by the Museum of Comparative Zoology. The inception of this paper
was the identification of this collection, which proved to contain a few
novelties, many rarities, and numerous range extensions.

As Snethlage's Aves Amazonicas was hopelessly out of date, and the
American Museum had no plans for working up and reporting on
their Amazonian collections except in connection with other projects,
it seemed possibly more useful to try and present a complete and up-
to-date list, with such systematic notes as the material at our disposal
made possible. In this connection the great Klages Collection at the
Carnegie Museum at Pittsburgh was a possible mine of interesting
information. We learned that our good friend, Mr. W. E. Clyde Todd,
had no immediate plans for preparing a report on this collection, in
the press of various other projects approaching completion.

We accordingly sought and obtained permission to study this great
collection and to include in this report the specimens in the Carnegie
Museum and all information in any way supplementary to that
already available. Such generosity and cooperation is most unusual,
and we here gladly express our sense of deep obligation and appre-
ciation to the Director of the Carnegie Museum, Dr. AvinofT, and our
colleague Mr. Todd, who did everything possible to assist us while at
Pittsburgh. By agreement with these two gentlemen we described
such new forms as were represented by specimens in both institutions,
and Mr. Todd published a series of papers on the novelties in the
Carnegie Museum alone.

PRINCIPAL COLLECTIONS

Hellmayer's classic paper on the birds of Para (Abh. Konigl. Bayer.
Akad. Wiss., Math. Phys. Klasse, 26, 1912) gives a summary of various
small collections in Europe from the vicinity of Para and Marajo
Island. These include such historic collections as Natterer's at Caju-
tuba in 1835, and Wallace's birds from the Rio Capim in the British
Museum. Earlv American collections were those of Prof. Steere on

86 bulletin: museum of comparative zoology

Caviana and Marajo Islands in 1871 and 1879 (reported on by Brod-
korb in 1937) and two made at Santarem, reported on by Ridgway,
Chapman and Riker.

The Goeldi Museum at Para was founded in 1891 by Dr. Goeldi,
who secured some material from Lower Amazonia, but Dr. Emilia
Snethlage for years explored up and down the Amazon River with
indefatigable energy and enthusiasm, and her important collections
culminated in her great work, Aves Amazonicas (1914).

Various other Brazilian ornithologists have contributed to our
knowledge of the birds of the Lower Amazon, among them C. O. da
Cunha Vieira, Dr. Miranda-Ribeiro and Pedro Pinto-Peixoto, all
reporting on small collections in Rio de Janeiro or Sa5 Paulo. Today
the leading ornithologist of Brazil is Dr. O. M. de Oliveira Pinto,
whose monumental catalogue of the birds of Brazil, Part I, lists the
specimens of a final collection made by A. M. Olalla in 1934, acquired
by the Museu Paulista. We are much indebted to him for compli-
mentary copies of his invaluable papers.

As already stated the Olalla collection purchased by the M.C.Z.
totals 4023 specimens. Fortunately, Mr. Zimmer of the American
Museum, in connection with his studies of Peruvian birds, has studied
and listed the material in New York of the difficult passerine families
in the superfamily Furnariides. The balance of the Olalla collections
in New York are not reported on.

The Klages Collection in Pittsburgh contains 7379 specimens from
our area in lower Amazonia; about nine thousand specimens from
upper Amazonia and seven thousand from French Guiana and adja-
cent Brazil afforded magnificent comparative material.

GEOGRAPHICAL NOTES
Area Included

As is almost invariably the case, it is quite impossible to select as a
unit an area in which political, geographical and faunal boundaries
coincide. Our principal object has been to compile a list of the rich
bird-life known to date from the Lower Amazon. The western limits
of the area are the western boundaries of the State of Para, which on
the south side of the Amazon, runs between the Rio Tapajoz and Rio
Madeira. On the main river, just a few miles west of this boundary
line, is the city of Villa Bella Imperatriz or Parintins, and the few
birds recorded from here are included. On the north bank, the town of

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 87

Faro on the Rio Jamunda is about on the State boundary line. Faun-
ally there is here no embarrassment or difficulty, as new genera and
species of birds are encountered the moment we proceed still further
west to the Rio Madeira and Rio Negro.

The eastern limits of our area must be selected on a more arbitrary
basis, as the coast of Brazil happens to run eastward from the mouth
of the Amazon. In this region lies the boundary between the States of
Para and Maranhao, and somewhere in here there is also an important
faunal boundary, as we pass from the great river forests of the Amazon
to the campos of Maranhao. Little or no exact detail is on record.

Theoretically, our northern and southern boundaries should include
the whole drainage system of the Amazon. Actually, this would in-
clude a gigantic area, stretching from the Brazilian Guianas to the
tableland of northern Matto Grosso, one of the greatest unexplored
areas of wilderness in the world, with scarcely a bird skin on record.
On the north bank of the Amazon, the rivers are relatively small, and
no collecting has been done more than a few miles north of the main
river. On the south bank, however, the great rivers Tapajoz, Xingu,
and Tocantins rise far south of the boundaries of the State of Para, but
we are here concerned only with the relatively few miles along which
ornithological collections have been made. Even the uppermost
points reached by Madame Snethlage in her traverse from the Xingu
to the Tapajoz are little more than half way to the Para-Matto
Grosso boundary.

As thus outlined, the area included in this report is the lowest
quarter of the main Amazon, and its tributary rivers to the degree to
which they have been ascended by ornithological collectors. It is
consequently a mere fraction of area in terms of square miles of the
State of Para, and a still smaller fraction of the drainage system of the
main Amazon.

COLLECTING LOCALITIES

In Lower Amazonia the rivers are the main highways of travel, and
the only means of access to most of the interior. Practically all towns,
villages and hamlets are on some well known river or one of its tribu-
taries. Thus, if you are at Para and wish to go to Santa Helena, you
must go up the Amazon to the mouth of the Rio Tapajoz, up the
Tapajoz to the mouth of the Rio Jamauchim, and up the Jamauchim
until you reach this particular hamlet. In much of Lower Amazonia,
an overland or "cross-country" trip between two rivers has never

88 bulletin: museum of comparative zoology

been done, or at least would require a well equipped and competently
led expedition in the dry season. The rivers are consequently used
almost entirely in orienting place names. First of all, you are on the
north or south side of the main Amazon; secondly, you are located
on the west or east bank of some tributary, or your locality is defined
in terms of the mouth of the nearest well known river. Thus, Aruman-
duba is on the north bank of the Amazon, between the Rio Pari! and
the Rio Jary.

It is important for readers to remember that the Brazilians fre-
quently use the terms right and left for river banks, in which case the
river is always thought of as being descended. This is quite simple in
the case of a river like the Tapajoz, flowing from south to north, where
the east bank is automatically the right bank, but this situation is
reversed in those rivers flowing into the Amazon from the north, and
one sometimes has to stop and think before realizing that the right
bank of the main Amazon is the south bank.

In this connection Brazilian Portuguese has adopted certain Indian
words into the language which frequently occur in place names.
Igarape means stream, creek, or bayou; cachocira means a rapid; assii
means big. Such local place names have often in the past been selected
as ornithological collecting localities, with exceedingly poor judgment.
Fortunately, we know approximately where Igarape-assu is. It is
nothing in the world but a large creek, an affluent of the Rio Acara,
above the town of Acara. Two or three seasons of particularly heavy
freshets and floods might completely eliminate some igarape.

Any atlas map of Brazil shows all the principal rivers and towns
mentioned in this report. Minor place names can readily be found by
consulting the map at the end of Snethlage's Aves Amazonicas. De-
tails of place names on the Rio Tapajoz can be found on the map in
Snethlage's article in Bol. Mus. Goeldi, 7, 1910. For the convenience
of the reader, the principal localities are roughly placed in the sche-
matic arrangement adopted below. We proceed from west to east.

I. North bank of main Amazon

Rio Jamunda — Faro (Snethlage and A.M.N.H.)
Obidos, a large city and port, with numerous minor collecting
stations near by. The Carnegie Museum has a large collec-
tion from some islands in the Amazon near Obidos. (Numer-
ous collectors)
Rio Maecuru — Cachoeira Muira and Igarape de Paituna
(Snethlage)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 89

Monte Alegre and Erere (chiefly Snethlage)

Rio Parii (no collections)

Arumanduba (Snethlage)

Rio Jary — So. Antonio de Cachoeira (Snethlage)

Macapa (Snethlage)

Amapa (Snethlage)

II. South bank of main Amazon

Villa Bella Imperatriz or Parintins (A.M.N.H.)
Lago Grande (M.C.Z.)
Rio Tapajoz (all collectors)

a. left bank — Villa Braga, Jtaituba, Pinhel, Boim

b. right bank — Apacy, Miritituba, Aveiros, Caxiricatuba,

Tauary, Pinhy

c. Above or upstream from these localities the Rio Jamauchim

comes in from the southeast.

Santarem, a small city and port (all collectors)

Cussary, about half way to the

Rio Xingu- Victoria, Boa Vista. Just upstream from these
towns, the Rio Iriri debouches from the southwest, and still
further inland the Rio Curua comes in from the southwest
(Snethlage and A.M.N.H.)

Rio Tocantins (Snethlage and A.M.N.H.)

a. left bank — Arumatheua, Alcobaca, Cameta

b. right bank — Baiao, Porto do Moz
Rio Mojii

Rio Acara-x\cara, Igarape-assu.

Rio Capim, which flows into the

Rio Guama — Ourem

Para, the capital city, also known as Belem, has numerous

suburbs, among them Providencia, Ananindeua and Val-de-

Caes.
Railroad from Para to Braganca on the Atlantic.

Well known stations are Benevides, Peixe-boi, and Quatipuru.

St. Antonio do Prata is a small town south of Peixe-boi.

III. Mouth of the Amazon

Innumerable islands, of which the very large one is Marajo.
Two small outside islands are Cavianna and Mexiana.

An excellent gazetteer of localities worked by older collectors can be
found in Hellmayr's classic paper in Abh. Konigl. Bayer. Akad. Wiss.,
Math. Phys. Klasse, 26, 1912, pp. 84-85. Of these the only important

90 bulletin: museum of comparative zoology

one, not on Snethlage's map, is Natterer's famous station Cajiituba,
a beach on the sea-coast northwest of Cintra. Here Natterer col-
lected 105 years ago various wading and sea birds, never since reported
from the region.

Some readers may like to get some idea of the geography of Lower
Amazonia. In addition to the standard books of travel on the Amazon
River, the following articles are recommended:

1. Snethlage, Bol. Mus. Goeldi, 6, 1910, pp. 226-235, deals with the
savannahs on the upper Tocantins.

2. Snethlage, ibid., 7, 1910, pp. 49-92, map and 15 plates, deals
with her famous trip up the Rio Xingti and her traverse across country
to the Rio Jamauchim.

3. Ducke, ibid., pp. 100-197, and 12 plates, while primarily botani-
cal, gives an excellent idea of the types of country around Faro and
Obidos.

4. Miiller, Abh. Konigl. Bayer. Akad. Wiss., Math. Phys. Klasse,
26, 1912, pp. 1-80, describes the savannahs of Marajo Island and the
country near the capital city of Para.

Collecting Localities of A. M. Olalla

Rio Tapajoz: Santarem

A locality situated at the junction of the Amazon with the river of
this name. A small city, with considerable commerce, and a port that
can accommodate any of the local ships. The built-up part is sur-
rounded by savannahs for a distance of about two kilometers, beyond
which the dense vegetation, high and savage, begins. The land is
relatively level as far as the Hacienda of Piquiatuba, where a slope
begins and ends with the plateaus of the farms of San Jose. The
climate is healthy, without plagues of mosquitos. Birds abundant,
but mammals rare.

Rio Tapajoz: Caxiricatuba

A locality situated on the eastern shores of the Rio Tapajoz about
21 miles from Santarem; the place is inhabited by two native families.
The terrain is varied, including Igapo on the shore of the river, and
there is dry land with virgin forest in the center. Birds and mammals
relatively common.

Rio Tapajoz: Tauary

Near Caxiricatuba, about 24 miles from Santarem, and is also on the
east side of the river. The igarape is navigable for small craft in the
winter (wet season) ; some natives live there. The terrain and flora

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 91

are similar to those of Caxiricatuba. Birds and mammals abundant.
An igarape, with a fairly long course from its source to its junction
with the Tapajoz.

Rio Tapajoz: Pataua

The central locality situated about 4 miles to the east of Tauary.
Region uninhabited; virgin forest; terrain varied and dry. Birds and
mammals abundant.

Rio Tapajoz: Pinky or Piny

About two miles from Tauary, upstream and on the same bank of
the Rio Tapajoz. A small igarape, navigable at the height of the wet
season, and inhabited by a few "caboclos." The terrain and flora show
the same characteristics as those of Tauary. Birds and mammals
abundant.

Rio Tapajoz: Boim

A small town situated opposite Tauary on the west bank of the Rio
Tapajoz. A port where a few small boats stop, with a little commercial
activity. Climate healthy; dry land up to the bank of the river; the
forest destroyed on the river bank, but intact within. Flora notable
for the abundance of "Castaneros" (Berthileta excelsa). Birds and
mammals relatively abundant.

Rio Tapajoz: Pinhel

A small town, almost abandoned, but shown on almost all maps;
located about 29 miles from Santarem on the west bank of the Tapajoz.
It is surrounded by relatively level land covered with dense vegeta-
tion; beaches are exposed on the river bank at all seasons. Birds and
mammals rare.
Rio Amazonas: Lago Grande

An extensive region, situated on the south shore of the Amazon
between the cities of Obidos and Santarem. A central region where
cattle raising is the principal occupation; low and swampy land in the
wet season ; the flora extensively destroyed because of the great natural
and artificial fields that circle the region; inhabited by quite a few
people; birds very common; mammals almost never found.

Rio Amazonas: Boca do Igarape Piaba

A locality situated at the entrance of the parana de Obidos (going
up the river) about 12 miles from the city of Obidos on the north shore
of the Amazon. The terrain completely level and swampy in the wet
season ; the flora typical of the lowlands of the Amazon, i.e. the vege-
tation low and sickly, destroyed in some places where artificial fields
for cattle raising have been made; birds and mammals abundant but

92 bulletin: museum of comparative zoology

very few forms represented. No collections have previously been made
in this locality.

Rio Amazonas: Lago Jauary: Lirramento

A locality situated about 6 miles to the northeast of Boca do Igarape
Piaba. In the wet season it is a large lake called Jauary; in the dry
season it is said to dry up, only a small channel navigable for canoes
remaining. There exist, also, some other small lagoons of little im-
portance. Quite a few people live in the region. The terrain and the
flora are just like existing at Boca do Igarape Piaba. In this region,
also, no collections have previously been made.

Rio Amazonas: Lago Cuipeuz o Cuipeua

A locality situated between the Boca do Igarape Piaba and Livra-
mento to the northwest, about 8 miles away. Distribution of the land :
to the north, elevated, dry, and above water at all seasons; virgin
vegetation, abundance of chestnuts, inhabited only at the season of the
exploitation of these nuts, i.e. March to June; to the south, east and
west, both the terrain and the vegetation are exactly similar to those
of other parts of this region. No collections have previously been
made in this region.

Rio Amazonas: Igarape Matta

A locality near Igarape Piaba on an island of dry land.
Rw Tocantins: Cametd

A small city situated on the left bank of the Tocantins River; a place
sufficiently well known to the scientific world, and is found on all
maps. Various collections have been made there.
Brazil, Bclcm, Bosque

This locality includes the surroundings of the city of Belem (Para.)
on the central side, where the terrain is in part high and in part sub-
ject to floods; the forest is destroyed in some places.

Brazil, Belem: Val-de-Caes

This locality is situated about a league away on the margin of the
river, down stream; a place relatively thickly populated but with suffi-
ciently wild vegetation in the places where there are no inhabitants.
The central part of this locality is completely wild. In these woods
the peccary, penelope and other mammals and birds, which live en-
tirely in uninhabited woods, have been seen.

Brazil, Rio Acard: Villa Acard

A locality situated on a bank of the river of this name; the region
unhealthy, the population decadent and completely out of commission

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 93

because of malaria and other diseases. The place is found on all
maps. The birds killed in this locality come exclusively from the
region near to the town. On this side the terrain is high, and with
woods in which hunting is easy, but destroyed in places near the town,
and wild in the center. The side opposite the river is impenetrable due
to the many palm trees, called "moriche", in which is found Ber-
lepschia rikeri (Ridgway). Some of these palms are also found in the
town.

ECOLOGICAL AND DISTRIBUTIONAL NOTES

In marked contradistinction to those parts of tropical America
which contain mountain systems, a gigantic area of eastern South
America has a uniformly humid tropical climate. As Chapin has so
clearly shown in his study of the avifauna of the Congo, ecological
factors are consequently of primary importance in the distribution of
birds. Quite the most important of these is the presence of natural
savannahs in the sea of primeval tropical rain forest.

This subject will receive further notice, but is mentioned here first
because it explains the occurrence of two distinct Faunas in Lower
Amazonia. The principal one, of course, is the Guiana-Amazonian
Fauna to which the very great majority of the birds belong. A totally
different one is encountered the moment we reach the higher "campos"
of central Brazil or the more arid coast of Ceara and Maranhao.
Birds characteristic of this fauna occur on the great savannahs of
Marajo and other islands at the mouth of the Amazon, on certain
savannahs on the upper reaches of the Rio Tocantins, and a very few
have been collected on the north bank of the Amazon at Monte
Alegre, where the savannahs deserve much more careful exploration.
(A very few are reported locally elsewhere.) A list of these birds is
given below; all are common and widely distributed in central and
southern Brazil.

Crypturellus obsoletus griseiventris Colaptes campestris chrysosternus

undulatus adspersus Chrysoptilus melanochloros mariae

Rhynchotus rufescens catingae Leuconerpes candidus

Polyborus plancus brasiliensis Lepidocolaptes angustirostris coro-
Uropelia campestris natus *

Guira guira Casiornis rufa

Polytmus guainumbi thaumantias Xolmis cinerea
Trogon v. variegatus " velata

Nystalus maculatus Suiriri affinis

94 bulletin: museum of comparative zoology

Mimus saturninus Sporophila leucoptera mexianae
Turdus amaurochalinus " bouvreuil

Archiplanus solitarius " caerulescens

Agelaius cyanopus

Returning now to the question of ecological habitats, Snethlage
(Journ. f. Ornith., 61, 1913, pp. 469-539) gives the following classifica-
tion.

I. Low land subject to flooding in the rainy season — the varzea of
the Brazilians.

a. Virgin forest

b. Savannahs or campos
II. High ground never flooded

a. Virgin forest

b. Savannahs or campos

III. Scrub growth of at least two types, caused by the destruction
of the original forest, the capoeira.

One point to emphasize here is the almost incredibly minute differ-
ences in habitat which affect the presence or absence of many humid
tropical forest birds. Some like the two species of Automolus and the
flycatchers, Knipolegus and Phaeotriccus, are strictly riparian. The
little Sicalis Columbiana goeldii is practically confined to grassy patches
on river banks. The virgin forest can be divided into at least four
"associations:" (1) the ground floor (2) the undergrowth and shrub-
bery (3) the trunks and lower branches (4) the "crown" or tops of the
trees, where recent studies in British Guiana have proved the occur-
rence of a special insect fauna. While many birds will be found in two
or more of these divisions, some are strictly confined to each one. The
same divisions apply to the varzea forest. Here a fascinating problem
presents itself — what becomes of the birds of the first two divisions
during the season of flood?

As the rivers are the highways of travel, and the principal cities
and towns are on the rivers, it follows that the varzea birds are much
better known and more abundantly represented in collections. Com-
paratively few expeditions have penetrated to points up the rivers
to high ground and have stayed there any length of time. The rarity
of many "high ground" forest species will prove to be more apparent
than real.

With hundreds of species of birds, however, there is as yet little or
nothing on record as to their habitat preferences and seasonal changes
in preference, if any. In other cases, botanical knowledge runs ahead

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 95

of the ornithological. Near Jamunda, Rio Faro, for instance, are
stretches of grassy sand dunes with special trees and shrubs. It would
be difficult, if not impossible, to produce a list of birds characteristic
of this habitat.

Of very great importance in the distribution of the birds are the
topographic boundaries provided by the main Amazon and its prin-
cipal tributaries. Many decades ago people began to be aware that
certain birds like the Trumpeters would not cross a river. But our
knowledge today of the degree to which birds are sedentary in tropical
forests is greatly increased. The Ant-birds furnish an excellent illus-
tration of innumerable subspecies, the ranges of which are confined to
the area between two great rivers. There are, however, many excep-
tions and anomalies, and in certain cases birds are crossing the Amazon
between Obidos and Santarem and further up river near Villa Bella
Imperatriz; common on one bank, they are rare and little known on
the other, where they would appear to have gained a bare "toe-hold"
only. The following principal groups can be distinguished.

1. Species of the Guianas south to the north bank of the Amazon.
In a small number of cases they cross this river at its mouth and occur
near Para, but do not range very far up-river on the south side.

2. A group of representative species on the south side of the Amazon
balances group 1 in part.

3. Genera and species of far upper Amazonia range eastward down
river for varying distances. Many stop at the Rio Madeira, others
reach the left bank of the Rio Tapajoz. Many stop at the Rio Negro
on the north side.

4. A group of representative species and subspecies on the south
bank in Lower Amazonia balances group 3 in part.

5. Many species range much more widely, from the Guianas to
eastern Brazil or farther. When subspecific variation occurs, the main
Amazon is the usual boundary. In other cases the population in the
Amazon valley is intermediate, but sufficiently distinct from either
extreme to bear a separate name. The modern tendency is to name all
these intermediate populations.

6. A small group of endemic species, practically confined to our area
on the south side of the Amazon, not clearly representative of anything
else. Examples are Pyrrhura rhodogaster, Conopophaga roberti, and
Pipra iris.

7. It follows that the greatest number of endemic species and sub-
species occur on the south bank in our area.

8. These various groups clearly fit in to the concept of an Amazonian

96 bulletin: museum of comparative zoology

Sea in the past. A relatively recent area has been invaded from the
north, west, and south, and the invasion may still be going on.

9. The enormous avifauna is due to the sedentary nature of many
of the birds, the bounding effects of the great rivers, and the ability
of the birds to adapt themselves to relatively minute ecological
habitats.

In the systematic list beyond we have been careful to point out all
cases where birds belong to one or another of these groups.

As a matter of interest a list of the known migrants and winter
visitants is appended. Many Shore-birds, the Gray-cheeked Thrush,
the Blackpoll and Connecticut Warblers and Bobolink migrate across
the Amazon west of our area. They are perhaps following an ancestral
route west of the Amazonian Sea.

MIGRANTS OR WINTER VISITANTS

Pelecanus occidentalis from the north

Phoenicoptcrus ruber

Pandion haliaetus carolinensis

Falco peregrinm anatum

Haematopus ostralegus palUatus

Pluvialis dominica

Squatarola squatarola

Charadrius semipalmatus

wilsonia
Bartramia longicauda
Numenius hudsonicus
T ring a flavipes

" melanoleucus

" solitaria
Actitis macularia
Catoptrophorus se mipalmatus
Arenaria interpres morinella
Limnodromus griseus
Crocethia alba
Ereunetcs pusiUus
Erolia minuta
" fuscicoUis
" mclanotos
Larus atricilla
Gelochelidon nilotica

GRISCOM AND GREENWAY : BIRDS OF LOWER AMAZONIA 97

Sterna hirundo from the north

" antiUarum
Muscivora tyrannus tyrannus south

? Tyrannus albogularis

m. mclancholicus
? Empidonomus varius varius
? " aurantio-atro-cristatus minor

Myiodynastcs maculatus solitarius
Empidonaz euleri
Progne sub is subis north

Pygochclidon cyanoleuca south

Riparia riparia north

Hirundo erythrogaster

Vireo chivi chivi south

" " vividior north

" calidris barbatula
Dendroica aestiva aestiva

SYSTEMATIC LIST

Family TINAMIDAE

1 Tinamus tao tao Temminck

Type locality: Paid

Para, Cussary (Snethlage): Santarem (Chapman & Riker); Santarem & Rio
Tapajoz (Pinto)

1 d\ 1 9 1? Rio Tapajoz; 1 d\ Rio Capim

1 d\ Apacy, Rio Tapajoz (Carnegie Museum)

A widely ranging, but little known Tinamou, with a relatively re-
stricted Amazonian range. Pinto records this species from Monte
Alegre on the north bank, but it might prove to be a different sub-
species. The species reappears in Venezuela, but this apparently dis-
continuous distribution may not prove real.

2. Tinamus major major (Gmelin)

Type locality: Cayenne

Obidos (Snethlage, as T. subcristatus) ; Obidos (Conover & Pinto)
1 c? imm., 1 9 , Obidos (Carnegie Museum)

These birds agree perfectly with Cayenne topotypes in the Carnegie
Museum.

98 bulletin: museum of comparative zoology

3. Tinamus major olivascens Conover, 1937

Type locality: Toure-Assu, Rio Acara\ Para

Rio Acara" and Rio Tapajoz, left bank (Conover); right bank (Pinto, as ser-
ratus)

1 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

Apparently a rare bird on the south bank of the Amazon in our area.
It would appear to be commoner on the Rio Purus. Two specimens
from there agree with the Tapajoz bird and fully support Conover's
conclusions on this very distinct race. Birds from the Rio Madeira
would presumably belong here also, though they pass as serratus
(Spix) ; but Hellmayr's detailed comments in his study of Spix's types
shows clearly that they cannot possibly be olivascens Conover. It is
obvious that this group requires further study. At present two races
have a discontinuous distribution and stagger each other in four river
valleys on the south bank of the Amazon.

4. Tinamus guttatus Pelzeln

Type locality: Borba, Rio Madeira

Marajo Island (Hellmayr); vicinity of Para (Hellmayr, Snethlage, Stone);
Santarem (Chapman & Riker)

1 9 , Benevides, 2 d% Santarem, 2 9 , Rio Tapajoz, left bank (Carne-
gie Museum).

This species is also known from far upper Amazonia, where speci-
mens should prove separable, judging by Salvadori's comments. (Cat.
Birds Brit. Mus., 27, p. 508).

There is still an interesting problem in the life histories of these big
Tinamous. No one seems to know how they divide the territory be-
tween them.

5. Crypturellus cinereus (Gmelin)

Type locality: Cayenne

Monte Alegre (Snethlage); Marajo Island (Snethlage); vicinity of Para
(numerous collectors); Santarem (Chapman & Riker)

1 9 , Benevides; 2 9 , Rio Tapajoz, left bank (Carnegie Mus.)

A relatively rare species, reaching its southern limits on the south
bank of the Amazon.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 99

6. Crypturellus obsoletus griseivektris (Salvadori)

Type locality: Santarem

Reported only from the type locality, and Caxiricatuba, Rio Tapajoz (Pinto)

1 d\ Rio Tapajoz, Pinhy

3 c? 4 9 , Santarem (Carnegie Mus.)

A rare and little known subspecies of a widely ranging species from
central Brazil southward.

7. Crypturellus soui soui (Hermann)

Type locality: Cayenne

Rio Jamunda (Faro), Cussary, Obidos (Snethlage); Obidos (Pinto)
1 C? 1 9 , Obidos (Carnegie Mus.)

The Carnegie Museum possesses 1 c? 3 9 from French Guiana, which
definitely represent typical soui. The two birds from Obidos agree
with these in that the female especially is a very rich tawny, rufescent
and buffy bird, and soui would appear to be the extreme in this direc-
tion of all the South American races.

8. Crypturellus soui decolor Griscom and Greenway, 1937

Type locality: Pinhy, right bank Rio Tapajoz, Para

Para (Elliott) in Brit. Mus.); numerous localities near Para (Hellmayr,
Snethlage, Stone, Pinto as soui); Boim, Rio Tapajoz (Snethlage); San-
tarem (Chapman & Riker)
1 d\ Rio Tapajoz, Pinhy

1 d\ Benevides, 3 cT 1 9 , Santarem, 2 9 , Rio Tapajoz, left bank
(Carnegie Mus.)

As pointed out in our original description Hellmayr had already
characterized this subspecies, but lack of typical soui made it im-
possible for him to go further. It proves to be strikingly different from
typical soui in that both sexes are duller colored, less tawny above; the
female paler and more ochraceous, less tawny below; males, as usual,
less different than females, but greyer and browner, less buffy and
ochraceous.

There are still many complications and uncertainties, however, in
the variations of this species in most of Amazonia. By inference only
from Hellmayr's comments, British Guiana specimens are not quite
the same thing as typical soui. Further extended comments by him on
birds from both banks of the Rio Madeira show clearly that these

100 bulletin: museum of comparative zoology

birds have nothing to do with decolor, but differ from British Guiana
specimens only in the color of the upper tailcoverts. They have since
been described as hoffmannsi Brabourne and Chubb, type from the
left bank. It will be apparent, therefore, that there may well be one or
possibly two valid subspecies here, when real series can be combined,
and that specimens from the north bank of the Amazon from Faro
might be one or another of these subspecies, rather than soui. Further
uncertainties develop as we proceed westward. Specimens from the
Rio Purus in the Carnegie Museum are once more dull and pale like
decolor, but on the Rio Solimoes we find another rich tawny bird, ex-
ceedingly close to soui!

It should now be clear why we are not rash enough to refer Rio
Purus specimens to decolor. It is certainly striking that the variations
indicated are exactly the same geographically as those pertaining to
Tinamus major.

9. Crypturellus undulatus adspersus (Temminck)

Type locality : State of Para, Brazil

"Para" (old specimens in Mus. Berlin and Monaco, fide Hellmayr); Rio
Maecuru and Rio Tapajoz (Snethlage) ; Rio Tapajoz (Pinto)
1 c? 1 9 , Rio Acara, Buenos Aires
1 a" 3 9 , Rio Tapajoz, Pinhy and Tauary (do.)

3 cT, Santarem (Carnegie Mus.)
1 c? 1 9 , Miritituba and Goyana Isl., Rio Tapajoz (do.)

This subspecies is the northeasternmost of the many races of
undulatus, ranging north to the upper Rio Branco and southern
British Guiana, and west to the Rio Madeira. The species is primarily
one of campos country or savannahs rather than heavy rain forest
which accounts for its absence from the vicinity of Para. According
to Hellmayr, birds from Borba on the right bank of the Rio Madeira
are practically identical with Temminck's type. Our two birds from
the Rio Acara, the easternmost point of definite record, are more
heavily barred on the neck than the Tapajoz series.

Madame Snethlage recorded japura (Spix) from the Rio Tapajoz
erroneously. It is not known east of the Rio Solimoes. From the left
bank of the Rio Madeira to the Rio Purus, we find confusus Brabourne
and Chubb. Seven specimens in the Carnegie Museum are more
heavily barred and vermiculated with blackish above, and on the
average somewhat greyer, less buffy, on flanks and under tailcoverts.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 101

10. Crypturellus variegatus variegatus (Gmelin)

Type locality: Cayenne
Obidos (Hellmayr, Snethlage)

1 c? 2 9 , Obidos (Carnegie Mus.)

11. Crypturellus variegatus transamazonicus Todd, 1937

Type locality: Santarem, Brazil

Vicinity of Para ; numerous records (Natterer, Hellmayr, Snethlage) ; mouth
of Rio Tapajoz (Pinto)

9, Santarem; Rio Tapajoz, left bank, 2 (Carnegie Mus.)

A very distinct subspecies, the separability of which was predicted
by Hellmayr (Novit. Zool., 1905, p. 305), although some of the
differences he noted are matters of age or individual variation. We
are, of course, familiar with the material in the Carnegie Museum, the
basis for Todd's valuable review of this species. To us, however, the
birds from the Rio Purus and Rio Solimoes are just half way between
the present form and salvini of east Ecuador, and we would not refer
them to transamazonicus for a moment. They may well be left un-
certain, until the type of variation on the Rio Madeira becomes known.

12. Crypturellus noctivagus erythropus (Pelzeln)

Type locality: Manaos, as restricted by Zimmer, 1938

Obidos (Hellmayr, Zimmer, Pinto); Rio Jamunda, Faro Snethlage, Zimmer)
10 d\ Obidos (Carnegie Mus.)

The series in the Carnegie Museum had enabled us to reach the
identical conclusions, just published by Zimmer, 1938, which show
that dissimilis Salvadori from British Guiana is a synonym.

13. Crypturellus noctivagus strigulosus (Temminck)

Type locality: Para

Near Para; numerous localities (Natterer, Goeldi, Hellmayr, Snethlage, Stone,

Pinto); Rio Acara (Snethlage); Rio Tapajoz (Pinto); Rio Tocantins and

Rio Xingu (Zimmer)

1 d" 2 9 ad., 1 9 irara., Rio Tapajoz, Pinhy

2 <? 1 9 ad., 1 d* juv., Santarem (Carnegie Mus.)
2 cf, Rio Tapajoz (do.)

By now a relatively well known Tinamou, ranging to the right bank

102 bulletin: museum of comparative zoology

of the Rio Madeira. Westward it is replaced by hellmayri Brabourne
and Chubb on the left bank of the same river, a subspecies still prac-
tically unknown. The species is apparently lacking on the Rio Purus
and the Rio Solimoes, and on the north bank of the Amazon, west of
the Rio Negro.

14. Crypturellus parvirostris (Wagler)

Type locality: Bahia, suggested by Hellmayr
Marajo Island (Snethlage); Santarem (Chapman and Biker).
1 d1 1 9 , Santarem (Carnegie Mus.)

A characteristic species of the "campo" of central and southern
Brazil, locally penetrating north to the south bank of the Amazon.

15. Rhynchotus rufescens catingae Reiser

Type locality: B,io Parnahyba, Piauhy

One specimen from Marajo Island (Pinto Beixoto, 1923)

This record would appear to have been completely overlooked. The
bird is referred to catingae on purely geographic grounds ; the validity of
the subspecies has been questioned, at least in part (cf. Pinto, 1938).

Family COLYMBIDAE

16. Poliocephalus dominicus speciosus (Arribalzaga)

Type locality : Buenos Aires

Cajutuba near Bara (Natterer) ; Monte Alegre near Bara, (Snethlage)
1 9 , Bio Tapajoz, Santarem (M.C.Z.)

Family PHALACROCORACIDAE

17. Phalacrocorax olivaceus olivaceus (Humboldt)

Type locality: near Banco, Bio Magdalena, Colombia

Marajo Island (Snethlage) ; Ucurituba, near Santarem (Hellmayr) ; Cajutuba
near Bara (Natterer, Graham)

1 d1 ad., Tapajoz Biver, Santarem

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 103

Family ANHINGIDAE
18. Anhinga anhinga anhinga (Linnaeus)

Type locality: Brazil

Rio Capim near Para (Hellmayr); Marajo Island (Snethlage); Para (Pinto).

2 cf ad., 2 c? imra., 1 9 ad., 1 9 imm., 1 ? imm.; Rio Tapajoz,

various localities

The arrival of this series from Brazil, the type locality, with two
others from the lower Amazon, made it possible to compare birds from
the United States for the first time. These last prove to average
smaller, with markedly shorter bills. There is also an excellent color
character. Brazilian birds have a much broader tail tip, at least
double that of North American birds, and on the outer tail feathers
much more than this. Intergradation is about as follows. An east
Ecuador specimen has a bill as long as Brazilian birds, but a short
wing, and the tail tip is intermediate. Panama and Yucatan birds
resemble Florida specimens in size, but have an intermediate tail-tip.
Cuban birds are like Florida series.

As it is more than possible that extreme western and northwestern
Brazilian birds might prove to be intermediate, like the east Ecuador
bird discussed above, we designate Rio Tapajoz, Para, Brazil as a
restricted type locality. The northern race will be known as Anhinga
anhinga leucogaster (Vieillot), type locality, Florida..

cf- 9

Brazil —wing 340-353; bill, 93-96 349 90.5

Florida— wing 322-345; bill, 80-88 323-338; 75-87

Family PELECANIDAE
19. Pelecanus occidentalis occidentalis (Linnaeus)

Type locality: Jamaica

Itaituba, Rio Tapajoz (Snethlage)

A very surprising record, as the Brown Pelican is unknown in South
America, except along the coast of Venezuela and the Guianas.

N.B. Frigate-Birds (Fregata) and Boobies (Sula ssp.) surely occur off the
coast of Para, but there are no definite records.

104 bulletin: museum of comparative zoology

Family ARDEIDAE

20. Ardea cocoi Linnaeus

Type locality: Cayenne

Mexiana Island (Hagmann and Snethlage); Marajo Island (Pinto-Peixoto) ;
Pataua and Lago Cuipeua (Pinto)

1 d ad., Rio Tapajoz (M. C. Z.)

2 9 , Lago Grande (do.)

21. Casmerodius albus egretta (Gmelin)

Type locality: Cayenne

Rio Capim (Goeldi); Mexiana Island (Hagmann and Snethlage); Santarem
(Chapman and Riker, Marajo Island (Pinto-Peixoto)

2 d, Rio Amazon, Lago Jauary, Livramento (M.C. Z.)
1 9 , Rio Tapajoz, Santarem (do.)

22. Leucophoyx thula thula (Molina)

Type locality: Chili

Mexiana Island (Hagmann and Snethlage); Santarem (Chapman and Riker)
1 d, R. Amazon, Lago Jauary (M. C. Z.)

23. Florida caerulea (Linnaeus)

Type locality: Carolina

Mexiana Island (Hagmann and Snethlage); Marajo Island (Snethlage);
Pataua and Lago Cuipeua (Pinto)

1 d imm., R. Amazon, Lago Jauary (M. C. Z.)

1 d ad., R. Amazon, Lago Jauary (do.)

1 9 ad., R. Amazon, Lago Grande (do.)

24. Hydranassa tricolor tricolor (P.L.S. Muller)

Type locality: Cayenne

Cajutuba, near Para (Natterer); Capanema (Snethlage)

1 d, R. Amazon, Boca do Igarape Piaba (M. C. Z.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 105

25. Agamia agami (Gmelin)

Type locality: Cayenne

Para (Layard and Snethlage); Monte Alegre (Snethlage); Rio Tapajos (Sneth-
lage)

Specimen A^,Bc?,R. Tapajoz, Pinhy (May) (M. C. Z.)
Specimen Cc? R- Tapajoz, Caxiricatuba (July) (do.)

Specimen D ? R. Tapajoz, Tauary (Oct.) (do.)

Specimen Ed" R- Tapajoz, Tauary (Nov.) (do.)

Specimen F cf R. Tapajoz, Tauary (Aug.) (do.)

G imm. d\ R- Tapajoz, Pinhy (do.)

H imm. a", R- Amazonas, Lago Grande (do.)

1 9 imm., Santarem (Carnegie Mus.)

Specimens C and E are very much bluer and less green on the back
and wing coverts than other adult specimens. It would seem probable
that there is an intermediate stage of plumage which is characterized
by the green back. A specimen from middle America in the collections
of the Museum of Comparative Zoology has new blue feathers molting
in on the back, while the greener feathers remaining on the upper back
appear to be falling out.

Specimen F has the brown underparts streaked with buff.

26. Nycticorax nycticorax hoactli (Gmelin)

Type locality: Valley of Mexico

Para; Marajo and Mexiana Islands (all Snethlage)

1 d1 imm., R. Amazon, Lago Jauary (M. C. Z.)

19 R. Amazon, Lago Jauary (do.)

27. Nyctanassa violacea cayennensis (Gmelin)

Type locality: Cayenne
Marajo Island (Snethlage)

28. Pilherodius pileatus (Boddaert)

Type locality: Cayenne

Mexiana Island (Hagmann and Snethlage) ; Monte Alegre (Snethlage)
1 imm. 9 , R. Tapajoz, Caxiricatuba (Aug.) (M. C. Z.)

The wings are whiter, not as gray, and the feathers of the neck not
as long or as decomposed as the mature bird.

106 bulletin: museum of comparative zoology

29. Butorides striatus striatus (Linne)

Type locality: Surinam

Para (Stone); Rio Capim (Goeldi); Mexiana Island (Hagmann); numerous
localities from Para to the mouth of the Amazon (Snethlage); Santarem
(Chapman and Riker)

A long series from Rio Tapajoz
1 c? Santarem (Carnegie Mus.)

30. TlGRISOMA LINEATUM LINEATUM (Boddaert)

Type locality: Cayenne

Para (Layard and Hellmayr) ; Rio Capim (Goeldi) ; Mexiana Island (Hagmann)
Ilha das Oncas and Marajo Island (Snethlage) ; Monte Alegre (Snethlage) ;
Santarem (Chapman and Riker)

lcf R. Amazonas, Lago Jauary, Livramento (M. C. Z.)
1 cf R. Amazonas, Boca do Igarape Piaba (do..)
1 9 , Santarem (Carnegie Mus.)

31. IXOBRYCHUS EXILIS ERYTHROMELAS (Vieillot)

Type locality: Paraguay

Para, Marajo Island, Monte Alegre, Cussary (Snethlage)

1 cf, Santarem (Carnegie Mus.)

32. Zebrilus undulatus (Gmelin)

Type locality: Cayenne

Para, Rio Tocantins (Aramatheua), Cussary, Rio Jamunda (Faro), all Sneth-
lage; Santarem (Chapman and Riker), Para (Pinto)

2 d\ Obidos; 1 cf , Santarem (Carnegie Mus.)

33. Botaurus pinnatus (Wagler)

Type locality: Bahia

1 d\ Rio Tapajoz, Tauary (M. C. Z.)

This specimen has the brown streaks on the neck and upper breast
somewhat lighter than two specimens from Colombia in this museum.

This rare bittern has never before been recorded from Amazonian
Brazil.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 107

Family COCHLEARIIDAE
34. Cochlearius cochlearius cochlearius (Linnaeus)

Type locality: Guiana

Rio Capim (Goeldi); Mexiana Island (Hagmann and Snethlage); near Para

(Snethlage-Stone) ; Marajo Island (Snethlage); Rio Tocantins, Cameta

(Snethlage) ; Utinga and Lago Cuipeua (Pinto)

1 d\ R. Amazon, Lago Jauary, Livramento (M. C. Z.)

Family CICONIIDAE
35. Mycteria Americana Linnaeus

Type locality: northeastern Brazil

Mexiana Island (Hagmann); Marajo Island (Snethlage)

36. Euxenura galeata (Molina)

Type locality: Chile

Mexiana Island (Hagmann) ; Marajo and Mexiana Islands (Snethlage)

37. Jabiru mycteria (Liehtenstein)

Type locality: Brazil, ex Marcgrave

Mexiana Island (Wallace and Hagmann); Marajo Island (Snethlage)

Family THRESKIORNITHIDAE

38. Theristicus caudatus caudatus (Boddaert)

Type locality: Cayenne

Mexiana Island (Hagmann); Marajo and Mexiana Islands (Snethlage), Lago
Cuipeua (Pinto)

39. Mesembrinibis cayennensis (Gmelin)

Type locality: Cayenne

Mexiana Island (Hagmann and Snethlage); Rio Tapajoz, Goyana (Snethlage)

1 9 , Rio Acara, Acara (M. C. Z.)

4 cf 1 9, Rio Tapajoz, Caxiricatuba and Tauary (do.)

108 bulletin: museum of comparative zoology

40. Phimosus infuscatus nudifrons (Spix)

Type locality: Rio Sao Francisco, Brazil

Rio Inhangapy (sight record by Bond and de Schauensee) Stone, 1928

41. Guara rubra (Linnaeus)

Type locality: Guiana

Cajutuba, near Para (Natterer); Mexiana Island (Hagmann and Snethlage);
Marajo Island (Hellmayr, Pinto, and Snethlage)

42. Ajaia ajaja (Linnaeus)

Type locality: South America

Cajutuba, near Para (Natterer); Mexiana Island (Hagmann); Marajo Island
(Snethlage)

Family PHOENICOPTERIDAE

43. Phoenicopterus ruber Linnaeus

Type locality: Jamaica

Cajutuba (Natterer); Macapa, 19 3?, collected by J. de Cavianna (Snethlage)

Family ANHIMIDAE

44. Anhima cornuta (Linnaeus)

Type locality: Brazil

Rio Maracana, Livramento, and Peixe-Boi (Snethlage) ; near Para (Stone, sight
record)

1 9 , Obidos (Carnegie Mus.)

Family ANATIDAE

45. Cairina moschata (Linnaeus)

Type locality: Brazil

Rio Capim (Goeldi); Mexiana Island (Hagmann); Marajo Island (Snethlage);
Cunany (Snethlage); Santarem (Chapman and Riker)

46. Sarkidiornis sylvicola Ihering

Type locality: Brazil

Atlantic coast of Marajo Island and Maguary (Goeldi)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 109

47. Dendrocygna viduata (Linnaeus)

Type locality: Cartagena, Colombia
Marajo Island (Goeldi)

48. Dendrocygna bicolor bicolor (Vieillot)

Type locality: Paraguay

Marajo Island (Goeldi and Snethlage); Maguary (Goeldi)

49. Dendrocygna autumnalis discolor Sclater and Salvin

Type locality: Maroni River, Surinam

Marajo Island (Hellmayr and Snethlage); Mexiana Island (Hagmann); PanS
(Pelzeln); Obidos (Hellmayr); Santarem (Chapman and Riker); Lago
Cuipeua (Pinto); Caviana Island (Brodkorb)

3 d1 3 9 , Rio Tapajoz, Pinhy and Caxiricatuba (M. C. Z.)
1 9 , Santarem (Carnegie Mus.)

50. Neochen jubata (Spix)

Type locality: Rio Solimoes, Brazil
Mexiana Island (Hagmann)

1 6" , Santarem (Carnegie Mus.)

51. Anas brasiliensis Gmelin

Type locality: Brazil

Para (Snethlage) ; Marajo Island (Goeldi and Snethlage)

1 cf, Rio Capim, Para (M. C. Z)

2 cf, Rio Acara, Acara (do.)
2 d71, Rio Tapajoz (do.)

1 d\ north bank of Amazon near Obidos (do.)
8 cf 1 9 , Apagy, Rio Tapajoz (Carnegie Mus.)

52. Anas bahamensis Linnaeus

Type locality: Bahamas

Cajutuba, near Para (Natterer); Marajo Island (Goeldi)

53. Nomonyx dominicus (Linnaeus)

Type locality: South America
Rio Acara (Snethlage)

110 bulletin: museum of comparative zoology

Family CATHARTIDAE

54. Sarcorhamphus papa (Linnaeus)

Type locality: Brazil

Para (Goeldi) ; Rio Maraca (Snethlage) ; Santarem (Chapman and Riker)

55. Coragyps atratus foetens (Liechtenstein)

Type locality: Paraguay

Para (Wallace, Layard, Snethlage, Stone); Rio Capim (Goeldi); Mexiana
Island (Hagmann)

56. Cathartes aura ruficollis Spix

Type locality: Interior of Bahia and Piauhy

Para (Snethlage) ; Marajo Island (Snethlage) ; Mexiana Island (Hagmann and
Snethlage) ; Santarem (Chapman and Riker)

1 9 , Rio Tapajoz, Tauary; 1 cf Santarem (Carnegie Mus.)

57. Cathartes urubitinga Pelzeln

Type locality: Southern and central Brazil

Rio Capim (Goeldi); Marajo Island (Snethlage); Para, Rio Guama and Rio
Inhangapy (Stone); Caviana Island (Brodkorb).

Family ACCIPITRIDAE

58. Elanoides forficatus yetapa (Vieillot)

Type locality: Paraguay

Para (Layard and Natterer); Rio Capim (Goeldi); Peixe-Boi and Quati-puru
(Snethlage); Mexiana Island (Hagmann); near Santarem (Hellmayr)

59. Odontriorchis palliatus (Temminck)

Type locality: Southern Bahia

Para and Marajo Island (Snethlage); Rio Acara (Hellmayr); Santarem (Chap-
man and Riker, Pinto)

1 cT 1 9 , Rio Tapajoz, Pinhy and Tauary (M. C. Z.)

60. Chondrohierax uncinatus (Temminck)

Type locality: near Rio

1 9 , Rio Tapajoz, Tauary (M. C. Z.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 111

61. Harpagus bidentatus bidentatus (Latham)

Type locality: Cayenne

Mexiana Island (Hagmann); Para, Marajo Island, Rio Tocantins, and Rio
Tapajoz, (Snethlage); Obidos (Hoffmans); Utinga (Pinto)

1 cf 1 9 , Rio Amazonas, Lago Cuipeuz (adults) (M. C. Z.)

2 d1 3 9 , Rio Tapajoz, Tauary (immature) (do.)
2 c? ad. 1 cf imm., Obidos (Carnegie Mus.)

1 cf imm., Santarem (Carnegie Mus.)

This nicely sexed series shows some color differences between the two
sexes when immature. Current descriptions apply to the female
primarily. The male is much less marked below, the black streaks,
rather than spots, confined to the chest except for fine shaft streaks on
the central feathers of breast and abdomen.

62. Harpagus diodon (Temminck)

Type locality: "Bresil"

Para (Hellmayr); Para and Rio Capim (Goeldi); Santarem (Chapman and
Riker)

1 d\ Apacy, Rio Tapajoz (Carnegie Mus.)

All recent check-lists overlook these records, and restrict the range
of this species to eastern and southeastern Brazil.

63. Ictinia plumbea (Gmelin)

Type locality: Cayenne

Para (Wallace) ; Peixe-Boi (Hellmayr) ; San Antonio do Prata (Snethlage) ; Rio
Tocantins, Arumatheua (Snethlage)

64. Rostrhamus sociabilis sociabilis (Vieillot)

Type locality: Corrientes

Para (Stone); Peixe-Boi, Marajo Island, Rio Tocantins (Arumatheua); all
Snethlage; Santarem (Chapman and Riker)
1 9 , Rio Tapajoz, Pinhy (M. C. Z.)

65. Helicolestes hamatus (Temminck)

Type locality: Brazil
Para (Snethlage)

4 c? 1 9 , Obidos (Carnegie Mus.)

112 bulletin: museum of comparative zoology

66. Elanus leucurus leucurus (Vieillot)

Type locality : Paraguay
Marajo Island (series, Snethlage)

67. Accipiter bicolor bicolor (Vieillot)

Type locality: Cayenne

San Antonio, near Para (Hellmayr); Bemfica (Steere); Santarem (Chapman
and Riker)

1 cf , Santarem (Carnegie Mus.)

68. Accipiter pectoralis (Bonaparte)

Type locality: Brazil
Par&, 1 9 (Snethlage)

69. Accipiter superciliosus superciliosus (Linnaeus)

Type locality: Surinam

Pard (Natterer and Stone); Benevides and Peixe-Boi (Snethlage)
1 9, Rio Tapajoz, Santarem (M. C. Z.)

70. Heterospizias meridionalis meridionalis (Latham)

Type locality: Cayenne

Mexiana Island (Wallace, Hellmayr and Snethlage); Marajo Island (Sneth-
lage) ; Rio Xingil, Victoria (Snethlage) ; Santarem (Chapman and Riker)

71. Buteo albicaudatus albicaudatus Vieillot

Type locality: Rio de Janeiro

Marajo Island (Snethlage); also by Brodkorb as colonus Berlepsch
1 tf\ Rio Tapajoz, Tauary (M. C. Z.)

The two subspecies meet in our area, and perhaps birds will prove to
vary individually towards both.

72. Buteo albonotatus abbreviatus Cabanis

Type locality: Pomeroon River, British Guiana
Marajo Island (Snethlage)

73. Buteo brachyurus Vieillot

Type locality: Cayenne

Para (Natterer); Nazare (Layard)

GRISCOM AND GREENWAY : BIRDS OF LOWER AMAZONIA 113

74. Buteo magnirostris magnirostris (Gmelin)

Para (Spix, Layard, Stone, Snethlage); Cajutuba (Natterer); Rio Capim
(Goeldi); Mexiana Island (Wallace, Hagmann and Snethlage); Marajo
Island (Snethlage); Santarem (Chapman and Riker); Obidos (Hellmayr);
Caviana Island (Brodkorb)

1 <? 1 9, Rio Tapajoz, Tauary (M. C. Z.)
1 d", Rio Tapajoz, Caxiricatuba (do.)
1 9 , Rio Amazonas, Lago Cuipeuz (do.)

1 9 , Rio Amazonas, Boca do Igarape Piaba (do.)

2 9 , Obidos and 1 d" 2 ?, Santarem (Carnegie Mus.)

This series is readily referable to the typical race, but differs from a
fine series from Surinam in averaging far more rufescent below, thereby
approaching nattereri (Sclater and Salvin) to some extent. On geo-
graphical grounds this is to be expected, as Hellmayr (Orn. of North-
eastern Brazil, p. 460) refers birds from Maranhao and Ceara to
nattereri.

75. Asturina nitida nitida (Latham)

Type locality : Cayenne

Para (Natterer, Snethlage, Stone); San Antonio (Hellmayr); Peixe-Boi, Ara-

piranga and Marajo Island (Snethlage); Santarem (Chapman and Riker);

Rio Tapajoz (Pinto)

1 cf imm., Santarem (Carnegie Mus.)

76. Leucopternis albicollis albicollis (Latham)

Type locality: Cayenne

San Antonio (Hellmayr) ; Rio Capim and Rio Tocantins (Snethlage) ; Itaituba
(Pinto)

1 d\ Obidos (Carnegie Mus.)

77. Leucopternis melanops (Latham)

Type locality: Cayenne
Para (Natterer)

1 c?, Rio Amazonas, Lago Cuipeuz (M. C. Z.)

1 d\ Obidos (Carnegie Mus.)

78. Leucopternis kuhli Bonaparte

Type locality: Para

Para (Natterer and Wallace); San Antonio (Hellmayr); Igarape-Assu (Hell-
mayr) ; Peixe-Boi (Snethlage) ; Rio Tapajoz (Pinto)
1 9 , Villa Braga, Rio Tapajoz (Carnegie Mus.)

114 bulletin: museum of comparative zoology

79. Leucopternis schistacea (Sundevall)

Type locality: Brazil

Rio Capim (Goeldi) ; Para, Maraca (Snethlage)
1 cf, Benevides, Para (Carnegie Mus.)

80. Hypomorphnus urubitinga urubitinga (Gmelin)

Type locality: Brazil

Rio Capim, Para (Goeldi); Mexiana Island (Wallace, Hagmann, Snethlage);

Marajo Island, Maraca, Cussary (Snethlage); Santarem (Chapman and

Riker, Pinto); Pataua (Pinto)

4 d\ Rio Tapajoz, Boim and Caxiricatuba (M. C. Z.)

81. Buteogallus aequinoctialis (Gmelin)

Type locality: Cayenne

Cajutuba (Natterer); Marajo Island (Snethlage)

82. Busarellus nigricollis nigricollis (Latham)

Type locality: Cayenne

Mexiana Island (Wallace, Hagmann); Marajo Island and Cussary (Snethlage);

Santarem (Chapman and Riker) ; Obidos (Hellmayr) ; Urucurituba (Hell-

mayr); Para (Stone); Pataua, Lago Cuipeua (Pinto)
1 9, Lago Grande (M. C. Z.)

83. Harpia harpyja (Linnaeus)

Type locality: Mexico

Para (Natterer); Castanhal (Stone); Peixe-Boi, Rio Guama, Rio Capim, Rio
Tapajoz (Snethlage)

84. Spizastur melanoleucus (Vieillot)

Type locality: Guiana

1 d\ Rio Tapajoz, Tauary (M. C. Z.)
1 cf , Obidos (Carnegie Mus.)

Apparently the only records for the Amazon Valley.

85. Spizaetus ornatus (Daudin)

Type locality: Cayenne

Santarem (Natterer, Chapman and Riker)

1 d\ Rio Tapajoz, Tauary (M. C. Z.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 115

86. Spizaetus tyrannus (Wied)

Type locality: Bahia, Brazil

Para (Layard); Rio Capim (Wallace); Marajo Island and Rio Jamauchim
(Snethlage)

1 d\ Villa Braga, Rio Tapajoz (Carnegie Mus.)

87. Circus buffoni (Gmelin)

Type locality: Cayenne

Santarem (Natterer); Marajo Island (Snethlage)

88. Geranospiza caerulescens (Vieillot)

Type locality: Cayenne

Marajo Island and Rio Cussary (Snethlage) ; Santarem (Chapman and Riker)

1 9 , Rio Amazonas, Lago Jauary (M.C.Z.)

1 9 , Obidos (Carnegie Mus.)

Birds from the south bank may prove referable to gracilis (Tem-
minck).

89. Pandion haliaetus carolinensis (Gmelin)

Type locality: Carolina
Marajo Island (Snethlage)

1 d% Rio Tapajoz, Santarem, Oct. 17, 1932 (M. C. Z.)

Family FALCONIDAE

90. Herpetotheres cachinnans cachinnans (Linnaeus)

Type locality: Surinam

Mexiana Island (Wallace, Hagmann, Snethlage); Marajo Island (Snethlage);
Rio Inhangapy (Stone); Santarem (Natterer); Caviana Island (Brodkorb).

1 9 , Rio Amazonas, Boca do Igarape Piaba (M. C. Z.)

1 d\ Obidos (Carnegie Mus.)

91. MlCRASTUR SEMITORQUATUS SEMITORQUATUS (Vieillot)

Type locality: Paraguay

Monte Alegre (Snethlage) ; Santarem (Pinto)

1 c? ad., Rio Amazonas, Lago Cuipeuz (M. C. Z.)

1 cf imm., Rio Tapajoz, Caxiricatuba (do.)

1 9 imm., Rio Tapajoz, Tauary (do.)

1 cf imm., Santarem (Carnegie Mus.)

116 bulletin: museum of comparative zoology

92. MlCRASTUR MIRANDOLLEI MIRANDOLLEI (Schlegel)

Type locality: Surinam

Para (Natterer) ; Rio Guama, Ourem (Snethlage)

1 9 , Rio Acara, Villa Acara, fully adult. (M. C. Z.)

1 9 , Villa Braga, Rio Tapajoz (Carnegie Mus.)

93. MlCRASTUR RUFICOLLIS GILVICOLLIS (Vieillot)

Type locality: Cayenne

Para (Natterer, Wallace, Stone); Para, Rio Xingu (Victoria), Rio Curua, Rio
Jary (Snethlage)

1 cf, Rio Tapajoz, Tauary (M. C. Z.)

1 d\ Obidos (Carnegie Mus.)

1 C? 1 9 , Benevides (do.)

2 9 , Villa Braga, Rio Tapajoz (do.)

94. Daptrius ater Vieillot

Type locality: Brazil

Para (Wallace); Cussary and Rio Jaumachim (Snethlage); Santarem (Pinto)

3 cf 1 9 , Rio Tapajoz, Caxiricatuba (do.)

1 9 , Rio Tapajoz, Pinhy
1 d* , Obidos XCarnegie Mus.)
1 cf, Santarem (do.)

One of these specimens is immature and represents Gymnops
fasciatus Spix.

95. Daptrius americanus americanus (Boddaert)

Type locality: Cayenne

Para (Wallace, Stone) ; Rio Capim (Goeldi) ; Peixe-Boi, Rio Guama, Rio Moju,
Rio Jamauchim (Snethlage)

1 cf, Obidos (Carnegie Mus.)
1 d\ Santarem (do.)

96. MlLVAGO CHIMACHIMA CHIMACHIMA (Vieillot)

Type locality: Paraguay

Rio Capim (Goeldi); Mexiana Island (Wallace, Hagmann); Marajo Island
(Snethlage); Santarem (Chapman and Riker)

1 9 , Rio Tapajoz, Caxiricatuba (M. C. Z.)

1 cf , Obidos (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 117

97. Milvago chimachima PALUDiVAGUS Penard

Type locality:
Amapa (Snethlage)

2 9 , Boca do Igarape-Piaba, near Obidos

1 cf , Obidos (Carnegie Mus.)

These birds are really intermediate between the two subspecies.
They are decidedly blacker above, like paludimgus, but only faintly
buffier, less white below. We have no idea to which subspecies speci-
mens from Mexiana and Marajo will prove to belong.

98. Polyborus plancus brasiliensis (Gmelin)

Type locality: Brazil

Mexiana Island (Wallace, Hagmann); Marajo Island (Snethlage); Santarem
(Chapman and Riker); Caviana Island (Brodkorb)

1 9 ad., Rio Amazonas, Lago Grande (M. C. Z.)

1 cT 1 9 irnm., Rio Tapajoz, Santarern (do.)

99. Gampsonyx swainsonii leonae Chubb

Type locality: Leon, Nicaragua

Braganza, Monte Alegre, Cussary (Snethlage); Santarem (Chapman and
Riker)

4 cf 4 9, Rio Tapajoz, Santarem (M. C. Z.)

1 d" 1 9 , Obidos (Carnegie Mus.)

2 cf , Santarem (do.)

This series agrees with a long series from Venezuela, British Guiana
and Colombia in averaging more rufescent on thighs, flanks and sides
than four birds from southeastern Brazil. The same slight average
difference in the series in the American Museum of Natural History
induced Miller and Griscom (Amer. Mus. Novit., no. 25, 1921, p. 13
and errata) to recognize two races in 1921 out of the three maintained
by Swann at that time. Our series from the eastern Amazon is not
typical of northern birds, one or two being practically indistinguishable
from Bahia birds, and none as extreme as the most rufescent birds from
Venezuela.

100. Falco peregrinus anatum Bonaparte

Type locality: New Jersey

Cajutuba, mangrove swamp, March, 1835 (Natterer)

118 bulletin: museum of comparative zoology

101. Falco deiroleucus Temminck

Type locality: Santa Catharina, Brazil

Para (Layard); Mara jo Island (Snethlage); Santarem (Chapman and Riker,
also Ribeiro)

1 9 , Santarem (Carnegie Mus.)

102. Falco albigularis albigularis Daudin

Type locality : Cayenne

Para (Hellmayr, Stone) ; Rio Capim (Goeldi) ; Peixe-Boi, Cunany, Rio Jamau-
chim (Snethlage); Santarem (Chapman and Riker); Obidos (Hellmayr)
1 c? 1 9 , Rio Tapajoz, Boim and Caxiricatuba (M. C. Z.)

1 9 , Obidos (Carnegie Mus.)
1 d* 1 9 , Santarem (do.)

1 d71, Villa Braga, Rio Tapajoz (do.)

It seems to us that albigularis Daudin applies definitely to this
species.

103. Falco fuscocaerulescens fuscocaerulescens Vieillot

Type locality: Paraguay

Mexiana Island (Wallace, Hagmann); Marajo Island (Snethlage)
1 9 , Rio Tapajoz, Santarem (M. C. Z.)

Family CRACIDAE

[Nothocrax urumutum (Spix)

This genus is characteristic of Upper Amazonia. The only record
for our area is Snethlage's "Para, (in the zoological garden)". As
Goeldi, however, states that the birds seen by him in the zoological
garden at Para came from the Rio Javary (in uppermost Amazonian
Brazil), more satisfactory evidence of the occurrence of this curassow
in our area is required.]

104. Mitu mitu (Linnaeus)

Type locality: northeastern Brazil

Para (Natterer); Rio Capim (Goeldi); Rio Muraiteua (Stone); Rio Acara
(Hellmayr)

2 cf , Rio Tapajoz, Tauary (M. C. Z.)

1 cf , Villa Braga, Rio Tapajoz (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 119

This is the characteristic Mitu of the Guiana-Lower Amazonian
region. On the Rio Negro northwestward it is replaced by M . tomen-
tosa (Spix) and in Amazonian Ecuador by M. salvini (Reinhardt).

105. Crax nigra Linnaeus

Type locality: Guiana

Pataua and Cuipeua, on the north bank (Pinto)
1 cf 1 9 , Obidos (Carnegie Mus.)

This is apparently the only evidence to show that nigra ranges south
to the Amazon. For that matter Dr. Oliveiro Pinto's record of fascio-
lata from Obidos is also the only evidence of that "species" on the
north bank.

106. Crax fasciolata fasciolata Spix

Type locality : Para, Brazil

Para (Spix); Rio Capim (Goeldi); Obidos (Pinto)

107. Crax pinima Pelzeln

Type locality: Para

Cajutuba (Natterer); Rio Capim (Goeldi)

Far too little is known about these two species of Crax, which are
characteristic of Lower Amazonia. They are apparently replaced in
upper Amazonia by C. nigra Linnaeus and C. globulosa Spix. By a
remarkable piece of inadvertence the present species is omitted in
Snethlage's Aves do Amazonas. She may, however, have thought that
pinima was a pure synonym of fasciolata. Needless to say it is by no
means certain that further material and study will endorse the validity
of the four "species" here mentioned, or the identifications of some of
them in our area.

108. Penelope marail (P.L.S. Muller)

Type locality : Cayenne

Rio Jamunda, Faro (Snethlage); near Obidos (Pinto)
1 cf 1 o71 juv., 1 9 , Obidos (Carnegie Mus.)

Apparently just reaching our area on the north bank of the Amazon.
It should be expected elsewhere.

120 bulletin: museum of comparative zoology

109. Penelope superciliaris superciliaris Temminck

Type locality: State of Para, Brazil

Igarape-Assu (Robert) ; Peixe-Boi and Ipitinga (Midler) ; Rio Capim (Goeldi) ;
Rio Acara; Rio Tocantins, Arumatheua; Rio Tapajoz, Boim (all Sneth-
lage)

5 <?, 2 9 , Rio Tapajoz, various localities (M. C. Z.)

1 9 , Santarem (Carnegie Mus.)
1 cf , 1 9, Villa Braga, Rio Tapajoz (do.)
1 cf , Miritituba, do. (do.)

This is the characteristic Penelope of the south bank of the Amazon,
west to the Rio Madeira. It is replaced by other races in central and
southern Brazil, Paraguay and northeastern Argentina.

If Neumann's views as to the type locality of true superciliaris
Temminck prove correct, this subspecies will be known as pseudonyma
Neumann, (cf. Bull. Brit. Orn. Club, 53, 1933, pp. 93-95.)

110. Penelope pileata Wagler

Type locality: State of Para, Brazil

Para (H. Sieber in Berlin Mus., and Natterer, cf. Pelzeln, pp. 283, 340);
Monte Christo, Rio Tapajoz (Pinto)

1 cf 2 9 , Rio Tapajoz, Tauary (M. C. Z.)

A little known species, apparently occurring in the same places with
the two preceding, as it is recorded from the Rio Madeira and Manaos
on the north bank.

111. Ortalis motmot motmot (Linnaeus)

Type locality: Cayenne

Obidos (Hellmayr, Pinto) ; Rio Maecuru, Monte Alegre (Snethlage)

5 cf 2 9 , north bank of Amazon, Lago Cuipeuz (M. C. Z.)
1 cf ad., Obidos (Carnegie Mus.)

112. Ortalis motmot ruficeps (Wagler)

Type locality: State of Para

Santarem (Chapman and Riker) ; Rio Tapajoz and Foz do Curua (Pinto)

1 cf , 3 9 , Rio Tapajoz, Pinhy

1 cf , 1 9 , Rio Tapajoz, Tauary

6 cf 1 9 , Santarem (Carnegie Mus.)

1 cf , Miritituba, Rio Tapajoz (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 121

This fine series amply validates Peter's reduction of ruficeps to a race
of motmot. A very interesting case, where the Amazon river is the
boundary between two distinct subspecies.

113. Ortalis spixi Hellmayr

Type locality: Maranhao, Brazil

Para, Rio Muria and Cajutuba (Xatterer); Castanhal (Stone, sight record);
Peixe-Boi and Ipitinga (Hellmayr) ; Rio Capim (Goeldi) ; Rio Tocantins,
Mazagao (Snethlage)

1 cf , Rio Capim, Resacca

1 9 , Para, Bosque

1 cf 1 9 , Rio Acara, Acara

3 cf 2 9 , Benevides (Carnegie Mus.)

While very distinct from motmot, it apparently represents it in
easternmost Amazonia. Southward this group of the genus is replaced
by 0. araucuan (Spix). Further west we find a different section of the
genus altogether, represented by 0. guttata (Spix).

114. Ortalis guttata guttata (Spix)

Type locality: Rio Solimoes

1 cf, Rio Tapajoz, Villa Braga (Carnegie Mus.)

1 cf , (do.), Itaituba (do.)

2 cf 1 9, (do.), Apacy (do.)

A range extension eastward from the Rio Madeira. These birds do
not differ from a series from the Rio Solimoes and the Rio Purus.

115. Pipile pipile cujubi (Pelzeln)

Type locality: Para

Para (Xatterer); Igarape-Assu (Hellmayr); Rio Capim (Goeldi); Rio Acara
(Snethlage); Santarem (Chapman and Riker); Monte Alegre (Snethlage);
Obidos (Pinto)

1 cf , Rio Tapajoz, Miritituba (Carnegie Mus.)
1 9 , (do.), Apacy (do.), perhaps this "species

Records from the north bank of the Amazon might be true pipile.

116. Pipile cumanensis subsp.
Santarem (Chapman and Riker) ; Rio Jamauchim, Santa Helena (Snethlage)

These records of this poorly known species were made long before
any subspecies were proposed, none of which are alleged to occur in our
area. The correctness of the identifications is questionable.

122 bulletin: museum of comparative zoology

Family PHASIANIDAE

117. Odontophorus gujanensis gujanensis (Gmelin)

Type locality : Cayenne

Para (Natterer) ; Rio Capim (Wallace and Goeldi) ; Igarape-Assu (Hellmayr) ;
Peixe-Boi and Ipitinga (Hellmayr); Santarem (Chapman and Riker);
Rio Tocantins, Mazagao; Cussary; Rio Tapajoz, Villa Braga and Boim;
Rio Jamauchim, Santa Helena; all Snethlage as marmoratus (Gould)
5 a71 3 9 , Rio Tapajoz, Tauary
2 9 , Obidos (Carnegie Mus.)

4 cf 4 9 , Santarem (do.)

1 cf 3 9 , Rio Tapajoz (do.)

This nice series is topotypical of rufinus (Spix), a name revived by
Chubb for a lower Amazon subspecies (cf. Ibis, 1919, pp. 25-29).
Peters assigned this form a questionable status in vol. 2 of his Check-
List, and suggested that the validity of the characters alleged required
confirmation. We can report our entire inability to confirm them, or in
fact to find any constant differences of any kind, and suggest that
Hellmayr in his study of the Spix types was entirely correct in regard-
ing rufinus (Spix) as a straight synonym of gujanensis.

Family OPISTHOCOMIDAE

118. Opisthocomus hoazin (P.L.S. Muller)

Type locality: Cayenne

Para (Wallace, Snethlage, Stone) ; Rio Inhangapy (Stone) ; Rio Capim (Goeldi) ;

Ilha das Oncas, San Antonio do Prata, Marajo Island (Snethlage, Pinto);

Santarem (Chapman and Riker)

2 cf , Para, Val-de-Caes

1 cf , Santarem (Carnegie Mus.)

Family ARAMIDAE

119. Aramus scolopaceus scolopaceus (Gmelin)

Type locality: Cayenne

Para (Hellmayr, Snethlage, and Stone); Mexiana Island (Hagmann); Marajo
Island (Snethlage)

1 cf 1 9 , Rio Tapajoz, Santarem
1 9 , Rio Tapajoz, Pinhy

1 cf , Rio Amazonas, Lago Jauary, Livramento
1 9 , Rio Amazonas, Lago Grande
1 cf , Rio Amazonas, Lago Cuipeuz
1 cf , Santarem (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 123

Family PSOPHIIDAE

120. Psophia crepitans crepitans (Linnaeus)

Type locality: Cayenne

1 ?, Lago Cuipeua, Obidos (Pinto)

A considerable southward extension of range, but to have been
expected.

121. Psophia viridis obscura Pelzeln

Type locality: Para

Para (Natterer and Wallace); Rio Capim (Goeldi); Rio Acara (Hellmayr and
Snethlage); Utinga (Conover)

122. Psophia viridis interjecta Griscom and Greenway, 1937

Type locality: Cameta, left bank, Rio Tocantins

1 d\ the type.

Combining the characters of obscura and viridis, but geographically
intermediate between obscura and dextralis. All four subspecies are
strikingly distinct. It will be of interest to see what Psophia turns up
on the Rio Xingti.

123. Psophia viridis dextralis Conover

Type locality: Tauary, right bank of Rio Tapajoz
Rio Tapajos, Tauary and Caxiricatuba (Conover)

2 cf 3 9 , Rio Tapajoz, Tauary and Pinhy

1 (J1, do. , Miritituba (Carnegie Mus.)

1 <f 1 9 , Santarem (do.)

Conover records a Psophia from the Rio Camaraipi as dextralis, but
it is obviously a connecting link between dextralis and obscura. This
river is between the Xingti and the Tocantins. If it is not the form
here described, it will probably prove to represent still another.

124. Psophia viridis viridis (Spix)

Type locality: Villa Nuova, Amazons (erroneously)
Left bank of the Rio Tapajoz at Boim (Conover)

1 cT 1 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

This form ranges west to the right bank of the Rio Madeira.

124 bulletin: museum of comparative zoology

Family RALLIDAE

125. Rallus longirostris subsp.

Marajo Island (Snethlage)

The Clapper Rail from the mouth of the Amazon still awaits sub-
specific determination. Typical longirostris is known from the Guianas;
crassirostris Lawrence from southern and central Brazil north to
Maranhao.

126. Pardirallus maculatus maculatus (Boddaert)

Type locality: Cayenne

3 9 in the zoological garden at Para, probably from the environs of the

capital (Snethlage); an old skin from the "Amazon River" (British

Museum); Para, (Stone)

127. Aramides cajanea cajanea (P.L.S. Muller)

Type locality: Cayenne

Rio Capim (Goeldi) ; Mexiana Island (Hagmann and Snethlage) ; Para (Gra-
ham in Brit. Mus., Snethlage and Stone); Marajo Island (Snethlage);
Santarem (Chapman and Riker)
1 9 , Rio Capim
1 cf, Rio Acara, Acara
4 9,1?, Rio Tapajoz, Pinhy and Tauary
2 d* 3 9 , Santarem (Carnegie Mus.)
1 9 , Obidos (do.)
1 9 , do., Goyana Isl. (do.)

We cannot see the slightest reason for recognizing grahami Chubb
from Lower Amazonia.

128. Amaurolimnas concolor castaneus (Pucheran)

Type locality: Brazil
Utinga and Santarem (Pinto)

1 cf, Rio Tapajoz, Boim

1 <?, Santarem (Carnegie Mus.)

This rare rail has hitherto been unknown between the Guianas and
southern Brazil. The male recorded above was so strikingly different
from the two currently recognized races that a third subspecies was
obviously involved. They are characterized as follows:

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 125

1. typical concolor (Gosse), 1847. Known only from Jamaica, and
apparently extinct. Of relatively large size, of paler coloration, the
olives and brown tones predominating over the rufescent. Specimens
examined, the type in Brit. Mus., 1 ad. Lawrence Coll. in New York,
2 ad. in Lafresnaye Coll., Mus. Comp. Zool., one badly faded.

2. guatimalcnsis (Lawrence). Very rare and local from Guatemala
through Central America to west Ecuador. Much smaller than typical
concolor, and strikingly darker, more olive less rufescent above, more
sooty brown below. Seven specimens including the type examined.
One from the Rio Solimoes (Carnegie Mus.) is transitional to the next.

3. castaneus (Pucheran). Definitely known only from Amazonian
Brazil southward to Bahia, Sao Paulo and Matto Grosso. As large as
concolor, but strongly rufous and chestnut, instead of olive and brown
with a rufescent tinge; strikingly distinct from concolor, and appearing
a different species when compared with guatimalcnsis. Only two
specimens examined.

A word about the nomenclature is in order. Rail us castancus
Cuvier is an MS name, based on a definite specimen from Brazil and
labelled castancus in Cuvier's handwriting. The name castancus
Cuvier has, of course, no nomenclatural status. It was taken up by
Lesson, but not properly validated by him, as there is no description.
It was, however, validated by Pucheran, (Rev. et Mag. Zool., 1851, p.
279) who gave a detailed description and critique of the "type", which
is clearly subspecifically identifiable today. While, therefore, castaneus
Pucheran (1851) is a synonym of the specific name concolor Gosse
(1847), it is available for the Brazilian subspecies. Rufirallus boecki
Bonaparte (1856) "Bolivia" given by Gray (1871), is clearly a synonym
of castaneus (Pucheran), but castaneus Bonaparte (1856) cannot be
definitely allocated, until Guiana specimens are identified sub-
specifically.
concolor— Wing: type 127, 2ad. 124, 1 ad. (9 ?) 119; tarsus. 43-46;

culmen 25-30.
guatimalensis—Wmg: & 118, 9 110-113; tarsus, c? 41, 9 35-39;

culmen, 25-27.
castaneus — 1 cf , wing 125, tarsus 46, culmen 26.

128. PORZANA FLAVIVENTER FLAVIVENTER (Boddaert)

Type locality : Cayenne

Rio Guamd, Ourem (Snethlage, 3d1 in Goeldi Museum)

126 bulletin: museum of comparative zoology

129. PORZANA ALBICOLLIS ALBICOLLIS (Vieillot)

Type locality: Paraguay

1 d\ north bank of Amazon near Obidos, Lago Cuipeuz

One of the many rare Rails, the distribution of which is poorly
known. This species is fairly well known in southern and eastern
Brazil only. The very distinct typhoeca Peters from Santa Marta
leaves much of eastern South America as a debatable ground. Our
specimen, while not typical albicollis, is assuredly nearer it than
to typhoeca.

130. Laterallus exilis exilis (Temminck)

Type locality: Cayenne

Para (Snethlage, Stone); Peixe-Boi (Hellmayr); Utinga (Pinto)
1 c? 3 9 , Rio Tapajoz, Tauary

131. Laterallus melanophaius lateralis (Lichtenstein)

Type locality: Bahia, Brazil

Para (Stone); Rio Guaraa (Layard and Snethlage); Igarape-Assu (Hellmayr)

132. Laterallus viridis viridis (P.L.S. Muller)

Type locality : Cayenne

Para (Natterer, Wallace, Snethlage and Stone) ; San Antonio and Igarape-Assu

(Hellmayr); Benevides, Rio Guama, Cussary (Snethlage); Rio Tapajoz,

Boim (Snethlage); Santarem (Chapman and Riker)

1 cf, Para, Bosque

4 c? , 1?, Rio Tapajoz, Tauary

2 9 , Obidos (Carnegie Mus.)
lc? 2 9, Benevides (do.)

2 c? 2 9 , Santarem (do.)

Birds from the south bank of the Amazon are inseparable from a
great Cayenne series in the Carnegie Museum.

133. Neocrex erythrops erythrops (Sclater)

Type locality: Lima, Peru

Rio Jamunda, Faro (Snethlage and Pinto)

This rare rail is best known from far upper Amazonia. It is doubt-
ful if any of these birds east of the Andes are typical erythrops. They
may prove to be olivascens Chubb, or a fourth subspecies.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 121

134. Gallinula chloropus galeata (Lichtenstein)

Type locality: Paraguay
Para (Snethlage and Pinto)

135. Porphyrula martinica (Linnaeus)

Type locality: Martinique

Para (Snethlage); Rio Tocantins (Pinto)

1 9 , Rio Capim, Ipanongo

2 d\ Rio Acara, Buenos Aires

1 cf 1 9 , Rio Tapajoz, Tauary

1 cf 1 9, Santarem (Carnegie Mus.)

136. Porphyrula parva (Boddaert)

Type locality: Cayenne

Para, Monte Alegre, Cussary (Snethlage); Para (Stone)

1 d\ Rio Amazonas near Obidos, Lago Cuipeuz

1 cf 1 9 , Santarem (Carnegie Mus.)

Family HELIORNITHIDAE

137. Heliornis fulica (Boddaert)

Type locality: Cayenne

Rio Capim (Goeldi); Para (also Stone), Cussary, Maraca, Monte Alegre
(Snethlage); Santarem (Chapman and Riker); Rio Tapajoz and Obidos
(Pinto)

6cf 3 9 , Rio Tapajoz, Pinhy
lcf, Rio Tapajoz, Santarem

Family EURYPYGIDAE

138. EURYPYGA HELIAS HELIAS (Pallas)

Type locality: Surinam

Cajutuba (Natterer); Rio Capim (Goeldi); Acara (Hellmayr); Mexiana Island
(Hagmann); Rio Guama, Marajo and Mexiana Islands, Monte Alegre
(Snethlage)

2 d" 1 9 , Rio Tapajoz, Caxiricatuba
1 cf, Rio Tapajoz, Santarem
1 9 , Rio Tapajoz, Pinhy
1 & 3 9 , Para, Val-de-Caes

128 bulletin: museum of comparative zoology

Family JACANIDAE

139. Jacana spinosa jacana (Linnaeus)

Type locality : Surinam

Para (Layard, Stone); Rio Capim (Goeldi); Mexiana Island (Hagmann);
Salvaterra, Marajo Island, Cussary (Snethlage); Santarem (Chapman and
Riker); Caviana Island (Brodkorb)
1 cf 4 9 , Rio Tapajoz, Pinhy
3 d" , 1 ?, Rio Tapajoz, Caxiricatuba
1 9 , Obidos (Carnegie Mus.)

1 9 , Santarem (do.)

2 d1 ad., 1 c? juv. 1 9 , Rio Tapajoz, Apacy (do.)

Family HAEMATOPODIDAE

140. Haematopus ostralegus palliatus (Temminck)

Type locality: Venezuela

Para (Snethlage); Cajutuba (Natterer)

The Oystercatcher ranges south to southeastern Brazil (Cape Trio,
Goeldi; Santa Catharina, Rogers).

Family CHARADRIIDAE

141. Belonopterus chilensis cayennensis (Gmelin)

Type locality: Cayenne

Para (Layard); Mexiana Island (Hagmann, Wallace, Snethlage); Marajo
Island (Snethlage); Santarem (Chapman and Riker); Caviana Island
(Brodkorb)

1 d\ Rio Tapajoz, Santarem

1 d\ Rio Amazonas, Lago Grande

1 d" 1 9 , Santarem (Carnegie Mus.)

It is, of course, possible that some of the records below might prove
referable to lampronotus (Wagler). (cf. Hellmayr, 1909, p. 491).

142. Hoploxypterus cayanus (Latham)

Type locality: Cayenne

Rio Capim (Goeldi); Mexiana Island (Hagmann); Rio Guama, Rio Tocantins,
Rio Tapajoz (Snethlage); Santarem (Hellmayr)

1 9 , Rio Tapajoz, Pinhel

7 d\ Rio Tapajoz, left bank (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 129

143. Pluvialis dominica dominica (P.L.S. Midler)

Type locality: Hispaniola
Marajo Island (Snethlage)

1 d", Rio Tapajoz, Pinhel, May 21, 1933

It is certainly interesting to find this bird in partial breeding plum-
age in eastern Brazil in late May.

144. Squatarola squatarola (Linnaeus)

Type locality: Sweden
Cajutuba (Natterer)

145. Charadrius semipalmatus Bonaparte

Type locality: New Jersey

Para (Stone); Cajutuba (Natterer); Mexiana Island (Wallace); Marajo Island
(Snethlage)

146. Charadrius collaris Vieillot

Type locality: Paraguay

Cajutuba (Natterer); Mexiana Island (Wallace, Hagmann and Snethlage);

Marajo Island (Snethlage); Quati-Puru; Rio Tapajoz, Goyana and Boim;

Rio Jamunda, Faro (all Snethlage) ; Santarem (Chapman and Riker)
4 cf 1 9 , Rio Tapajoz, Pinhel and Boim

2 o" 1 9 , Santarem (Carnegie Mus.)

147. Charadrius wilsonia wilsonia Ord

Type locality : Cape May, New Jersey
Catjutuba and Rio Muria (Natterer)

The occurrence of Wilson's Plover in Brazil seems to have been
overlooked in recent years by everyone except Hellmayr. It has been
taken as far south as Bahia (Dr. Wucherer in Brit. Mus.). For the
reasons for regarding these birds as North American migrants, cf.
Hellmayr, 1929, p. 492.

Family SCOLOPACIDAE
148. Bartramia longicauda (Bechstein)

Type locality: North America

1 cf, Rio Tocantins, Baiao (Snethlage)

1 9 , Santarem, Sept. 15, 1920 (Carnegie Mus.)

130 bulletin: museum of comparative zoology

149. Numenius phaeopus hudsonicus (Latham)

Type locality: Hudson Bay

Cajutuba (Natterer) ; Para (R. Graham) ; Marajo Island (Snethlage)

150. Tringa flavipes (Gmelin)

Type locality: New York

Cajutuba (Natterer); Mexiana Island (Wallace and Hagmann); Marajo Island
(Snethlage)

4 9 1?, Rio Tapajoz, Santarem, Oct. 4-8, 1932
2 cM 9 , Santarem, Sept. 16, 1919 (Carnegie Mus.)

151. Tringa melanoleuca (Gmelin)

Type locality: Labrador

Cajutuba (Natterer) ; Marajo Island (Snethlage) ; Para (Graham and Stone)

152. Tringa solitaria solitaria Wilson

Type locality: Pennsylvania

Para (Graham, Layard, Stone); Peixe-Boi (Hellmayr); Mexiana Island
(Wallace); Capanema, Braganca, Rio Guama, Marajo Island, Rio Tapajoz
(Snethlage)

1 9 , Rio Tapajoz, Tauary, Dec. 1, 1933

1 9 , Rio Tapajoz, Pinhel, May 18, 1933

1 9 , Amazon River near Obidos, March 2, 1933

4 d> 1 9 , Santarem, March 26, April 12-17, Aug. 2, 1919 (Carnegie

Museum)
1 9 , Obidos, May 2, 1921 (Carnegie Mus.)
1 cf, Rio Tapajoz, Feb. 24, 1921

153. Actitis macularia (Linnaeus)

Type locality: Pennsylvania

Cajutuba (Natterer); Mexiana Island (Wallace); Para, (Graham and Stone);
Marajo Island, Cunany (Snethlage) ; Santarem (Chapman and Riker)

1 cf, Rio Tapajoz, Pinhy, May 20, 1933

1 9, Santarem, Oct. 14, 1919 (Carnegie Mus.)

154. Catoptrophorus semipalmatus semipalmatus (Gmelin)

Type locality: New York

"Single birds, rare, 2 specimens, March, 1835", Cajutuba (Natterer)

This Brazilian record for the Willet has been completely overlooked
by recent American authors, and also by Snethlage.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 131

155. Arenaria interpres morinella (Linnaeus)

Type locality: Georgia

Para" and Cajutuba (Natterer, February, March, and April, 1835)

156. Limnodromus griseus griseus (Gmelin)

Type locality: New York

Cajutuba, April, 1835, 5 specimens, Cajutuba (Natterer); Marajo Island
(Snethlage)

157. Capella paraguaiae paraguaiae (Vieillot)

Type locality: Paraguay

Para" (Stone, Graham); Braganca, Marajo Island, Monte Alegre (Snethlage);
Santarem (Chapman and Riker); Caviana Island (Brodkorb)

1 d1, Obidos (Carnegie Mus.)

1 9 , Santarem (do.)

158. Crocethia alba (Pallas)

Type locality: North Sea

Cajutuba, November, 1835 (Natterer)

159. Ereunetes pusillus (Linnaeus)

Type locality: Santo Domingo

Cajutuba, March and April, 1835 (Natterer); Mexiana Island (Wallace);
Marajo Island (Snethlage)

1 d\ Rio Tapajoz, Santarem; Nov. 18, 1932

160. Erolia minuta (Vieillot)

Type locality: Nova Scotia

Para (Layard) ; Mexiana Island (Wallace, Hagmann) ; Marajo Island (Sneth-
lage)

161. Erolia fuscicollis (Vieillot)

Type locality: Paraguay

Rio Tocantins (Wallace); Braganca (Snethlage)

1 tf1 5 9 , Rio Tapajoz, Pinhel, May 16-19, 1933

132 bulletin: museum of comparative zoology

162. Erolia melanotos (Vieillot)

Type locality : Paraguay

1 d* 2 9 , Rio Tapajoz, Santarem, Oct. 4, 1932

1 d\ Rio Amazonas, Lago Grande, Sept. 6, 1932

These are the only records for Lower Amazonia. The species is well
known, however, just west and south of our area in Upper Amazonia
and Matto Grosso.

Family RECUR VIROSTRIDAE

163. Himantopus himantopus mexicanus (P.L.S. Miiller)

Type locality: Mexico

Cajutuba (Natterer); Mexiana Island (Wallace, Hagmann, Snethlage); Marajo

Island (Snethlage) ; Monte Alegre (Snethlage) ; Santarem (Chapman and

Riker); Caviana Island (Brodkorb)

Natterer's specimen was reported by Pelzeln as melamirus Vieillot.
There is no satisfactory evidence as yet that any Stilt breeds in
northern Brazil, or that melamirus ranges anywhere nearly so far
north as the Amazon Valley.

Burhinus bistriatus vocifer (L'Herminier) is recorded by Snethlage
from "Para (zoological garden)". It is a bird of the savannah regions,
Colombia, Venezuela and Guiana, and the nearest definite record to
our area is the upper Rio Branco in northwestern Brazil. As a definite
locality, Para in this sense cannot be taken literally.

Family LARIDAE
164. Larus atricilla Linnaeus

Type locality: Bahamas

Cajutuba, Feb. 20, 1835 (Natterer); Marajo Island (Snethlage)

165. Phaetusa simplex simplex (Gmelin)

Type locality : Cayenne

Para (Layard, Stone); Cajutuba (Natterer); Mexiana Island (Wallace); Quati-
puru, Marajo Island, Monte Alegre (Snethlage) ; Santarem (Chapman and
Riker)

1 d\ 1 9 , Rio Tapajoz, Tauary and Caxiricatuba
2 9 , Santarem (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 133

166. Gelochelidon nilotica gronvoldi Mathews

Type locality: South America
Marajo Island (Snethlage)

The reported breeding of this species on Mexiana Island goes back
to Hagmann. Hellmayr, however, has shown that his identification
was erroneous (cf. Hellmayr, 1910, p. 122, footnote 6).

167. Sterna hirundo Linnaeus

Type locality: Sweden

Para, March 3, 1936; Marajo Island, Jan. 3, 1936 (cf. Lincoln, Bird-Banding,
Oct., 1936, pp. 146 and 147)

168. Sterna superciliaris Vieillot

Type locality: Paraguay

Cajutuba (Natterer); Peixe-Boi (Snethlage); Rio Tocantins (Wallace); Para
and Rio Tapajoz (Pinto)

3 o"7 9, Rio Tapajoz, Santarem and Pinhel

1 9 , Obidos (Carnegie Mus.)
1 o71 1 9 , Santarem (do.)

169. Sterna fuscata fuscata Linnaeus

Type locality: Santo Domingo

Mouth of the Amazon (Saunders Coll., in Brit. Mus.)

170. Sterna albifrons antillarum (Lesson)

Type locality: Guadeloupe
Marajo Island (Snethlage)

171. Thalasseus maximus maximus (Boddaert)

Type locality: Cayenne
Para, 1 <? (Snethlage)

172. Thalasseus sandvicensis acuflavidus (Cabot)

Type locality: Yucatan

Cajutuba (Natterer); Mexiana Island, breeding (Hagmann)

134 bulletin: museum of comparative zoology

173. Anous stolidus stolidus (Linnaeus)

Type locality: West Indies

"At sea off north Brazil, Oct. (Brit. Mus.)

Family RYNCHOPIDAE

174. Rynchops nigra cinerascens Spix

Type locality : Amazon River

Para (Stone); Cajutuba (Natterer); Mexiana Island (Wallace); Marajo
Island (Snethlage)

1 ?, Iiio Tapajoz, Villa Braga (Carnegie Mus.)

Family COLUMBIDAE

175. COLUMBA SPECIOSA GMELIN

Type locality: Cayenne

Para (Natterer and Wallace); Rio Muraiteua (Stone); Ipitinga and San
Antonio (Hellmayr); Rio Capim (Goeldi); Monte Alegre, Cussary, Rio
Tapajoz, Rio Jamunda (Snethlage); Santarem (Chapman and Riker,
Pinto); Marajo Island (Brodkorb)
1 cf, Rio Tapajoz, Tauary
1 9 , Santarem (Carnegie Mus.)
1 d\ Rio Tapajoz, Villa Braga (do.)

176. Columba rufina rufina Temminck

Type locality : Cayenne

Rio Jamunda, Faro (Snethlage); Caviana Island (Brodkorb)
1 d71, 1 9 , Rio Amazonas, Lago Cuipeuz

177. Columba rufina sylvestris Vieillot

Type locality : Paraguay

Para (Stone); Mexiana Island (Wallace & Snethlage) Benevides, Marajo
Island, Rio Tocantins, Rio Tapajoz (Snethlage); Santarem (Chapman and
Riker)

13 cf 8 9 , Rio Tapajoz; various localities, east bank
2 cf 1 9 , Rio Tapajoz, west bank. (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 135

We entirely agree with Hellmayr that the Amazon River is the
boundary between typical rufina of the Guianas, and syhestris of
northern Argentina northward (cf. Birds of Northeast Brazil, p. 464).
No birds from the Amazonian basin are really typical of either, but are
variously intermediate. The long series from the Rio Tapajoz has an
obviously bicolored tail, but the majority not as sharply contrasted
as Argentina and Sao Paulo specimens. Similarly, the two birds from
the north bank of the Amazon are much nearer true rufina.

ITS. COLUMBA PLUMBEA WALLACEI Chubb

Type locality: Rio Capim, Para, Brazil

Rio Capim (Wallace, Goeldi) ; Para (Stone) ; San Antonio and Ipitinga (Hell-
mayr); Rio Jamauchim, Santa Elena (Snethlage); Santarem (Chapman
and Riker).

1 9 imm., Obidos (Carnegie Mus.)

2 d" , 2 9 , Rio Tapajoz. Villa Braga (do.)

Todd, 1937, has straightened out the great local confusion in these
pigeons. Having examined the same material, we fully endorse his
main conclusions. The series from the Rio Purus, representing pal-
lescens Snethlage, proves the lower Amazon birds to be another sub-
species, as Hellmayr long ago suggested. There is, however, the remote
possibility that locutrix Wied, based on Bahia birds, might prove to be-
long to the present form. In far upper Amazonia, Zimmer (Birds
Peruvian Exped., 1930, p. 256) has resurrected the name delicata
Berlepsch and Stolzmann for the birds from northern Bolivia to Peru,
Ecuador and Colombia, of which propinaua Cory and andicola Chubb
are synonyms. Based on meager material this name would appear
to contain two elements, a larger, darker bird from Bolivia, and a
smaller, paler one northward, which has yet to be compared with
authentic pallescens.

179. COLUMBA SUBVINACEA RECONDITA Todd, 1937

Type locality: Colonia do Mujuy, Santarem, Brazil
Rio Gurupy, Rio Tocantins (Snethlage)

1 cf 1 9 , Obidos (Carnegie Mus.)
1 cf, Santarem (do.)

1 o71 1 9 , Rio Tapajoz, Villa Braga (do.)

This bird is the Columba purpureotincta of Snethlage and Pinto, ncc
Ridgway, a little known subspecies of British Guiana and adjacent

136 bulletin: museum of comparative zoology

Venezuela, which closely resembles bogotensis in color, but differs only

in its dwarf dimensions. Wing measurements are appended.

bogotensis— 4 d* 163-173; 5 9 165-175

recondita— Rio Purus, 3d" 164-170; 3 9 153-156

recondita — Obidos, 1 cf 151 ; 1 9 154 (both immature)

recondita — Santarem, 1 c? 160

recondita — Rio Tapajoz, 1 d" 163; 1 9 155 (immature)

recondita — Rio Tocantins 2 c? 155; (fide Snethlage)

jmrpureotincta 4 ? 146.5-149.5 (fide Ridgway)

In both species, plumbea and subvinacea, immature specimens are
smaller than adults. They are recognizable in having some buffy edg-
ings to the feathers of the mantle and nape, and rusty edgings to the
under tailcoverts.

ISO. Zenaida auriculata marajoensis Berlepsch

Type locality: Marajo Island

San Juao, Pard, (Layard); Mexiana Island (Wallace, Hagmann and Snethlage);
Marajo Island (Berlepsch and Snethlage)

181. Zenaida auriculata jessie.e Ridgway

Type locality: Diamantina, Rio Tapajoz

Diamantina (Ridgway); Santarem (Chapman and Riker); Erere (Snethlage);
Paricatuba (Schulz in Frankfort Museum); Rio Tapajoz (Pinto).
2 cf 2 9 , Santarem (Carnegie Mus.)

For the latest opinion on the races of this species see Naumburg,
Amer. Mus. Novit., no. 648, 1933.

182. COLUMBIGALLINA PASSERINA GRISEOLA (Spix)

Type locality: Amazon River

Para (Layard, Natterer Stone); Bemfica (Steere); Rio Capim (Goeldi); San
Antonio (Hellmayr); Santarem (Chapman and Riker); Para, Quatipuru,
Maraca, Monte Alegre, Rio Xingii, Victoria (Snethlage)
8 d" , 16 9 Rio Tapajoz, various localities
4 d", Rio Amazonas, Lago Cuipeuz
2 9 , Rio Amazonas, Boca do Igarape Piaba
3 cM 9 Obidos (Carnegie Mus.)

3 d" 1 9 Benevides (do.)

1 d" Santarem (do.)

4 cf 1 9 Rio Tapajoz, both banks (do.)

GRISCOM AND GREEN WAY: BIRDS OF LOWER AMAZONIA 137

183. Columbigallina talpacoti talpacoti (Temminck)

Tj'pe locality: Brazil

Para (Natterer); Castanhal (Stone); Nazare (Layard); Peixe-Boi (Hellmayr);
Mexiana Island (Hagmann); Para (Snethlage); Arumanduba, Ererc, Rio
Maecuru (Snethlage); Rio Tocantins, Alcobaca (Snethlage); Rio Tapajoz,
Goyana (Snethlage) and Santarem (Chapman and Riker); Marajo Island
(Brodkorb)

13 cf, 8 9 Rio Tapajoz, various localities
2o", 3 9, Rio Acara, Acara

1 cf Obidos (Carnegie Mus.)

1 a* Santarem (do.)

8 cT 6 9 , Rio Tapajoz, both banks (do.)

With the greatly increased material in this museum both from
Surinam and southern Brazil since Bangs and Penard described C.
arthuri, we agree absolutely with Hellmayr (Birds Northeast Brazil,
p. 468) that matters of individual variation are involved. The present
fine series shows this quite graphically. Several birds resemble one
from Sao Paulo in having more rufous on the primaries than the type
of arthuri; perhaps half have some rufous, and the remainder are solid
black. It is consequently quite impossible to divide this series into two
artificial "species". Twenty-one specimens from the Guianas divide
in a similar fashion, so that it seems impossible to recognize arthuri
even as an intermediate race connecting the Central American rufi-
pemiis, a mere geographical representative, with talpacoti.

184. Uropelia campestris (Spix)

Type locality : Bahia
Marajo Island (Snethlage)

One of the many characteristic campo birds of central and southern
Brazil, which ranges north to this island, but is absent from the
forested sections of Lower Amazonia.

185. Clara vis pretiosa (Ferrari-Perez)

Type locality: Brazil
Rio Maecuru (Snethlage)

1 d\ Santarem (Carnegie Mus.)

1 o\ Rio Tapajoz, Villa Braga (do.)

138 bulletin: museum of comparative zoology

186. Leptoptila verreauxi approximans Cory

Type locality : Serra de Baturite, Ceara, Brazil

Monte Alegre and Rio Jamunda (Faro); Mexiana and Marajo Islands (all
Snethlage); Rio Tapajoz and Santarem (Pinto)
3 d71 3 9 , Rio Tapajoz, various localities
5 cf 2 9 , Rio Amazonas, Lago Cuipeuz
1 d", Benevides (Carnegie Mus.)
1 d\ Santarem (do.)

The characters of this race have been fully discussed by Hellmayr
(Birds Northeast Brazil, p. 471), who also shows that the Guiana
race must be known as brasiliensis Bonaparte. Our series from the
north bank of the Amazon is inseparable from the Tapajoz series,
showing that brasiliensis (=tenella Penard, type before us) does not
extend so far south as the Amazon, in this section at least.

187. Leptoptila rufaxilla rufaxilla (Richard and Bernard)

Type locality: Cayenne

Obidos and Rio Jamunda, Faro (Snethlage); Mexiana Island (Wallace, Hag-
mann, Snethlage); Caviana Island (Brodkorb); Para (Stone); Rio Muria
(Natterer); Rio Tapajoz, Mararu, Goyana (Snethlage); Santarem (Chap-
man and Riker)

2 9 Obidos (Carnegie Mus.)

1 ? Santarem (M.C.Z.), 1 <? do. (Carnegie Mus.)

3 a1 ad., 1 a71 imm., 2 9 ad, 1 9 imm., Rio Tapajoz, right bank
(M.C.Z.)

2 cf 2 9 , Rio Tapajoz, both banks (Carnegie Mus.)

The fine series from Cayenne in the Carnegie Museum proves that
lower Amazonian birds are inseparable, and that Dutch and British
Guiana specimens, long assumed to represent true rufaxilla, are in
reality a different subspecies, hypochroos Griscom and Greenway.

188. Oreopeleia violacea violacea (Temminck and Knip)

Type locality: South America
San Antonio (Hellmayr)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 139

189. Oreopeleia Montana (Linnaeus)

Type locality: Jamaica

Para (Wallace Stone); Rio Capim (Goeldi); Mocajatuba, Ananindeua, Santa
Isabel, Benevides, Peixe-Boi (Snethlage); Rio Tocantins (Cameta), Rio
Curua, and Rio Tapajoz, Boim (Snethlage); Obidos (Snethlage); Obidos
and Santarem (Pinto)

2 cf 2 9 , Rio Tapajoz, Tauary
1 J1, Obidos (Carnegis Mus.)
1 cf, Santarem (do.)

Family PSITTACIDAE

190. Anodorhynchus hyacinthinus hyacinthinus (Latham)

Type locality: none given by Latham; "Brazil" of later authors; we designate

Rio Tapajoz, Tauary
Rio Capim (Goeldi); Monte Alegre (Snethlage); Santarem (Chapman and

Riker); Rio Tapajoz (Bates in Brit. Mus.)
1 c? 1 9 , Rio Tapajoz, Tauary

The lovely Hyacinthine Macaw has largely been overlooked in
recent decades in the lower Amazon valley and has become associated
in people's minds with central and southwestern Brazil. The receipt of
specimens from the lower Amazon shows that the Matto Grosso bird
is quite distinct. Latham's original description was based on a bird in
a private museum with no locality. The chances, however, are that a
specimen would have reached Europe prior to 1790 from Para rather
than the interior of southern Brazil, and we consequently restrict the
type locality on this basis. .Psittacus augustus Shaw, 1792, no definite
locality, is based on a live bird belonging to Lord Orford. By the same
reasoning, we restrict this name to "vicinity of Para, in lower Ama-
zonia." The name of the Matto Grosso race will consequently be
Anodorhynchus hyacinthinus maximiliani Spix, type collected Oct. 20,
1827, at Rio das Flechas, Matto Grosso. This subspecies is the one
generally represented in most collections. The typical form is ob-
viously a brighter blue both above and below, more ultramarine, less
purplish, most conspicuous on the underparts; lower mandible 5-8 mm.
longer, and width of bill at gape 5 mm. + narrower, the whole bill
consequently relatively longer and slenderer. We are indebted to the
American Museum in New York for the opportunity of comparing our
Tapajoz skins with their fine series from Matto Grosso (5 cf 4 9 )•

140 bulletin: museum of comparative zoology

191. Ara ararauna (Linnaeus)

Type locality: Brazil

Mexiana Island (Wallace) ; Santarem (Allen)
1 d" , Santarem (Carnegie Mus.)

192. Ara macao (Linnaeus)

Type locality: Pernambuco

Para (Natterer); Rio Capim (Goeldi); Mexiana Island (Wallace and Hag-
mann); Rio Guama, Ourem; Rio Tocantins, Aramatheua; Rio Jamunda,
Faro (the last three, Snethlage) ; Rio Tocantins (Pinto)

193. Ara chloroptera Gray

Type locality: Guiana

Para (Natterer); Rio Capim (Goeldi); Santarem (Chapman and Riker); Rio
Maraca, Rio Tocantins (Aramatheua), Rio Jamauchim (Snethlage).
1 o71 1 9 , Rio Tapajoz, Tauary and Caxiricatuba

194. Ara severa (Linnaeus)

Type locality: Amazon River

Mexiana Island (Hagmann, Snethlage); Santarem (Pinto)

3 cf , Rio Tapajoz, Santarem

1 d1, Santarem (Carnegie Mus.)

195. Ara maracana (Vieillot)

Type locality: Paraguay

Cajutuba (Natterer); Marajo Island (Snethlage); Santarem (Chapman and
Riker)

196. Ara manilata (Boddaert)

Type locality : Cayenne

Monte Alegre, Cussary, Marajo Island (Snethlage); Pataua (Pinto)
1 cf , Benevides (Carnegie Mus.)
1 cf 1 9 , Santarem (do.)

197. Ara nobilis cumanensis (Lichtenstein)

Type locality: Brazil-Cuman, Maranhao
Para (Wallace); Cajutuba (Natterer)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 141

This small Macaw seems to have a range, which is practically
identical with that of Anodorhynchus. Further north, it is replaced by
typical nobilis, formerly better known as hahni Souance, which is not
known nearer our region than the Rio Branco.

198. Aratinga guarouba (Gmelin)

Type locality; northeastern Brazil

Para (Natterer, Wallace Stone); Peixe-Boi (Hellmayr and Snethlage); San

Antonio do Prato, Rio Tocantins, Rio Xingu (Snethlage); Rio Tocantins

(Pinto)

Outside of our area this conure is known only from the adjacent
state of Maranhao.

199. Aratinga solstttialis (Linnaeus)

Type locality: Cayenne

Monte Alegre and Erere (Snethlage); Santarem (Pinto)

This species replaces the last from the north bank of the Amazon
north to the Guianas.

200. Aratinga leucophthalmus leucophthalmus (P.L.S. Midler)

Type locality: Guiana

Para and Cajutuba (Natterer); Mexiana Island (Hagmann); Marajo Island,

Rio Tocantins, Rio Jamunda (Snethlage); Santarem (Chapman, Riker

and Pinto)

4 o71 4 9 , Rio Tapajoz, Santarem
1 a" 1 9 , Obidos (Carnegie Mus.)

2 d\ Santarem (do.)

A common and widely distributed parrakeet, represented by the
race callogenys Salvadori in extreme upper Amazonia.

201. Aratinga aurea aurea (Gmelin)

Type locality: Brazil

Para (Graham); Mexiana Island (Wallace and Hagmann); Caviana Island
(Brodkorb); Marajo Island (Hellmayr and Snethlage); Monte Alegre,
Erere, Igarape de Paituna, Rio Jamunda (Snethlage); Santarem (Chap-
man, Riker and Pinto)

5 c? 2 9,1?, Rio Tapajoz, Santarem
8 d" 3 9 , Santarem (Carnegie Mus.)

142 bulletin: museum of comparative zoology

Another common parrakeet in eastern South America, passing in
the extreme south into major Cherrie and Reichenberger. Unlike
leucophthalmus, however, it is unrecorded west of the Rio Madeira
basin.

202. Pyrrhura picta amazonum Hellmayr

Type locality: Obidos

Rio Tocantins, Cussary, Monte Alegre, Rio Tapajos (Snethlage); Obidos
(Hellmayr and Snethlage); Santarem (Chapman and Riker, Hellmayr);
Obidos and Santarem (Pinto)

3 cf , 1 9, Rio Tapajoz, Caxiricatuba
1 cf, Rio Tapajoz, Caxiricatuba
10 cf , 6 9, Obidos (Carnegie Mus.)
7 cf , 6 9 , Santarem (do.)

This variable species is represented by true picta in the Guianas and
Venezuela and by lucianii Deville in far upper Amazonia. According
to Snethlage, birds from the Rio Madeira are another undescribed
race, but Hellmayr doubts this.

[Pyrrhura melanura (Spix) is a well known species of upper Ama-
zonia, which is not definitely known east of the Rio Solimoes and the
Rio Negro. There is a specimen in the old British Museum collection
by Bates from "Tocantins", Amazon River. This "Tocantins" (really
Tonantins) must not be confused with the Rio Tocantins in our area. ]

203. Pyrrhura perlata lepida (Wagler)

Type locality: Amazon River

Para (Natterer); San Antonio (Hellmayr); Rio Capim (Wallace); Peixe-Boi

(Hellmayr); Benevides (Snethlage); Igarape-Assu (Hellmayr); Utinga

(Pinto)

204. Pyrrhura perlata anerythra (Neumann)

Type locality: Rio Tocantins, Arumatheua

from type station, 2 cf (Neumann); Cametd, 1 cf (Sieber in Berlin Museum)

Only known from the specimens listed above. Prof. Neumann's
revision (Verh. Orn. Ges. Bayern, 17, no. 4, 1927) shows that typical
perlata (Spix) is only known from the two cage-bird types, which may
be a mere cage variety, as suggested by Hellmayr.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 143

205. Pyrrhura rhodogaster Sclater

Type locality: Borba, Rio Madeira

1 cf , Rio Jamauchim (Snethlage); 1 9, Rio Arapiuns (Pinto)

1 cf, Rio Tapajoz, Boim

2 cf 1 9 , Rio Tapajoz, Apagy (Carnegie Mus.)

This rare parrot is only known from the localities mentioned above.

206. Forpus modestus modestus Cabanis

Type locality: British Guiana

1 9 irnm., Rio Tapajoz, Aveiros (Carnegie Mus.)

This bird agrees with a series from the Rio Purus of this very dark
species with the blackish upper mandible. We have, however, no
typical material.

207. Forpus passerinus subsp.

Para, Rio Jamauchim, Recreio and Porto Seguro (Snethlage)

There is still some uncertainty about Forpus in Lower Amazonia.
Snethlage is the only person who has seen a series from Para. These
birds she distinguishes from deliciosus Ridgway (Santarem), in that
adult males have bright ultramarine blue rumps and a lot of yellow
on the forehead and sides of the head. She calls these birds modestus
Cabanis, but it would appear certain that she did not understand that
species. Her description leads one to infer that her birds will turn out
to be fiavissimus Hellmayr (Maranhao). Stone records a single female
from Para as modestus also.

208. Forpus passerinus deliciosus (Ridgway)

Type locality: Diamantina Creek, near Santarem

Santarem (Chapman and Riker, Snethlage); Monte Alegre, Igarape" de
Paituna, Rio Jamunda (Snethlage); Obidos (Hellmayr and Pinto)

5 cf 3 9, north bank of Amazon near Obidos and south bank, Lago
Grande

1 9 , Rio Tapajoz, Tauary

2 cf , Obidos (Carnegie Mus.)
12 cf 13 9 , Santarem (do.)

144 bulletin: museum of comparative zoology

209. Brotogeris versicolurus versicolurus (P.L.S. Miiller)

Type locality: Cayenne

Para (Natterer Stone); Benevides (Steere); Rio Inhangapy, Rio Guama

(Stone); Mexiana Island (Wallace, Hagmann, Hellmayr); Marajo Island

(Hellmayr, Snethlage); Ilha das Oncas, Serra de Paituna, Rio Jamunda,

Rio Tocantins (Snethlage); Santarem (Chapman and Riker, Hellmayr)

3 9 , Para, Val-de-Caes

1 9 , Rio Tapajoz, Pinhel

2 9 , Lago Grande

2 d71 1 9 , Obidos (Carnegie Mus.)
8 c? 5 9 , Santarem (do.)

1 cf 1 9 , Rio Tapajoz (do.)

210. Brotogeris chrysopterus tuipara (Gmelin)

Type locality: northeastern Brazil

Para (Layard, Natterer, Wallace Stone); Ourem (Stone); Igarape-Assu (Hell-
mayr); Rio Capim (Goeldi); Peixe-Boi and Ipitinga (Hellmayr); Ilha das
Oncas, Rio Barcarena, Marapanim, San Antonio do Prata, Providencia,
Rio Tocantins (all Snethlage); Santarem (Chapman and Riker); Marajo
Island (Brodkorb)

5 cf 5 9 , Para, Val-de-Caes and Bosque

7 d1 4 9 , Rio Tapajoz, Tauary and Caxiricatuba

7 d1 2 9 , Santarem (Carnegie Mus.)

1 d1 1 9 , Benevides (do.)

211. Brotogeris chrysopterus chrysopterus (Linnaeus)

Type locality: Guiana

Rio Jary, San Antonio da Cachoeira, and Monte Alegre (Snethlage); Obidos
(Pinto)

3 cT 1 9 , Obidos (Carnegie Mus.)

These two subspecies represent each other on the south and north
banks of the Amazon. They are replaced by chrysonema on the Rio
Madeira.

212. Brotogerys st. thomae takatsukasae (Neumann)

Type locality: north bank of the Amazon opposite Santarem
Monte Alegre (Snethlage, Neumann); Maraca (Snethlage)
6 d1 4 9, north bank of Amazon near Obidos

1 9 , Rio Tapajoz, Santarem
2d" 19, Obidos (Carnegie Mus.)
13 d" 12 9, Santarem (do.)

GRISCOM AND GREENWAY : BIRDS OF LOWER AMAZONIA 145

Typical st. thomac ranges east to the Rio Madeira. We confirm
Neumann's color characters. Specimens seen by us from eastern
Ecuador in this museum and New York are radically larger than our
lower Amazonian series.

213. Amazona farinosa farinosa (Boddaert)

Type locality: Cayenne

Pard (Natterer); Castanhal (Stone); Peixe-Boi (Hellmayr and Snethlage);
Rio Capim (Goeldi); Mexiana Island (Hagmann); Rio Jamauchim
(Snethlage) ; Obidos and Para (Pinto)
8^29, Rio Tapajoz, Tauary

2 9 , north bank of Amazon near Obidos
1 d\ Obidos (Carnegie Mus.)
1 d" 1 9 , Santarem (do.)
1 d\ Rio Tapajoz (do.)

214. Amazona amazonica amazonica (Linnaeus)

Type locality: Surinam in error — "les pays des Amazones"
Cajutuba (Natterer); Rio Capim (Goeldi); Mexiana Island (Hagmann); Ilha
das Oncas, Marajo Island, Amapa, Rio Jamauchim (Snethlage)

1 d\ Para, Val-de-Caes

1 9 , imm., Rio Tapajoz, Villa Braga (Carnegie Mus.)

A very widely ranging species represented by micra Griscom and
Greenway on the Surinam coast, which may prove to have a wider
distribution.

[Amazona aestiva aestiva (Linnaeus) is listed in Snethlage's Aves do
Amazonicas on the basis of specimens in the zoological garden at Para.
It is not known north of Pernambuco, and Para, cannot be regarded
as a definite locality for this species. |

215. Amazona ochrocephala subsp.

Santarem (Chapman and Riker) ; Caxiricatuba, Rio Tapajoz (Pinto)
1 9 , Obidos (Carnegie Mus.)
1 9 , Santarem (do.)

216. Amazona ochrocephala xantholaema Berlepsch

Type locality: Marajo Island

Marajo Island (Berlepsch and Snethlage)

The status of this species in Lower Amazonia still remains to be
determined. Snethlage was about to describe this race, as a new

146 bulletin: museum of comparative zoology

species, apparently referring all Brazilian records and those from
"neighboring countries to the north" to it. The species was collected
by Natterer on the Rio Branco ; apparently the Santarem record given
above is the only basis for the occurrence of the species between
northwestern Amazonia and Marajo Island. The relationship of
alleged ochrocephala from the south bank of the Amazon to nattereri
from the Rio Madeira should also be investigated.

217. Amazona festiva festiva (Linnaeus)

Type locality: Guiana

Para (Graham in Brit. Mus.); Mexiana Island (Hagmann and Snethlage);
Monte Alegre (Snethlage); Santarem (Chapman and Riker); Lago
Pataua and Lago Cuipeua (Pinto)

3 d" 5 9 , north bank of Amazon near Obidos

1 ?, Rio Tapajoz, Santarem.
1 d" 1 9 , Obidos (Carnegie Mus.)
1 d" 1 9 , Santarem (do.)

One of these birds is chloronota Souance in having a green rump and
red at the bases of the outer tail feathers ; another has the red in the
tail, but a crimson rump; two others have a few red feathers in an
otherwise green rump. It will be apparent, therefore, that we have
here either an age or a mutational variation so common in these
parrots; not only is chloronota not a distinct species, but it seems to us
to have no status as a subspecies either (see Cory, Cat Birds Amer.,
pt. 1, 1918, p. 88).

218. Graydidasculus brachyurus insulsus
(Griscom & Green way, 1937)

Type locality: Lago Grande, south bank of Amazon

Para (Hellmayr) ; Amapa, Monte Alegre, Rio Jamunda (Snethlage) ; Santarem
(Chapman & Riker) ; Pataua (Pinto)
1 c? , Rio Tapajoz, Santarem

1 d\ 2 9 1 ?, north bank of Amazon near Obidos
4 d\ 2 9 south bank, Rio Amazonas, Lago Grande
1 d1 4 9 , Obidos (Carnegie Mus.)
11 cf 6 9, Santarem (do.)

Typical brachyurus is a much larger bird of Upper Amazonia, with a
proportionately smaller and weaker bill.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 147

219. Pionus menstruus menstruus (Linnaeus)

Type locality: Surinam

Cajutuba (Natterer); Igarape-Assu (Robert); Ilha das Oncas, Benevides, Santa
Antonio do Prata, Peixe-Boi, Rio Acara, Rio Tocantins (Snethlage);
Santarem (Chapman and Riker); Maraca, Obidos, Rio Jamunda, Campos
de Ariramba (Snethlage)

5 cf, 3 9,1?, Rio Tapajoz, Pinhy, Tauary, Boim
1 c? 1 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)
1 <? 1 9 , Obidos (do.)

1 cf 3 9 , Santarem (do.)

220. Pionus ftjscus (P.L.S. Miiller)

Type locality: Cayenne

Para (Natterer and Wallace) ; Rio Capim (Goeldi) ; Igarape-Assu and Ipitinga
(Hellmayr); Mocajatuba, Providencia, Benevides, Peixe-Boi, Rio Acara,
Rio Tocantins (Snethlage); Santarem (Chapman and Riker)

2 cf 2 9 , Rio Tapajoz, Tauary

1 9 , north bank of Amazon near Obidos

2 rf, Obidos (do.)

1 cf 2 9 , Santarem (do.)

2 cf 3 9 , Rio Tapajoz, various localities (Carnegie Mus.)

221. Deroptytjs accipitrintjs accipitrinus (Linnaeus)

Type locality: Cayenne by subsequent designation Maraca and Obidos (Sneth-
lage)

1 d\ north bank of Amazon near Obidos
1 & 3 9 , Obidos (Carnegie Mus.)

222. Deroptytjs accipitrinus fuscifrons Hellmayr

Type locality: Rio Acara, Igarape-Assu

Para (Natterer, Stone); Benevides (Steere); Igarape-Assu and Peixe-Boi
(Hellmayr); Rio Capim (Goeldi); Rio Jarriauchim (Snethlage)
1 cf, Rio Acara, Acara
1 9 , Rio Tapajoz, Tauary

1 9 , Benevides (Carnegie Mus.)
1 cf 1 9 , Rio Tapajoz, Miritituba (do.)

These two very distinct races are supposed to replace each other on
the two sides of the main Amazon. According to Hellmayr, the present
subspecies ranges westward for an undetermined distance. In Brazil
there are apparently no definite records west of the Rio Tapajoz, but a

148 bulletin: museum of comparative zoology

specimen in the British Museum from Sarayacu, east Ecuador is
fuscifrons. Hellmayr regarded this locality as "very doubtful", and
Chapman does not record the genus from Ecuador. Pinto claims that
two specimens from Santarem are obviously the northern race!

223. Pionopsitta caica (Latham)

Type locality: Cayenne

2 ^,1 9 , Rio Jamunda, Faro (Snethlage); Rio Atabany (Pinto)
1 9 , Obidos (Carnegie Mus.)

These are the only records for this common Guiana species in our
area. Another member of the genus, barrabandi (Kuhl), just reaches
extreme northwestern Brazil, and one of the tributaries of the Rio
Madeira.

224. Gypopsitta vulturina (Kuhl)

Type locality: Brazil

Para (Wallace, Stone); Castanhal, Rio Gurupy (Stone); Igarape-Assu and
Peixe-Boi (Hellmayr); Rio Capim (Goeldi); Providencia, St. Antonio do
Prata, Rio Moju, Rio Tocantins, Rio Tapajoz (Snethlage)
1 d", Rio Acara, Acara
1 d\ Benevides (Carnegie Mus.)
8 cf 4 9 , Santarem (do.

1 9 , Rio Tapajoz, Villa Braga (do.)

This genus is practically endemic in our area, but on the south bank
of the Amazon only. Westward it ranges to the right bank of the Rio
Madeira. In coloration it so exactly resembles Pionopsitta barrabandi
as to raise the presumption that the two birds are representative forms,
with perfectly complementary ranges. We know of no better illustra-
tion in tropical American parrots to endorse the suspicion that undue
weight has been given to striking color differences, and that there are
far too many "species" and "genera."

225. Touit purpurata purpurata (Gmelin)

Type locality : Cayenne

Para (Natterer and Snethlage); Rio Capim (Wallace); Ipitinga (Hellmayr)
1 d", Benevides (Carnegie Mus.)

Touit hueti (Temminck). Type locality: Peru. Para (Snethlage).

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA. 149

This is a species of far upper Amazonia, and there is no other
definite record for Brazil. The genus is exceedingly difficult to collect
and is easily overlooked, but Snethlage's locality "Para" would
certainly seem to need confirmation. Pinto (1937) probably correctly
omits this species from the Brazil list.

226. PlONITES MELANOCEPHALA MELANOCEPHALA (Linnaeus)

Type locality: Cayenne

Maraca and Obidos, north side of the Amazon (Snethlage, Pinto)
3 d" 4 9 , Obidos (Carnegie Mus.)

227. PlONITES LEUCOGASTER LEUCOGASTER (Kuhl)

Type locality: Brazil

Para (Natterer); Rio Muraiteua, Utinga (Stone); Igarape-Assu, San Antonio
do Prata, Peize-Boi (Hellmayr); Providencia and Rio Acara (Snethlage)

1 9 , Para, Bosque

1 d", Santarem (Carnegie Mus.)

This species is replaced by xanthurus Todd on the Rio Purus, and by
xanthomerius (Sclater) in extreme upper Amazonia east to the Rio
Solimoes. Hellmayr's record (1910) of leucogaster xanthomerius from
the Rio Madeira probably belongs to xanthurus Todd. All are sub-
species, and it is more than likely that all three should be regarded as
representative forms of melanocephala. We have examined the series in
the Carnegie Museum, and note that no mention of xanthurus is made
in Peter's Check-List, (vol. 3) or in Pinto's Catalogue.

Family CUCULIDAE

228. Coccyzus minor minor (Gmelin)

Type locality: Cayenne
Cajutuba (Natterer)

The most southern record for this species. It should be sought along
the coast, and a Brazilian series should be compared with Guiana
topotypes.

150 bulletin: museum of comparative zoology

229. Coccyzus melacoryphus Vieillot

Type locality: Paraguay

Para, Rio Tocantins, Rio Xingu, Monte Alegre, Erere, Rio Maecuru, Rio
Jamunda (Faro.)

4 c? 2 9 , Rio Tapajoz, Santarem.

4 o71 4 9 , Santarem (Carnegie Mus.)

230. Coccyzus euleri (Cabanis)

Type locality: Cantagallo, southeast Brazil
Santarem (Chapman and Riker) ; Pard (Snethlage)

1 d71 , Rio Tapajoz, Santarem

1 9 , Santarem (Carnegie Mus.)

There is still considerable uncertainty regarding the status and
proper name of this species. There is no doubt that C. euleri Cabanis
is definitely a rare cuckoo of Argentina and southern Brazil, ranging
north to the south bank of the Amazon, perhaps as a winter visitor
only. What we do not know is its relationship to C. americanus, of
which it may be only a representative form. Ridgway inclined to this
view, and believed that the type of C. julieni Lawrence, from Som-
brero Key, Lesser Antilles, was an earlier name for euleri, and that the
breeding bird of the West Indies was probably the same thing. There
are two difficulties involved. In the first place breeding adults from
Santo Domingo are not separable from typical americanus. In the
second place the type of julieni was a fall migrant only on Sombrero, a
barren rock with a lighthouse. It follows that julieni can only be a
migrating individual of americanus, and the senior author, who is
familiar with Lawrence's type, regards it as an immature americanus
of minimum size, and certainly not euleri of southern Brazil. We con-
sequently endorse Hellmayr's recent contention (1929, p. 432), not to
call euleri, americanus julieni. We have examined the type of lindeni
Allen from Santarem, which is certainly a synonym of euleri. Some
Cuckoo, either americanus or euleri, is recorded from the Guianas
and Trinidad, though there are no specimens on record in America.
They will almost surely prove to be euleri as a winter visitant, or the
West Indian breeding stock of americanus, which is as yet unknown
definitely east of Venezuela. Snethlage's record of C. americanus from
Para presumably belongs here.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 151

231. Piaya cayana cayana (Linnaeus)

Type locality: C

"ay

enne

Para, Bosque

Rio Acara, Acara

7 tf" 4

Rio Tapajoz, various localities east bank

, Obidos (Carnegie

Mus.)

d1,

, Benevides (do.)

3c?2

Santarem (do.)

2 c? 3

Rio Tapajoz (do.)

The identity of the squirrel cuckoos of Lower Amazonia has been
left in abeyance for some years, due to the inability of any student to
compare a topotypical series of obscura Snethlage from the Rio Purus
with adequate material from any other part of Amazonia. Thus
various museums, lacking a Rio Purus series, have referred specimens
from the Rio Madeira to obscura on geographic grounds, and on the
other hand have acted similarly with birds from northeastern Peru.
The latest comment on Lower Amazon birds is that of Hellmayr (1929,
p. 434), who showed that. Para birds were a different subspecies from
any in Brazil just south of the Amazon valley, and intermediate be-
tween pallescens Cabanis and Heine and obscura Snethlage to which
he assigned all birds from the Rio Madeira to eastern Peru. These
birds have just been named hellmayri by Pinto (1938).

We have before us a nice series of topotypes of obscura from the Rio
Purus. It turns out that obscura of recent authors, nee Snethlage, is a
composite of two races. Rio Madeira birds in the American Museum
are obscura, but specimens in New York and Chicago from eastern and
northeastern Peru are not obscura, immediately separable in being
much darker below, and constitute a connecting link between obscura
and boliviano, Stone. Whether these birds have sufficient geographic
range and clear cut characters to merit the description of still another
subspecies must be left to Mr. Zimmer, who alone has adequate ma-
terial at his command. But our impression of very inadequate material
is that it is as distinct a race as many another now currently recognized.

We may now return to the identity of Lower Amazonian birds. With
the darker Peruvian element extracted from our concept of obscura, we
see no necessity for describing a lower Amazonian race. Birds from
Para and the Rio Tapajoz are clearly distinct from pallescens, but they
are not clearly distinct from obscura. Perhaps one in four specimens
approaches pallescens either in paler upper parts, paler throat, or
paler gray ventral surface, but it will be apparent that lower Ama-
zonian birds have no color character of their own, and are very much

152 bulletin: museum of comparative zoology

nearer obscura on the average than anything else. There is a slight
tendency for tail length to increase as the westward limits of the range
are approached, but the difference between the extremes of the two
series is less than half the difference between extreme individuals in
either series.

Rio Purus <? 280-294; 9 243-278

Rio Tapajoz tf1 271-290, the majority below 2S0 mm.; 9 245-265

Para region 9 256-270

The character used by Madame Snethlage to distinguish her 3
specimens of obscura from Para specimens of what she called P. cayana
(Linnaeus) was the much darker color of the under tail-coverts. This
is apparently a matter of individual variation. One Rio Purus bird is
indeed quite the darkest of all Amazonian specimens seen. Five others,
slightly paler, are identical with ten more eastern specimens. One is still
paler and agrees with four more eastern specimens. It is interesting to
note that the darkness of coloration of the thighs and under tailco verts
does not vary proportionately with the relative darkness or paleness
of the abdomen. We have no hesitation, therefore, in referring all
birds from the south side of the Amazon from the Rio Purus eastward
to obscura Snethlage.

It now remains to determine the relations between obscura and true
cayana from Cayenne. A very fine series in the Carnegie Museum from
French Guiana is at hand, which gives an adequate idea of the con-
siderable individual variation that occurs. It seems to us impossible to
separate obscura. Half the series differ from the majority of true
cayana in having minutely darker under tailcoverts. There are no
other color differences whatever, and there proves to be no difference
in tail length. This seems to us to be far too slight a character for
nomenclatural recognition. The birds of Lower Amazonia can all be
referred to typical cayana, it being understood that eastward an in-
creasing percentage of specimens approach paUcscens in one or another
characteristic.

We give no locality records, as this abundant bird has been obtained
in every part of our area by all collectors, with the single exception of
Marajo Island.

232. PlAYA MELANOGASTER MELANOGASTER (Vieillot)

Type locality: Cayenne, by subsequent designation
2 cf 1 9 , Obidos (Carnegie Mus.)

3 9 , Rio Tapajoz, Villa Broga (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER MAAZONIA 153

Not previously recorded east of the Rio Madeira. The Obidos birds
agree perfectly with a fine series from Cayenne. Those from the
Tapajoz and a fine series from the Rio Purus and Rio Solimoes do not
appear to differ constantly in color characters, but the bills are much
darker, apparently more red, less orange or yellow in life.

233. Coccycua minuta minuta (Vieillot)

Type locality: Cayenne

Para (Layard, Wallace, Stone); Utinga (Hellmayr); Mexiana Island (Hag-
mann and Hellmayr); Marajo Island, Cussary, Rio Jary, Monte Alegre,
Rio Jamunda (Snethlage); Santarem (Chapman and Riker, Hellmayr);
Rio Cunany (Pinto)

1 9 , Rio Amazonas, north bank near Obidos
1 cf 1 9 , Obidos (Carnegie Mus.)
1 9 , Benevides (do.)
3 9 , Santarem (do.)
1 c? 2 9 , Rio Tapajoz, various localities (do.)

Birds from the Lower Amazon are not separable from a fine series
from Cayenne in the Carnegie Museum.

234. Tapera naevia naevia (Linnseus)

Type locality: Cayenne

Para (Layard Stone); Igarape-Assu (Hellmayr); Castanhal" (Stone); Mexiana
Island (Hagmann, Hellmayr, Wallace); Marajo Island, Quati-Puru,
St. Antonio do Prata, Maraca, Monte Alegre (Snethlage)
1 d" 1 9 , Rio Tapajoz, Santarem
3 d" 2 9 , Obidos (Carnegie Mus.)
1 c? 1 9 , Benevides (do.)
1 d\ Santarem (do.)
1 d", Rio Tapajoz, Itaituba (do.)

235. Neomorphus geoffroyi geoffroyi (Temminck)

Type locality: Brazil

Para (Natterer and Snethlage); Igarape-Assu and St. Antonio do Prata (Hell-
mayr); Rio Capim (Goeldi); Cussary (Snethlage)
According to Temminck's account in the Planches Col., his descrip-
tion was based on birds in two private collections and on specimens
in the museums of Paris, Berlin, Vienna and Leyden (given in the
order mentioned). As one of the private collections was that of Prince
Neuwied, Temminck obviously had Bahia as well as Amazonian

154 bulletin: museum of comparative zoology

specimens. As the Bahia bird has been described as dulcis Snethlage,
1927, the question arises which of the two subspecies involved should
actually be typical gcoffroyi. In 1905, (p. 298) Hellmayr designated
Bahia ex Wied as are stricted type locality, but in 1929, (p. 436), after
the description of dulcis by Snethlage, suggested that the application
of Temminck's name should await "the examination of the type in
Leyden." There being no holotype, and Leyden being the last of the
numerous listed collections to be mentioned, we can see no reason for
applying the name on the basis of a specimen in the Leyden Museum.
As a matter of fact we have before us an Amazon specimen of gcoffroyi
and an example of dulcis Snethlage from the interior of Bahia. They
are very different subspecies, and Temminck's detailed description
and excellent colored plate apply very definitely to the bird from
around Para, which has long been known as gcoffroyi. We suggest
Para, therefore, as a restricted type locality and as a probable source
of many of the older skins in the great collections. We might add that
all detailed descriptions of gcoffroyi in the literature, based on Para
birds, agree with Temminck's original description and plate in all the
respects, which we now know to be of subspecific importance.

236. Neomorphus squamiger squamiger Todd

Type locality: Colonia do Mojuy, Santarem, Brazil

Previously known only from the four specimens in the type series in the
Carnegie Museum, examined by us

1 a"1 1 ? (both adult), Rio Tapajoz, Tauary (right bank)

The rarity of the Cuckoos of this genus is readily seen by the fact
that Klages, the discoverer of squamiger, "sought for it in vain along
the Tapajoz." Further comments on this form will be found under
the next.

237. Neomorphus squamiger iungens subsp. nov.

Type. — No. 173564, Mus. Comp. Zool.; 9 ad.; Boim, left bank of the Rio
Tapajoz, Para, Brazil; Jan. 10, 1933; A.M. Olalla

Characters. Differing from squamiger in having the feathers of fore-
head and pileum buffy brown tipped with dull bluish and with bluish
centers; mantle slightly more bronzy green; secondaries greener, less
coppery red; auricular region much deeper cinnamon buff; chin and
throat uniform rich buff, instead of soiled whitish or grayish; dark
subterminal area of breast feathers less extensive; narrow pectoral

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 155

collar much more distinct. Differing from lepidophanes in the buffy
forehead and pileum; secondaries bronzy green, not coppery red; chin
and throat buffier, less white; breast squamated, and much narrower
pectoral collar.

In Todd's original description of his two supposed species (Proc.
Biol. Sec. Wash., 38, 1925, p. 112), lepidophanes was compared with
pucherani Deville, but squamiger was not compared with any member
of the genus. His only comment was that squamiger was unique in the
restriction of the bare area on the side of the head. Our birds, how-
ever, show that this is an age or individual variation, as they have
just as big a bare orbital space as geoffroyi and bigger than several
specimens of salvini.

Many years ago (1910) Hellmayr predicted that geoffroyi, pucherani
and salvini would probably turn out to be representative forms. The
discoveries of recent years amply support his contention. These rare
terrestrial cuckoos are an old group of undoubted Old World origin,
which have probably occupied their present ranges for a very long
time. We now have definite evidence that, like the Trumpeters, the
larger rivers are barriers they will not cross, and we find very different
birds on opposite banks of these rivers. Our prediction is, therefore,
that still other races of these cuckoos remain to be discovered, and that
they will turn out still more closely to connect birds currently regarded
as specifically distinct.

At the present state of our definite knowledge, it must be admitted
that at first sight geoffroyi and squamiger appear very different. How-
ever, it should be noted that some at least of the geoffroyi characters
reappear in other races further west. In particular, iwigens is in
general coloration a complete "throw back" to geoffroyi, and salvini
aequatorialis Chapman is nearer in general coloration to geoffroyi than
any of the intermediate "species". We present the following synopsis.
I. Whole side of head and neck barred.

a. Forehead and pileum cinnamon; breast light cinnamon
buff; chest browner; subterminal bars on feathers of breast
dusky, narrower. Typical geoffroyi. Para region to the Rio
Acara. (ranging further westward?)

b. Forehead and pileum paler, buffy; breast paler, buffy; chest
grayer; subterminal bars of breast feathers much broader.
dulcis Snethlage. Eastern and southeastern Brazil.

II. Side of head and neck uniform, never barred.

a. Throat and breast strongly squamated, caused by black
subterminal bars.

156 bulletin: museum of comparative zoology

1. Forehead and pileum mostly bluish like crest; secondaries
coppery green, rather than olive green as in geoffroyi; chin
and throat and chest soiled whitish; subterminal bars on
breast very broad, sometimes the entire feather black
with a grayish tip; scarcely any pectoral collar, squamiger
Todd. Right bank of the Rio Tapajoz, for an unknown
distance eastward.

2. Forehead cinnamon buff as in geoffroyi; secondaries olive
green as in geoffroyi; chin, throat and chest rich buffy; sub-
terminal bars on breast feathers narrower; a definite but
very narrow pectoral collar ; (in these respects intermediate
between geoffroyi and squamiger). iungens Griscom and
Greenway. Left bank of the Rio Tapajoz to the right
bank of the Rio Madeira.

3. Forehead bluish as in squamiger; secondaries coppery red;
chin and throat whitish as in squamiger; breast and chest
clay color much as in geoffroyi; underparts rich buff as in
iungens; squamation on breast much as in iungens;
pectoral collar broad, lepidophanes Todd. Rio Purus to
right bank of Rio Solimoes, perhaps east to left bank of
Rio Maderia.

b. Breast faintly squamated, the feathers with pale tips, but

subterminal bar, if present, only faintly darker than rest
of feather.

1. Forehead rich cinnamon rufous as in geoffroyi; second-
aries coppery, intermediate between squamiger and
lepidophanes; underparts intermediate between lepido-
phanes and pucherani, the color of the belly and under-
tailcoverts much less rich and dark than any preceding
race, and less contrasted with chest; pectoral collar broad
and complete as in (lepidophanes and pucherani) . aequatori-
alis Chapman. Known only from a few localities in Ama-
zonian Ecuador.

2. Forehead much paler, less rufescent; pectoral collar
narrower and incomplete, salvini. Southern Central
America and parts of Colombia.

c. No squamation of any kind; feathers of breast and chest
entirely uniform.

1. Forehead bluish or purplish, as in squamiger and lepido-
phanes; secondaries rich coppery red as in lepidophanes;
chin to chest uniform clay color, the extreme in this

GRISCOM AND GREEN WAY: BIRDS OF LOWER AMAZONIA 157

direction of tendencies in aequatorialis and lepidophanes;
entire balance of underparts also clay color, the vent and
under tailco verts only faintly darker; broad complete
pectoral collar, pucker ani Deville. Rio Ucayali in
northeastern Peru east to the left bank of the Rio Soli-
moes (fide Todd), but series from these two extremes
should be compared.
In conclusion we wish to point out that no Neomorphus is known
from the Rio Acara to the Rio Tapajoz, and the genus is as yet unre-
corded in the enormous area on the north side of the main Amazon from
west of the Rio Negro to Amazonian Ecuador. We also suggest that
the differences between radiolosus and rufipennis and any member of
the group here discussed are the criteria for valid specific differences in
this genus.

238. Dromococcyx pavoninus Pelzeln

Type locality: Araguaya, Brazil

1 9 imm., Obidos (Carnegie Mus.)

While not previously recorded from Lower Amazonia, the occurrence
of this little known species was to have been expected.

239. Crotophaga major Gmelin

Type locality: Cayenne

Ipitinga (Hellmayr) ; Rio Capim (Wallace and Goeldi) ; Rio Inhangapy (Stone)';
Mexiana Island (Hagmann); Para, Ilha das Oncas, Benevides, Marajo
Island (Snethlage); Santarem (Chapman and Riker); Caviana Isl. (Pinto)
3 cf 1 9 , north bank of Amazon near Obidos
2 d\ Rio Tapajoz, Santarem and Caxiricatuba
1 cf , Obidos (Carnegie Mus.)
1 d71, Santarem (do.)
1 cf 1 9 , Rio Tapajoz (do.)

240. Crotophaga ani Linnaeus

Type locality: Brazil

Para (Layard) ; Igarape-Assu (Hellmayr) ; Rio Capim (Goeldi) ; Rio Inhangapy
(Stone); Mexiana Island (Wallace and Hagmann); San Antonio do Prata,
Marajo Island (Snethlage) ; Santarem (Chapman and Riker)

2 d1 1 9 , south bank of Amazon, Lago Grande
1 cf 1 9 , Rio Tapajoz, Santarem and Pinhy

1 9 , Obidos (Carnegie Mus.)
1 <f , Santarem (do.)

158 bulletin: museum of comparative zoology

241. Guira guira (Gmelin)

Type locality: Brazil

Mexiana Island (Wallace and Hagmann); Marajo Island, Capanema, Quati-
Puru (Snethlage); Caviana Island (Brodkorb)

One of the many characteristic birds of southern Brazil, which
ranges north along the coast to the mouth of the Amazon.

Family TYTONIDAE

242. Tyto alba hellmayri Griscom & Greenway, 1937

Type locality: Paramaribo, Surinam

Para (Hellmayr, Snethlage, Pinto); Marajo Island (Snethlage); Mexiana
Island (Hagmann); Santarem (Hellmayr)
1 oM 9 , Rio Tapajoz, Santarem

A much larger bird than the southern tuidara, the distinctness of
which Hellmayr suspected years ago.

Family STRIGIDAE

243. Asio stygius stygius (Wagler)

Type locality: Brazil

1 cf 2 9 , Rio Tapajoz, Pinkhy and Tauary

These birds constitute the first definite record for the Lower Amazon.
[Rhinoptynx clamator can confidently be expected in our area, and
should be sought for carefully]

244. Bubo virginianus subsp.

Vicinity of Para (Snethlage)

This bird, if not representing an undescribed race, might be deserti
Reiser. The horned owl should be sought on Mexiana and Marajo
Islands.

245. Pulsatrix perspicillata perspicillata (Latham)

Type locality:

Para (Natterer); Marajo Island, Monte Alegre, Rio Tocantins, Rio Tapajoz
(Snethlage); Santarem (Chapman and Riker)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 159

246. Otus choliba crucigerus (Spix)

Type locality: Amazon River

Mexiana Island (Wallace, Hagmann); Para (Stone); Para, Marajo Island, Rio
Tocantins (Snethlage); Obidos (Hellmayr)
2 tf 1 9 , 1 ?, Rio Tapajoz, Pinhy

1 9 , north side of Amazon near Obidos
1 d\ Obidos (Carnegie Mus.)
1 cf , juv., Santarem (do.)
1 9 , Rio Tapajoz (do.)

This is the widely distributed race of the Guiana — Amazon region.
In the campo country further south it is replaced by decussatus
(Lichtenstein).

247. Otus watsonii usta (Sclater)

Type locality: Ega, upper Amazon

Rio Tapajoz, Pinhel (Snethlage) ; Utinga (Pinto)

1 cf 1 9 , Rio Tapajoz, Tauary and Caxiricatuba

In default of adequate material, these birds are subspecifically
identified only provisionally, but Chapman (Birds Ecuador, 1926, p.
246) has restricted the type locality of watsonii Cassin to the "Napo,
Region, east Ecuador", while ustus Sclater was from "Ega on the
upper Amazon". Compare also Naumburg, Birds of Matto Grosso,
1930, p. 117, who inadvertently uses the combination "usta watsoni".

248. Lophostrix cristata cristata (Daudin)

Type locality: Guiana

Para (Wallace); Santarem (Chapman and Riker); Monte Christo and Obidos
(Pinto)

2 cf , Rio Tapajoz, Tauary and Caxiricatuba
1 cf , Santarem (Carnegie Museum)

249. Ciccaba virgata superciliaris (Pelzeln)

Type locality: Brazil

Para (Wallace); Ipitinga (Hellmayr); Rio Curua (Snethlage); Murutucu
(Pinto)

1 cf , Rio Tapajoz, Tauary

This specimen agrees with the Para bird described by Sharpe in the
Catalogue of Birds in having a black and white barred tail. Pinto

160 bulletin: museum of comparative zoology

records the Murutucu specimen as true virgata, which is quite im-
possible. Presumably his bird represents the dark phase of super-
ciliaris.

250. Ciccaba huhula (Daudin)

Type locality: Cayenne

1 9 , Obidos (Snethlage)

It is very doubtful if huhula and nigrolineatum are really specifically
distinct.

251. Glaucidium brasilianum brasilianum (Gmelin)

Type locality: Brazil
Para (Snethlage)

According to Hellmayr typical brasilianum ranges north to the south
bank of the Amazon. The species will surely be found on the north
bank, where there is open country. Such birds might prove to be
phalamoides.

Family NYCTIBIIDAE

252. Nyctibius griseus cornutus (Vieillot)

Type locality : Paraguay

Para (Natterer and Snethlage) ; Rio Capim (Layard) ; Rio Jamauchim (Sneth-
lage); Murutucu (Pinto)

2 cf, Rio Tapajoz, Tauary and Caxiricatuba
1 9 , Santarem (Carnegie Mus.)

Amazonian birds are really intermediate between the large cornutus
and the very small griseus of the Guianas. Our birds are distinctly
nearer the former, however, and have more pronounced barring on the
inner webs of the primaries. It is possible that specimens from the
north bank of the Amazon might prove referable to griseus.

253. Nyctibius longicaudatus (Spix)

Type locality: Rio Japura, Brazil

1 cf, Rio Capim, Resacca (Snethlage)

254. Nyctibius ^ethereus ( Wied)

Type locality: Bahia

Mexiana Island (Hagmann, fide Hellmayr)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 161

In default of specimens we can only cite the records for these two
species, which are still very rare in collections. Our belief, however, is
that they are two representative subspecies, and that the two records
from our area really deal with an intermediate population which is a
connecting link between the two.

255. Nyctibius grandis (Gmelin)

Type locality: Cayenne

Para, Rio Moju, Marajo Island (Hellmayr and Snethlage); Obidos (Hellmayr);
Santarem and Pataua (Pinto)

2 d\ 5 9 , north bank of Amazon near Obidos
1 cf , 1 ?, Rio Tapajoz, Santarem

2 9 , Santarem (Carnegie Mus.)

Family CAPRIMULGIDAE

256. Chordeiles acutipennis acutipennis (Hermann)

Type locality: Cayenne

Cajutuba (Natterer) ; Para, Marajo Island, Monte Alegre, Rio Jamunda, Rio
Tocantins, Rio Xingii (Snethlage)

9 d" 1 9 , Santarem (Carnegie Mus.)

3 d" 1 9 , Rio Tapajoz, Apaey (do.)

As with many widely ranging South American birds, this Night-
hawk is larger in the south. Guiana birds are the smallest, and
those from the extreme southern limits of the range are appreciably
larger. It follows that intermediate birds occupy the greater part of
the range in eastern South America, and we see no point in recognizing
a southern race, for which brasilianus (Gmelin) is the earliest of
several available names, which probably apply to this species.

257. Chordeiles rupestris rupestris (Spix)

Type locality: Rio Negro
Rio Tapajoz (Snethlage)

1 cf 19, Rio Tapajoz, Pinhel and Caxiricatuba
10 o71 11 9 , Rio Tapajoz, various localities (Carnegie Mus.)

The subspecific variations of this species remain to be determined,
and those proposed require confirmation with a good series of topo-
types from the Rio Negro, as well as series from the type localities

162 bulletin: museum of comparative zoology

of zaleucus Oberholser (Pebas, Peru) and xyostictus Oberholser
("Bogota", Colombia, where the species does not occur!). As a whole it
is an upper Amazonian species, and on the south side of the Amazon
has been found eastward only on the Rio Madeira and the Rio Tapajoz.
The record of "Para" in Ihering's Cat. Fauna Bras., 1907, p. 132, cer-
tainly requires confirmation.

If there really is any racial variation in this species, one would
expect that the birds from the south side of the Amazon would differ
from those of the Rio Negro. Hellmayr (1910) claimed that the
differences between Peruvian and Rio Madeira specimens were "in-
significant". It is possible, therefore, that our birds from the Rio
Tapajoz may prove to be zaleucus.

258. Nyctiprogne leucopyga (Spix)

Type locality: Amazon River

Monte Alegre and Rio Jamunda, Faro (Snethlage)

1 d" 2 9 , 1 ?, Rio Tapajoz, Pinny

1 9, Santarem (Carnegie Mus.)

3 cf 6 9 , Rio Tapajoz, various localities (do.)

A characteristic river forest Nighthawk, which reaches the southern
limit of its range on the south side of the Amazon. In the more open
campo country it is replaced by Nannochordeiles pusillus, which is
unknown in the Amazonian forest area.

259. PODAGER NACUNDA NACUNDA (Vieillot)

Type locality: Paraguay

Para (Hellmayr, Stone) ; Rio Capim (Wallace) ; Quati-puru and Rio Tocantins
(Snethlage)

3 cf 6 9 , R-io Tapajoz, various localities.

6 c? 15 9 , Santarem (Carnegie Mus.)

The wings of our females measure 234-246 mm., thus running
slightly smaller than birds from southern Brazil. It is possible that
specimens from the northwestern corner of our area (Rio Jamunda to
Obidos) might prove referable to minor Cory of the Rio Branco,
Brazil, northward.

260. Lurocalis semitorquatus nattereri (Temminck)

Type locality: Brazil

Para (Wallace); Strada Braganca (Layard); Ilha das Oncas (Snethlage)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 163

A very little known bird in our area, apparently more common
southward. Typical semitorquatus of the Guianas is even less known,
and its southern limits remain to be determined. A specimen collected
by Natterer on the Rio Icanna is currently referred to this race. This
river is a small tributary on the west side of the Rio Negro in extreme
northwestern Amazonas, near both the Colombian and Venezuelan
boundaries.

261. Hydropsalis crasiliana crasiliana (Gmelin)

Type locality: northeastern Brazil

Rio Xingu and Santarem (Snethlage); Santarem (Pinto).

1 oM 9 , Rio Tapajoz, Santarem

2 9 , Santarem (Carnegie Mus.)

Another little known species, which ranges west to the Rio Madeira
and south to Bahia and Matto Grosso. Still further south it is re-
placed by the larger and paler furcifera (Vieillot).

262. Hydropsalis climacocerca canescens
Griscom & Greenway, 1937.

Type locality: Lago Grande, south bank of Amazon, west of Rio Tapajoz

2 cT 2 9 , Lago Grande; 1 d" , Rio Tapajoz, Pinhel

1 cT, Rio Tapajoz, Itaituba (Carnegie Mus.)

As yet only known from the localities listed above, and possibly
Manacapuru, Rio Solimoes.

263. Hydropsalis climacocerca pallidior Todd, 1937.

Type locality: Santarem

14 d* 5 9 , Santarem (Carnegie Mus.)

264. Hydropsalis climacocerca intercedens, Todd, 1937.

Type locality: Islands in Amazon, near Obidos
1 cf 3 9, Obidos (Carnegie Mus.)

This distinct subspecies is, in our opinion, an obvious connecting
link between climacocerca and the dark schomburgki of British Guiana.

In spite of the two recent papers on this species, the writers do not
agree wholly in their interpretation of the same material, and other

164 bulletin: museum of comparative zoology

points still await proper series and study. Mr. Todd agrees with us
that a fourth very distinct, buffy subspecies occurs on the Rio Purus.
He refers Rio Solimoes birds to canescens, while we suspect that they
are subspecifically distinct. The question still remaining to be settled
is what are the exact subspecific characters of true climacocerca from
the lower Ucayali River, Peru? We assumed that the Rio Solimoes
birds might represent it. Mr. Todd thinks that the Rio Purus birds
probably do. On the other hand, we were not prepared to separate
birds from the two banks of the Rio Tapajoz as different subspecies.
It follows, therefore, that good series from the Rio Ucayali and the
Rio Solimoes may alter the present picture in one of several possible
ways.

Incidentally the British Museum Catalogue records a specimen by
Wallace from the "Rio Tocantins". There is every possibility that
Tonantins in Upper Amazonia is actually intended. In no case was
Pinto (1938) justified in ascribing this record to typical climacocerca.

265. Nyctidromus albicollis albicollis (Gmelin)

Type locality: Cayenne

Recorded from every collecting locality in our area
5 cf , Rio Tapajoz, various localities.

1 cf 1 9 , Obidos (Carnegie Mus.)

2 cf 1 9, Santarem (do.)

4 9 , Rio Tapajoz, Apacy (do.)

266. Nyctiphrynus ocellatus ocellatus (Tschudi)

Type locality: Peru

1 9 , Rio Tapajoz, Boim (Snethlage)
1 9 , Rio Tapajoz, Tauary (M.C.Z.)

As shown in our revision (1937) this rare Whippoorwill is repre-
sented by brunnescens Griscom and Greenway in southern Brazil and
by lautus Miller and Griscom in Central America.

267. Nyctipolus nigrescens nigrescens (Cabanis)

Type locality: British Guiana

Para (Natterer and Wallace) ; Utinga (Pinto) ; Rio Acara, Rio Tocantins, Rio
Tapajoz, Rio Jary, Rio Jamunda (Snethlage)

3 cf 1 9 , Rio Tapajoz, Pinhel and Caxiricatuba

1 cf , Obidos (Carnegie Mus.)

2 d71 , Santarem (do.)

1 9 , Rio Tapajoz, Villa Braga (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 165

See Griseom and Greenway, 1937, for our understanding of racial
variation in this species.

268. Caprimulgus rufus rufus Boddaert

Tjfpe locality: Cayenne

Para (Natterer); Santarem (Pinto)

2 cf, Rio Tapajoz, Tauary and Pinhy

6 d\ Santarem (Carnegie Mus.)

As we understand it, true rufus is restricted to the Guianas and
northeastern Brazil, south possibly to Bahia. Still further south we
find a paler less rufescent bird, for which the name rutilus Burmeister is
probably correct.

269. Caprimulgus sericeocaudatus (Cassin)

Type locality: South America, Brazil or Venezuela

1 9 , Santarem, Nov. 6, 1919 (Carnegie Mus.)

This Caprimulgus is definitely known only from the type, which is
presumably a male. The female, listed above, probably belongs here,
when due allowance is made for the sex differences known in the related
species of the genus. The holotype has not, however, been examined in
the present connection.

270. Caprimulgus maculicaudus (Lawrence)

Type locality: Para

Para (Hellmayr, Stone); Rio Acara (Hellmayr); Marajo Island and Arumand-
uba (Snethlage)

1 d\ Marajo Island (Linden in M.C.Z.)

2 cf 1 9 , Santarem (do.)

1 d\ Obidos (Carnegie Mus.)
12 d" 6 9 , Santarem (do.)

This species is still rare in collections. Hartert in the Cat. Birds
Brit. Mus., 16, p. 575, gave its range as the Andes from Colombia to
Peru, regarding Lawrence's type locality as incredible. Hellmayr's
notes on this species (1907, p. 397) should be consulted.

271. Caprimulgus parvulus parvulus Gould

Type locality: Rio Parane

Para, Benevides, Rio Xingu (Snethlage) ; Santarem (Chapman and Riker)

1 cf imm., Para, Bosque

1 9, Santarem (Carnegie Mus.)

166 bulletin: museum of comparative zoology

Family MICROPODIDAE

272. Chaetura spinicauda spinicauda (Temminck)

Type locality: Cayenne

1 9 , Obidos (Carnegie Mus.)

273. Chaetura spinicauda aethalea Todd, 1937.

Type locality: Benevides, Para, Brazil.
Para (Layard and Natterer) ; Santarem (Wickham)
6 cf 1 9 , Benevides (Carnegie Mus.)
1 cf, Santarem (do.)

274. Chaetura brachyura Jardine

Type locality: Tobago

Para (Layard and Snethlage)

2 cf 1 9 , Rio Tapajoz, Apacy, Itaituba (Carnegie Mus.)

275. Chaetura chapmani subsp.

1 cf Benevides (Carnegie Mus.)

In default of comparative material, no subspecific identification is
attempted.

276. Reinarda squamata squamata (Cassin)

Type locality: British Guiana

1 cf Santarem (Carnegie Mus.)

277. Panyptila cayannensis (Gmelin)

Type locality: Cayenne

Nazare (Layard); Para (Stone); Para, Apehu, Rio Jamunda (Snethlage)

Family TROCHILIDAE

278. Threnetes leucurus medianus Hellmayr

Type locality: Tury-assu, Maranhao, Brazil

Para and S. Antonio do Prata (Hellmayr); Snethlage for both these localities;
Para, Rio Inhangapy (Stone)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 167

This recently proposed form (1929, p. 381) connects the two sup-
posed species leucurus and cervinicauda, but is a geographic extreme.
Hellmayr refers Rio Madeira birds to typical leucurus of the Guianas.

279. Glaucis hirsuta hirsuta (Gmelin)

Type locality: northeastern Brazil

Para (Layard, Hellmayr, Snethlage, Stone); S. Antonio do Prata (Hellmayr);
Marajo Island (fide Simon); Utinga, Murutucu (Pinto)

1 cf, Benevides, (Carnegie Mus.)

2 d\ Rio Tapajoz (do.)

The Benevides bird does not differ from Guiana specimens.

280. Glaucis hirsuta subsp.

1 cf1 imm., north bank of Amazon near Obidos

As Hellmayr has correctly remarked, the racial variation of this
species is still unsettled in eastern South America, due to inadequate
material. Five specimens from Surinam differ from true hirsuta in two
characters claimed for roraimop Boucard by Simon, the brighter
rufescent chest contrasted with the paler belly and the more extensive
black subterminal area on the outer rectrices. Specimens from Trini-
dad and Grenada do not show these characters, however, and we can
see no difference between them and a typical hirsuta. Our one bird
from the north bank of the Amazon is intermediate between hirsuta
and the Surinam birds in the characters mentioned, but is almost as
dark below as affinis from Panama and north Colombia.

There are also nomenclatural difficulties. The earliest name for any
race north of the Amazon would be mazeppa (Lesson) from "Guiana".
A study of the colored plate and description, however, in comparison
with the description and plate of hirsuta (what Lesson called errone-
ously the 9 of Trochilus superciliosus L.) strongly arouses suspicion
that it might be a much earlier name for- Hetcroglaucis philippinae
Penard, the proper status of which still remains to be determined.

281. Phoethornis superciliosus superciliosus (Linnaeus)

Type locality: Cayenne

Rio Jary, San Antonio do Cachoeira (Snethlage)

2 cf , Obidos (Carnegie Mus.)

168 bulletin: museum of comparative zoology

282. Phoethornis superciliosus mulleri Hellmayr

Type locality: Para

San Antonio do Prata, Para, Rio Inhangapy (Stone); Peixe-Boi, and Ipitinga

(Hellmayr); Mocajatuba, Ananindeua, Santa Isabel, Rio Moju, Rio

Tocantins (Snethlage); Murutucu (Pinto)

1 9 , Para, Bosque

2 9 , Rio Acara, Buenos Aires

2 d\ 1 ?, Rio Tapajoz, Pinhy and Tauary

1 cf , Benevides (Carnegie Mus.)
1 c? 2 9 , Santarem (do.)
1 o71 1 9 , Rio Tapajoz, Miritituba (do.)

283. Phoethornis superciliosus insignis Todd, 1937

Type locality: Itaituba, left bank Rio Tapajoz, Brazil

5 d1 1 9 , Rio Tapajoz, Villa Braga and Itaituba, (Carnegie Mus.)

A well marked subspecies, presumably ranging west to the right
bank of the Rio Madeira. Across that river to the Rio Solimoes we
find ochraceiventer Hellmayr, and still further west moorei Lawrence.
The distribution of the species on the north bank of the Amazon is still
practically unknown.

284. Phoethornis bourcieri bourcieri Lesson

Type locality: Brazil

1 d", Rio Tapajoz, Caxiricatuba

This species has been completely overlooked in our area, and in
spite of the type locality, the only record for Brazil is Natterer's from
Marabitanas on the Rio Negro. Birds passing as bourcieri are now
well known from the Peruvian, Ecuador and Colombian Amazons,
much rarer northeastward to the Guianas. These latter birds are all
strikingly different from our specimen from the Rio Tapajoz and an
east Ecuador series, in being much less white and grey below, the
underparts almost uniform pale mouse brown, paler on chin, throat
and belly, instead of whitish and greyish with a faint brownish tinge
on the sides. Thanks to the authorities of the American Museum we
have examined ample material listed below. The two races are as
follows :

1. typical bourcieri (Lesson). We designate Rio Tapajoz as a more
restricted type locality. Whiter and greyer below. One record from
lower Amazonia ; 10 cT 2 9 from Amazonian Ecuador belong here also,

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 169

and possibly the Rio Negro specimen described by Pelzeln as Ame-
trornis abnormis, nee Reichenbach nomen nudum.

2. whitelyi Boucard, 1891, ex British Guiana. Below nearly uniform
pale mouse brown. 4 specimens from British and Dutch Guiana, 1 cf
from Mt. Duida, Venezuela. 4 cf from Caqueta, Colombia are inter-
mediate.

This name, Phaethomis whitelyi Boucard, Hummingbird, 1, 1891,
p. 18 (British Guiana) must not be confused with Eremita whitelyi
Boucard, Genera Hummingbirds, 1895, p. 390 (Kanuku Mts., British
Guiana) which is a synonym of episcopus Gould, now a race of ruber.

285. Phoethornis rupurumii amazonicus Hellmayr

Type locality: Santarem

Santarem (Hellmayr) ; Rio Tapajoz, Arumanduba, Monte Alegre (Snethlage) ;
Obidos (Hellmayr)

4 cf 1 9 , north bank of Amazon near Obidos
4 cf 1 9 , Rio Tapajoz, Santarem and adjacent localities
4 cf\ Santarem (Carnegie Mus.)

286. Phoethornis ruber ruber Linmeus

Type locality: Surinam

Para (Layard, Stone); S. Antonio do Prata and Peixe-Boi (Hellmayr); Provi-
dencia, Ananindeua, Maguary, Sta. Isabel, Quatipuru, Rio Tocantins, Rio
Tapajoz (Snethlage); Santarem (Hellmayr); Murutucu, Utinga (Pinto)
1 cf , Para, Bosque
4 c? , Rio Tapajoz, various localities
1 9 , Santarem (Carnegie Mus.)

287. Campylopterus largipennis largipennis (Boddaert)

Type locality: Cayenne

1 cf, Obidos (Carnegie Mus.)

288. Campylopterus largipennis obscurus Gould

Type locality: Amazon valley

Para (Layard, Wallace, Hellmayr, Stone); S. Antonio do Prato (Hellmayr);

Castanhal, Rio Inhangapy (Stone, Snethlage); Peixe-Boi (Hellmayr);

Mocayatuba, Apehu, Sta. Isabel, Benevides, Rio Mojii, Marajo Island

(Snethlage) ; Utinga, Murutucu (Pinto)
5^4 9,3?, Para, Bosque

1 c? 2 9 , Rio Acara, Acara and Buenos Aires

2 cf 1 9 , Benevides (Carnegie Mus.)

2 c? 1 9,1? Rio Tapajoz, Miritituba, Itaituba, Apacy (Carnegie
Mus.)

170 bulletin: museum of comparative zoology

This well known hummer only differs from largipennis (Boddaert)
of the Guianas in having narrower white tips to the lateral rectrices.
It has a remarkably restricted range in northeastern Brazil. This
species is as yet unrecorded between the Rio Acara and the Rio
Madeira, where aequatorialis Gould is supposed to occur. The speci-
mens in the Carnegie Museum prove that this disposition of the case is
unsatisfactory. The Rio Purus birds agree with aequatorialis in two
important particulars: (1) the underparts are a lighter grey, almost
white on the belly and (2) the tail tips are white, not greyish, and
much wider, especially the next to the outermost. The tail above,
however, is coppery green, not bluish green as in aequatorialis. The
Tapajoz bird has the same tail tips as the Purus birds and aequatorialis
but is otherwise like obscurus.

289. Eupetomena macroura macroura (Gmelin)

Type locality: Cayenne

Para, Ilha das Oncas, Monte Alegre (Snethlage); Mexiana Island (Wallace and

Hagmann); Marajo Island (Hellmayr, Snethlage, Pinto); Santarem

(Chapman and Riker)

1 o71, Santarem (Carnegie Mus.)

290. Florisuga mellivora (Linnams)

Type locality: Guiana

Para (Layard, Wallace, Stone); Rio Inhangapy (Stone); S. Antonio do Prata,

Peixe-Boi and Ipitinga (Hellmayr); Benevides, Sta. Isabel, Rio Guama,

Marajo Island, Rio Tocantins (Snethlage)

2 d" 1 9 , Para, Bosque

4 cf 1 9 , Rio Tapajoz, various localities

1 9 , Benevides (Carnegie Mus.)

2 cf , Rio Tapajoz, Apacy (do.)

291. Agyrtrina versicolor nitidifrons (Gould)

Type locality: Para

S. Antonio do Prata and Ipitinga (Hellmayr); Rio Tocantins (Snethlage),
either this or the next race

1 d", Rio Tapajoz, Tauary

1 9 , Santarem (Carnegie Mus.)

This race of the common versicolor of central and southern Brazil
occupies a very restricted area from the south side of the mouth of the
Amazon to Ceara, Piauhy and Maranhao.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 171

292. Agyrtrina versicolor subspecies

5 d" 10 9 2?, near Obidos
1 9 , Obidos (Carnegie Mus.)

Characters. Connecting A. milleri (Bourcier) of the Rio Negro
northwestward with versicolor nitidifrons (Gould) of Para; throat and
breast pure white as in milleri, instead of glittering green as in nitidi-
frons; crown plaque nearly pure green, not bluish green as in milleri,
and restricted as in nitidifrons; bill short, 13.5-15 mm. as in nitidifrons,
instead of 16-16.5 mm. as in milleri.

Remarks. Specific lines have been greatly overdone in this genus,
as with so many other groups of Hummingbirds. Hellmayr (Birds
Northeast Brazil, p. 396) has recently shown that nitidifrons and
versicolor are conspecific, and in a valuable footnote on Simon's types
of mcracnla and laglaizi, alludes to milleri as a subspecies also, without,
however, assigning any definite reasons. The present form, also a
geographic intermediate, provides abundant proof of the correctness
of this view. In two specimens in our series of 17 from the region
around Obidos, there are two or three glittering green feathers on the
otherwise white throat, giving final evidence of complete intergrada-
tion in the most striking and obvious difference between milleri and
nitidifrons.

There are hopeless nomenclatural complications due to the indefinite
"nitidifrons meracula" Simon (Hist. Nat. Troch., pp. 114, 329) and
"milleri laglaizei" Simon. Simon's concept of specific lines was quite
different from that of previous workers. He kept milleri and nitidifrons
as a distinct species, and while giving all the differentiating characters,
did not consider that the white versus glittering green throat was the
outstanding difference between them. Accordingly meracula is briefly
diagnosed on p. 114 as differing from nitidifrons in having the throat
and breast white as in milleri, the bill 15 mm. long. The question
naturally arises whether our bird from Obidos could possibly be
meracula. Two footnotes throw important light on this question.
Simon knew meracula from two sources (1) 2 specimens from the
Wiener mission labelled "Napo" [ = east Ecuador], but the locality
regarded by Simon as erroneous, as the make of the skins suggested
Guiana trade-skins. These specimens differ from nitidifrons [and inci-
dentally milleri) in having green, not bluish green, crown plaques.
(2) a bird from the Orinoco, very much bluer, in some feathers wholly
blue, suggesting A. hollandi Todd, also from Venezuela, a much
earlier name.

172 bulletin: museum of comparative zoology

It will be apparent, therefore, that meracula is a vague composite of
two elements. The differences between it and milleri laglaizei are com-
pletely in the air, as far as Simon's treatment is concerned. In these
circumstances the name must await more definite application. We
note that Hellmayr (loc. cit.) doubts the locality Rio Napo, but
Guiana is equally uncertain, the facts being that it would surely be
remarkable if this species had completely escaped detection in both
regions.

Previous records of this race in our area have been listed as Agyrtrina
milleri. These are Obidos (Hellmayr) and Rio Jamunda, Faro (Sneth-
lage). Hellmayr 's note on his female from Obidos clearly describes the
present subspecies. True milleri (Bourcier) ranges from the Rio Negro
northwestward and northward to Colombia and Venezuela, where
further racial variation may prove to take place.

293. Agyrtrina leucogaster leucogaster (Gmelin)

Type locality: Cayenne
Pan! (Snethlage)

The typical race of this species ranges from the Guianas to north-
eastern Brazil, and the poorly defined bahice Hartert from Pernambuco
southward. True leucogaster does not appear to be uncommon just
south of our area in Maranhao and Piauhy, and the dearth of records
near Para is consequently surprising.

294. Agyrtrina fimbriata fimbriata (Gmelin)

Type locality: Cayenne

Maraca, Monte Alegre, Igarape de Paituna (Snethlage) ; Obidos (Hellmayr)
3 d1, Rio Amazonas, north bank near Obidos

295. Agyrtrina fimbriata nigricauda (Elliot)

Type locality: Bahia

Mexiana Island (Wallace); Marajo Island (Hellmayr); Quati-Puru, Rio
Tocantins, Rio Iriri, Rio Tapajoz (Snethlage) ; near Santarem (Hellmayr,
Pinto).

1 d1 , Para, Bosque

16 d71 7 9 , Rio Tapajoz, various localities

GRISCOM AND GREENWAY : BIRDS OF LOWER AMAZONIA 173

This fine series enables us to complete Hellmayr's recent critique of
the variations of this species in eastern South America (cf. Birds
Northeast Brazil, p. 394). A series of 36 topotypes from Surinam com-
pared with 8 from British Guiana show that the characters for nitidi-
cauda Elliot have no geographic constancy. Hellmayr proves to be
more or less right in his assumption that the Amazon is the boundary
between true fi '.mbriata and nigricauda, though all birds from the south
bank are intermediate. Thus our three birds from the north bank near
Obidos are typical fimbriata. The great series from the Tapajoz agrees
with nigricauda and differs obviously from true fimbriata in having
pure white under tailcoverts or at most a fine streak of grey at the base
of the shaft. All, however, agree with fimbriata rather than nigricauda
in the coloration of the central tail feathers. A specimen from Para
belongs here also, as might be expected, but. curiously enough four old
specimens, purporting to come from Pernambuco, belong here also.
They are trade skins with no data and of course may not have come
from the immediate vicinity of Pernambuco. We see no basis for
proposing an intermediate race. Hellmayr refers a bird from the Rio
Madeira to typical fimbriata.

296. Hylocharis cyanus viridiventris Berlepsch

Type locality: Venezuela, Merida
San Antonio and Santarem (Hellmayr)

2 cf , Rio Amazonas, Lago Cuipeuz near Obidos

1 cf , Itaituba (Carnegie Mus.)

1 9 , Santarem (do.)

297. Hylocharis sapphirina (Gmelin)

Type locality: Guiana

Para (Stone); San Antonio (Hellmayr); Pard, Ananindeua, Sta. Isabel, Marajo

Island, Rio Tocantins, Rio Tapajoz, Monte Alegre (Snethlage); Utinga

(Pinto)

3 cf 6 cf imm., 1 9 , Rio Tapajoz, left bank

1 cf 1 9 , Rio Amazonas, Lago Cuipeuz near Obidos

2 cf imm., Pard,, Bosque

1 cf 2 9 , Rio Tapajoz, Itaituba (Carnegie Mus.)

174 bulletin: museum of comparative zoology

298. Chlorestes notatus cyanogenys (Wied)

Type locality: eastern Brazil

Para (Hellmayr, La yard, Wallace, Snethlage, Stone); S. Antonio do Prata
(Hellmayr); Mexiana Island (Hagmann); abundant throughout (Sneth-
lage) ; Obidos (Hellmayr) ; Santarem (Chapman and Riker)
10 c? ad., 4 cf imm., 6 9 , Rio Tapajoz, east bank
9 d1 ad., Rio Amazonas, Lago Cuipeuz
10 d* imm., Para, Bosque
2 d\ Benevides (Carnegie Mus.)
1 d71 , Santarem (do.)
1 cT, Rio Tapajoz, Itaituba (do.)

This fine series differs very markedly and strikingly from 56 speci-
mens from Surinam, Trinidad and east Ecuador (including certain
trade skins, "Demerara", "Guiana", and "Brazil") in being glittering
bluish green on most of the underparts, the chin violet passing rapidly
to bluish green, only the lower edge of the abdomen glittering green or
golden green. This is very different from typical notatus, in which the
violet chin spot is sharply contrasted with the throat, breast and
abdomen, which are green or even golden green, rarely with a faint
bluish tinge on the upper throat and sides of neck.

There are unusual nomenclatural points involved. In 1913 Riley
described Chlorostilbon prasinus puruensis, which two years later
(Proc. Biol. Soc. Wash., 1915, p. 179) he showed should be called
Chlorestes caeruleus puruensis. The race puruensis differed supposedly
from a small series of skins from Lower Amazonia in having slightly
longer bills. This difference proves not to hold, as some birds from the
Guianas have bills up to 19 mm., and others from upper Amazonian
Ecuador have shorter bills than Riley's Rio Purus birds. With
puruensis transferred to Chlorestes, it must, however, be considered
with the recognition of an Amazonian race. It is preoccupied several
times over. The earliest name is cyanogenys Wied, renamed wiedi
Lesson, whose description emphasizes the bluish green throat. Agyr-
tria meliphila Pelzeln, Rio Negro, possibly belongs here, and Eucephala
subcaerulea discussed below, comes next in chronological order!

We have examined the unique Eucephala subcaerulea Elliott in the
American Museum. In this bird the violet chin passes into a deep
blue throat and chest, and the underparts do not become bluish green
until the center of the abdomen. The skin is of typical "Bahia" make,
and as perfectly ordinary notatus occurs commonly around Bahia, we
are convinced that subcaeruleus is a hyperchromatic aberration of
C. notatus cyanogenys Wied.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 175

The folly of basing ranges and systematic or racial characters upon
trade skins is well exemplified in the present connection. The Museum
of Comparative Zoology happens to have an ample series from au-
thentic localities to check against many trade-skins. Thus a
Trinidad series collected by Cherrie at Caparo cannot be separated
from those from Surinam in the Penard collection. But a series of
trade skins from "Trinidad" are more bluish green below than the
series from the south bank of the Amazon ; consequently they cannot
possibly have come from this island. Another long run of trade-skins
from "Brazil" is even more illuminating. "Bahia" trade skins are sup-
posed to have more golden green upperparts and the blue chin spot
more sharply defined. There is nothing in this, as some "Bahia" skins
before us do not show these characters, and other specimens authen-
tically from Surinam or Trinidad do.

In this connection we may briefly consider the almost unknown
C. hypocyanus (Gould), the type and one other specimen, both trade
skins of Rio or Bahia "make." Many years of exploration have failed
to produce either an authentic specimen or a range for this bird. In
Hummingbirds this strongly raises the presumption of hybrid origin.
Thanks to Gould's beautiful plate, Simon's and Hellmayr's critiques
(Novit. Zool., 1908, p. 11) and the utter confusion of generic lines in
this group of Hummers, we are able to suggest a possible explanation
of C. hypocyancus (Gould) . It will be found to be a perfect combination
of Hylocharis cyanus and Chlorestes notatus, both common and occur-
ring over large areas of Brazil together. The more golden green back
and coppery upper tailcoverts come from the first parent. The dull
crown comes from the second parent, and the glittering blue throat
and breast is an exact combination of the violet of Hylocharis cyanus
and the green or bluish green of Chlorestes notatus.

299. Chlorosttlbon prasinus prasinus (Lesson)

Type locality: "Brazil", in error, fide Hellmayr
Mexiana Island (Hellmayr) ; Maraca (Snethlage)
1 cf , Rio Amazonas, Lago Grande

Lack of material makes it impossible to do anything with this group,
the nomenclature and racial variations of which have been thrown
into hopeless confusion by Simon. We follow Hellmayr, however, in
applying prasinus to this species and not to the aureoventris-pucherani
group with deeply forked tails. In the concept maintained above,
jirasinus, subfurcatus Berlepsch, brevicaudatus Gould, daphne Gould,

176 bulletin: museum of comparative zoology

vitticeps Simon, and peruanus Gould are all undoubtedly forms of one
species, but how many of them are synonyms of each other remains to
be finally determined.

300. Thalurania furcata furcata (Gmelin)

Type locality: Cayenne
Obidos (Snethlage)

1 d\ Obidos (Carnegie Mus.)

301. Thalurania furcata balzani Simon

Type locality: Yungas, Bolivia

Rio Tapajoz, various localities on the left or west bank (Hellmayr and Sneth-
lage)

1 d\ Rio Tapajoz, Miritituba (Carnegie Mus.)
1 d1 1 9 , Rio Tapajoz, Villa Braga (do.)

Hellmayr has shown (1910, p. 376) that this form reaches the north-
ern limits of its range on the south side of the Amazon between the
Rio Madeira and the Rio Tapajoz. Further west it is replaced by
simoni Hellmayr along the Rio Solimoes. In central and southeastern
Brazil this group is represented by baeri Hellmayr and eriphile (Les-
son).

302. Thalurania furcata furcatoides Gould

Type locality: Para

Para (Layard, Wallace, Stone); Rio Inhangapy, Rio Guama (Stone); Souza,
Peixe-Boi, Igarape-Assu, S. Antonio do Prata, Mexiana Island (Hellmayr);
Marajo Island and numerous localities near Para (Snethlage); Rio
Tocantins (Snethlage); Utinga, Murutucu (Pinto)
6 cf 4 9 , Para, Bosque
8 d1 3 9 , Rio Tapajoz, various localities east bank.

1 cf1 1 9 , Benevides (Carnegie Mus.)

2 cf 1 9 , Santarem (do.)

The Para birds are topotypes of furcatoides Gould, as Hellmayr
showed years ago. Our series from the right bank of the Tapajoz is
inseparable from the Para series. Both series show conclusively that
the characters of intermedia Snethlage from the Rio Tocantins are
those of younger males, such specimens occurring both in our Para,
and Rio Tapajoz series. We have here a welter of species which are
really races. In the far west nigrofasciata (Gould), tschudii Gould, and
jclskii Taczanowski are three obvious representative forms, which are

GRISCOM AND GREENWAY : BIRDS OF LOWER AMAZONIA 177

connected with the whole furcata complex of eastern South America
by balzani Simon and simoni Hellmayr. The colombica-fannyi series
comes perilously close to the western extremes discussed above, and
wdtertoni (Bourcier) is little more than a remarkable long-tailed
extreme of the eastern group of subspecies.

303. Avocettula recurvirostris (Swainson)

Type locality: Cayenne

S. Antonio do Prata (Hellmayr); Monte Alegre, Rio Tocantins (Snethlage);
Rio Guama (Stone); Santarem (Pinto)

1 cT ad., 4 c? imm., 1 9 , Rio Amazonas. Lago Cuipeuz

304. Anthracothorax viridigula (Boddaert)

Type locality: Guiana

Para (Stone); S. Antonio do Prata (Hellmayr); Cunany, Monte Alegre, Para
(Snethlage); Mexiana Island (Wallace)

1 9 , Rio Amazonas, Boca do Igarape Piaba

2 cT 7 9 , Obidos (Carnegie Mus.)
1 c? ad. 1 c? imm., Santarem (do.)

Apparently all previous records from Brazil are from Para and
possibly Maranhao.

305. Anthracothorax nigricollis nigricollis (Vieillot)

Type locality: Brazil

Para (Layard); S. Antonio do Prata, Rio Acara (Hellmayr); Mexiana Island
(Hagmann and Wallace) ; Marajo Island, Itacuan, Rio Tocantins (Sneth-
lage)

5 d* 11 9 , Rio Tapajoz, various localities
1 cf 19, Rio Amazonas, Boca do Igarape Piaba
3 d" ad., 1 cf imm., Santarem (Carnegie Mus.)

Excellent series from all parts of the range of this species fail to show
any constant geographic variation. It is apparent from the data above,
that the two species occur together in the same locality. There is pre-
sumably some ecological or habit requirement separating them which
remains to be discovered.

306. Chrysolampis elatus (Linnseus)

Type locality: Cayenne

S. Antonio do Prata (Hellmayr) ; Para (Snethlage)

1 d1 imm., north bank of Amazon near Obidos

178 bulletin: museum of comparative zoology

307. PsiLOMYCTER THERESIAE THERESIAE (Da Silva)

Type locality: Para

Para (Cabanis and Heine); S. Antonio do Prata (Hellmayr); Sta. Isabel, Rio
Xingu, Rio Tapajoz (Snethlage)
7 cf 1 9 , Rio Tapajoz, Pinhy
1 d1 2 9 , Benevides (Carnegie Mus.)
1 cf , Rio Tapajoz, Villa Braga (do.)

This subspecies ranges up the Amazon to the Rio Madeira and
Manaos, replaced westward and northwestward by leucorrhous
(Sclater and Salvin).

308. POLYTMUS GUAINUMBI THAUMANTIAS (Lijmaeus)

Type locality: Sergipe, northeast Brazil
Marajo Island (Snethlage)

This hummer has a predilection for more open country and is
absent from the forested Amazon valley.

309. TOPAZA PELLA MICRORHYNCHA Butler

Type locality: Utinga, Para, Brazil

S. Antonio do Prata, Ipitinga (Hellmayr); Mocajatuba, Apehu, Rio Moju,
Rio Acara (Snethlage) ; Para, Castanhal, Rio Muraiteua (Stone)
1 9 , Para, Bosque

1 9, Benevides (Carnegie Mus.)

Now a very uncommon species, the race unknown outside the
localities recorded above.

310. Heliothrix auritus auritus (Gmelin)

Type locality: Cayenne

2 9 , Obidos (Carnegie Mus.)

311. Heliothrix auritus phainolaema Gould

Type locality: Para

Igarape-Assu, S. Antonio do Prata, Peixe-Boi (Hellmayr) ; Para, Providencia,
Rio Gurupy, Rio Tocantins, Rio Jamauchim (Snethlage); Rio Inhangapy
(Stone)

1 cf , Rio Tapajoz, Caxiricatuba
1 cf , Benevides (Carnegie Mus.)
1 9 , Santarem (do.)
1 9 , Rio Tapajoz, Aveiros (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 179

The subspecific variation of this species is relatively unusual, as
auriculatus (Nordmann) not only occurs in Brazil just south of our
area, but also in the Amazon valley on the Rio Madeira.

312. Anthoscenus longirostris longirostris

(Audebert and Vieillot)

Type locality: Trinidad

Rio Jamunda, Faro and Rio Tocantins (Snethlage) ; Santarem (Pinto)

313. Calliphlox amethystina (Gmelin)

Type locality: Cayenne

Para (Stone) ; S. Antonio do Prata (Hellmajo-) ; Rio Tocantins (Snethlage)
1 cf 1 9 , north bank of Amazon near Obidos

A species of the Guianas and Venezuela, apparently rare in our
area, again common southward.

314. Lophornis gouldii (Lesson)

Type locality: Para

S. Antonio do Prata (Hellmayr); Providencia, Braganca, Rio Guama, Rio
Tocantins (Snethlage); Utinga (Pinto)

1 9 , Para, Bosque

1 ? , Rio Acara, Buenos Aires

1 cf imm., Benevides (Carnegie Mus.)

Outside our area this species is definitely known from Maranhao
only. The indefinite record "Matto Grosso" by Pelzeln requires con-
firmation.

315. Discosura longicauda (Gmelin)

Type locality: Cayenne

S. Antonio do Prata (Hellmayr); Rio Tocantins, Cameta (Snethlage)

Family TROGONIDAE

316. Pharomachrus pavoninus viridiceps

Griscom and Greenway, 1937.

Type locality: Lower Amazon, Brazil
1 9 , Rio Tapajoz, Tauary

180 bulletin: museum of comparative zoology

Readily distinguishable from true pavoninus of far upper Amazonia.
The specimen listed above is quite the easternmost locality of record
for the species.

317. Trogon collaris Vieillot

Type locality: Cayenne

Rio Jary and Rio Jamauchim (Snethlage); "lower Amazons" (Wallace in Brit.
Mus.)

2 cf 2 9 , Rio Tapajoz, Miritituba and Villa Braga (Carnegie Mus.)

To be perfectly clear on vexed matters of nomenclature, this species
is the T. curucui Linnaeus of Hellmayr and Pinto.

318. Trogon curucui curucui Linnseus

Type locality: Cayenne
Obidos, Rio Jamunda, Faro (Snethlage)
1 c\ Obidos (Carnegie Mus.)

319. Trogon curucui sulphureus Spix

Type locality: Tabatinga, Rio Solimoes, Brazil

S. Antonio do Prata (Hellmayr) ; Castanhal (Stone) ; Rio Guama, Rio Tapajoz
(Snethlage)

3 cf 2 9 , Rio Tapajoz, various localities

1 d1 1 9 , Para, Bosque

2 d\ Benevides (Carnegie Mus.)

1 d" 2 9 , Santarem (do.)

2 d\ Rio Tapajoz (do.)

We adopt Ridgway's nomenclature for the vexed question as to the
proper name of this species, and we designate Cayenne as the type
locality. We also agree with Ridgway as to the racial variation in this
species. Hellmayr in his study of Spix's types (p. 596) has already
shown that the bird of southern Brazil should be known as chryso-
chlorus Pelzeln. While endorsing the markedly larger size of the
southern bird on the basis of specimens before us, we would also report
the narrower bars on the outer tail feathers, both black and white.
This results in a 50% increase in the total number of bars per tail
feather. In the same paper Hellmayr did not regard sulphureus Spix
as separable, apparently comparing males only. WTith a fair series
from both Surinam and the Lower Amazon, we agree in finding males
inseparable. But there is an excellent difference in the tail feather
pattern of females. In Amazonian females the solid black basal por-

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 181

tion of the inner web of the outermost rectrix is greatly extended, so
that there are 2-3 black bars between the solid black area and the
pure white tip. In typical curucui the solid black area is reduced, re-
sulting in 5-8 black bars before the white tip is reached. The same
relative difference exists on the 2nd and 3rd rectrices from the outer-
most. Finally we would remark that the female of chrysochlorus has
a similar tail pattern to sulphureus, which furnishes still another
distinctive character.

320. Trogon strigilatus strigilatus Linnaeus

Type locality: Cayenne

Rio Capim (Wallace and Goeldi); Rio Guama, Rio Inhangapy (Stone);
Ipitinga (Hellmayr); Santarem (Chapman and Riker); throughout
(Snethlage); Caviana Island (Brodkorb, as albiventer of Envier, on highly
dubious characters)

8 d71 4 9 , Rio Tapajoz, various localities

1 9 , Rio Acara, Acara

1 d\ Obidos (Carnegie Mus.)

1 d\ Santarem (do.)

1 cf1 1 9 , Rio Tapajoz, Villa Braga (do.)

We have compared this series with a very fine one from Surinam,
and find no essential difference between these birds, as well as others
from Trinidad, the interior of Venezuela and upper Amazonian
Ecuador. Birds from eastern Brazil are, however, an easily recognizable
subspecies. Females have greatly reduced wThite tips to the outer
rectrices and there is a marked difference in size; 9 from Surinam,
140-149 mm., from Bahia, 155-159 mm. Of the various names ap-
plied to this Trogon, all are clearly synonyms of true strigilatus except
melanopterus Swainson, which was for a brief period applied to the
species as a whole, before the Linnsean names were identified. This
name is based on a male collected by Swainson himself while in Brazil.
The description is a very brief one and is not subspecifically identifiable.
The only possible clue, the length of the wing, is entirely unsatisfactory,
as the measurement given is smaller than the smallest Surinam bird in
our series. Swainson's travels in Brazil were (1) around Pernambuco,
(2) around Bahia, and (3) Rio de Janeiro. It will be apparent, there-
fore, that the chances are very great that he had the larger east
Brazil race. We consequently designate Bahia as a more restricted
type locality, and the southern form will be known as T. strigilatus
melanopterus Swainson.

182 bulletin: museum of comparative zoology

321. Trogon variegatus variegatus Spix

Type locality: Brazil
Cajutuba (Natterer)

The only record for this "campo" form in our area.

322. Trogon variegatus bolivianus Grant

Type locality: Cosnipata, Peru
Rio Tapajoz, Goyana on west bank

1 9 , Rio Tapajoz, Apacy (Carnegie Mus.)

There is still some confusion in the races of variegatus. According to
Hellmayr, the present form ranges from northern Bolivia to eastern
Colombia, east to the Tapajoz. This far outlying station is connected
with the rest of the range only by records from Matto Grosso, which
Mrs. Naumburg refers to behni Gould, a much more austral form ac-
cording to Hellmayr. It is remarkable that the species should occur at
all in our area, and that the only two specimens come from opposite
ends of it, and belong to two different subspecies.

323. Trogon violaceus crissalis (Cabanis and Heine)

Type locality: Bahia, in error

Para (Goeldi, Brigham, Stone); Ipitinga (Hellmayr); Benevides, Rio Guama,
Rio Tocantins, Rio Xingu, Rio Tapajoz, Rio Jamauchim, Obidos (Sneth-
lage) ; Santarem (Chapman and Riker)

1 c? 1 9 , Rio Tapajoz, Santarem and Caxiricatuba
1 9 , Rio Acara, Acara

1 d1 imm., Santarem (Carnegie Mus.)

2 d", Rio Tapajoz, Apacy (do.)
1 d\ Benevides (do.)

The racial variation of this species in eastern Brazil still awaits
final determination, and the subspecific name used above is tentative
only. The very few records from northeastern Brazil have been called
ramonianus Deville and Desmurs, a name based on a bird from north-
east Peru, and which occurs along the base of the eastern Andes in
Ecuador. The birds before us agree with ramonianus, and differ from
the violaceus-caligatus series, in having the blackish, very minutely
freckled wing-coverts. They differ from true ramonianus of eastern
Ecuador in having smaller bills. This is one of the characters assigned
by Ridgway to crissalis Cabanis and Heine, based on a "Bahia" trade

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 183

skin. The character of the relative amount of freckling on the wing
coverts is reversed in our specimens, so it is apparently worthless.
Ridgway saw only one bird from Para, and assuming that the Bahia
bird would not be the same, proposed goeldii for the Para bird, should
it prove different! The probabilities are, however, that crissalis did
not come from Bahia, as there is no authentic record of the species
south of Amazonian drainage. The chances consequently are that the
type of crissalis will prove to represent lower Amazonian birds, and
that Ridgway's goeldii, proposed without any proper diagnosis, will be
reduced to the synonymy it so richly deserves.

324. Trogon melanurus melanurus Swainson

Type locality: Guiana

Para (Natterer, Wallace, Stone); Rio Guama (Stone); Peixe-Boi and S. An-
tonio do Prata (Hellmayr); Rio Iriri, Cussary, Monte Alegre, Obidos,
Rio Jamunda, Marajo Island (Snethlage); Santarem (Riker and Chap-
man)

3 d" 2 9 , Amazon river near Obidos, various localities
2 c? 1 9 , Rio Tapajoz, various localities

1 d\ Obidos (Carnegie Mus.)

2 9 , Santarem (do.)

5 cT, Rio Tapajoz, Villa Braga (do.)

Family ALCEDINIDAE

325. Ceryle torquata cyanea (Vieillot)

Type locality: Paraguay

3 9 , Rio Tapajoz

1 d" 4 9 , Amazon River near Obidos

1 cf, Santarem (Carnegie Mus.)

Numerous records throughout the region.

326. Chloroceryle amazona (Latham)

Type locality: Cayenne

2 cf 2 9 , Amazon River near Obidos
3 9 , Rio Tapajoz
1 o\ Obidos (Carnegie Mus.)
2 c? 2 9, Santarem (do.)

Numerous records throughout the region.

184 bulletin: museum of comparative zoology

327. Chloroceryle Americana Americana (Gmelin)

Type locality: Cayenne

1 9 , Amazon River near Obidos
3 c? 3 9, Rio Tapajoz

1 9 , near Para, Val-de-Caes

2 d1 2 9 , Santarem (Carnegie Mus.)

3 9, Rio Tapajoz, Apacy (do.)

Numerous records throughout the region.

328. Chloroceryle inda (Linnaeus)

Type locality: Cayenne

3 d1, Rio Tapajoz, Pinhy

1 9 , Benevides (Carnegie Mus.)

Numerous records throughout, including Caviana Island (Brod-
korb).

329. Chloroceryle aenea aenea (Pallas)

Type locality: Surinam

3 d" 2 9 , Amazon River near Obidos
5 d" 3 9 , Rio Tapajoz, various localities

1 9 , Obidos (Carnegie Mus.)

1 9 , Benevides (do.)
4(^2 9 , Rio Tapajoz (do.)

1 cf, Santarem (do.)

Numerous records throughout.

Family MOMOTIDAE

330. Baryphthengus martii martii (Spix)

Type locality: Para

Rio Tapajoz, Villa Braga (Snethlage)

2 cf, Rio Tapajoz, Tauary and Caxiricatuba
2 9 , Santarem (Carnegie Mus.)

2 cf 3 9 , Rio Tapajoz, various localities (do.)

The right bank of the Rio Tapajoz marks the eastern extension of
the range of this species in Brazil. It has been taken on the Rio Purus
and the Rio Madeira.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 185

331. Momotus momota momota (Linnaeus)

Type locality: Cayenne

Obidos (Hellmayr, Snethlage, Pinto) ; Monte Alegre and Rio Jamunda (Sneth-
lage)

5 cf 5 9 , Obidos (Carnegie Mus.)

The north bank of the Amazon marks the southern limit of the
typical form.

332. Momotus momota parensis Sharpe

Type locality: Paid

8 cf 4 9 , Pard and Rio Acara

2 c? 2 9 , Benevides (Carnegie Mus.)

Recorded by all collectors from Para to the right bank of the Rio
Tocantins, and extending southward into Maranhao and Piauhy. Its
absence from Mexiana and Marajo Islands should be particularly
noted. On the left bank of the Rio Tocantins we find cametensis
Snethlage, which will presumably range west to the right bank of the
Rio Xingti. From the Rio Tapajoz west to the Rio Purus, it is replaced
by simplex Chapman.

333. Momotus momota cametensis Snethlage

Type locality: Cameta, Rio Tocantins

Rio Tocantins, Cameta and Araumatheua (Snethlage)

3 cf 2 9 , Rio Tocantins, Cameta

These birds are topotypes of cametensis, of which Chapman had no
specimens when he reviewed the group (Bull. Amer. Mus. Nat. Hist.,
(48), 1923, p. 45). Three show perfectly the characters ascribed to the
race, and are, indeed, astonishingly distinct from parensis and simplex.
Two are indistinguishable from parensis, and arouse suspicion as to
whether some mislabelling has not occurred.

334. Momotus momota simplex Chapman

Type locality: Santarem

Santarem (Chapman and Riker, Hellmayr)

8 cf 4 9 , Rio Tapajoz, various localities

5 cf 5 9 , Rio Tapajoz, various localities (Carnegie Mus.)

5 d1 7 9 , Santarem (do., including type)

186 bulletin: museum of comparative zoology

335. Electron platyrhynchum orientale Todd, 1937

Type locality: Villa Braga, left bank, Rio Tapajoz, Brazil
1 c? 1 9, type locality (Carnegie Mus.)
At the time Mr. Todd and we were working on the Carnegie Museum
collections, all completely overlooked Miranda Ribeiro's description of
chlorophrys from the Rio Tocantins in Goyaz and Tramagin, Matto
Grosso. There is, of course, the possibility that the two subspecies
might be the same, but it is impossible to guess from the original
diagnosis of chlorophrys, the main characters used being true of any
immature Electron as compared with any adult!

Family GALBULTDAE

336. Urogalba dea dea (Linnpeus)

Type locality: Surinam

7 d* 3 9 , Obidos (Carnegie Mus.)

337. Urogalba dea amazonum Sclater

Type locality: upper Amazonia
1 9 , Para, Bosque

5 o" 1 9, Rio Tapajoz, various locality
9 cf 5 9 , Benevides (Carnegie Mus.)

1 cf , Santarem (do.)
12 cf 6 9 , Rio Tapajoz, right bank (do.)
1 cf 2 9 , Rio Tapajoz, left bank (do.)

Recorded throughout our area, except Mexiana and Marajo Islands.
These birds differ strikingly from a series of typical dea from Surinam
in just the respects ascribed to amazonum. We have, however, no upper
Amazonian material for comparison.

338. Galbula galbula Linnaeus

Type locality: Brazil

Obidos (Hellmayr and Snethlage); Rio Jamunda, Monte Alegre, Erere, Pai-
tuna, Rio Maecuru, Rio Jary, Arumanduba, Cunani (Snethlage); San-
tarem (Chapman and Riker) and Rio Tapajoa (Snethlage)
1 cf 3 9 , north bank of Amazon near Obidos
3 cf 3 9 , Obidos (Carnegie Mus.)

6 cf 7 9 , Rio Tapajoz (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 187

339. Galbula rufoviridis rufoviridis Cabanis

Type locality: Brazil

Rio Guama (Stone); Marajo Island, Rio Tocantins, Rio Tapajoz, Rio Jam-
auchim (Snethlage); Rio Tocantins (Wallace); Santarem (Chapman and
Riker); also Monte Alegre (Snethlage); Caviana Island (Brodkorb).
' 4 <? 2 9 , Para (Val-de-Caes)
13 d1 7 9 , Santarem (Carnegie Mus.)

It is quite surprising that there are no previous records of this
common Jacamar from the vicinity of Para. It ranges west to the
Rio Madeira.

Snethlage, alone, reports rufoviridis from Monte Alegre. One
wonders if she did not really have transitional or atypical specimens.

340. Galbula albirostris albirostris Latham

Type locality: Guyana

Obidos (Hellmayr and Snethlage)

9 cf 6 9 , Obidos (Carnegie Mus.)

Obidos is the southern limit of this species, which doubtless occurs
elsewhere on the north bank of the Amazon. West of the Rio Negro it
is represented by chalcocephala Deville.

341. Galbula cyanicollis Cassin

Type locality: Pard,

8o"5 9 , Rio Tapajoz, various localities
3 d" 1 9 , Para, Val-de-Caes
6 c? 5 9 , Santarem (Carnegie Mus.)
6 cf 4 9 , Rio Tapajoz (do.)

Recorded by all collectors throughout our area on the south side of
the Amazon, but absent on Mexiana and Marajo Islands. It ranges
westward to the Rio Jurua and Rio Pjrrus.

342. Galbula leucogaster leucogaster Vieillot

Type locality: Guiana

Rio Maecuru, Cachoeira (Snethlage)

This species has never been recorded from the south bank of the
Amazon, where it is represented by the recently described form below.

188 bulletin: museum of comparative zoology

343. Galbula leucogaster viridissima
Griscom & Greenway, 1937
Type locality: Rio Tapajoz, Pinhy
3 cf 1 9 , type locality
2 cf 1 9 , Rio Tapajoz, both banks (Carnegie Mus.)

As shown in the original description chalcothorax is merely another
representative form of this species.

344. Brachygalba lugubris lugubris (Swainson)

Type locality: Brazil

Monte Alegre, Rio Maecuru, Rio Acara (Snethlage); Rio Tocantins (Wallace
and Snethlage)

2 d" 2 9 , Para, Val-de-Caes

Until very recently a very little known bird in Brazil. So far as we
know adequate series from Guiana and Brazil have never been com-
pared.

345. Brachygalba lugubris melanosterna Sclater

Type locality: Goyaz
Rio Curua (Snethlage)

This record, just within our limits, marks the extreme northern
limit of the subspecies, which ranges from central Brazil southward.
In fact typical lugubris ranges south of the Rio Curua, further east-
ward in Parnahyba. Both forms are represented by fulviventris in
Amazonian Colombia and Ecuador. The species has yet to be col-
lected in the intervening area, but it is sufficiently rare so that the
negative evidence to date is by no means conclusive.

[Brachygalba albigularis (Spix) was described as from "sylvis ad
urbem Param". With the exception of the type the only specimens
come from the Rio Javari (Bates) and the Rio Purus (Snethlage).
There is no reason to suppose that ''Para" is really correct.]

346. Jacamerops aurea aurea (P.L.S. Miiller)

Type locality: Guiana

Peixe-Boi (Hellmayr) ; Rio Capim (Wallace); Rio Guama, Rio Acara, Rio
Tapajoz (Snethlage); Pataua (Pinto)

1 9 , Rio Tapajoz, Pinhy

4 cf 2 9 , Obidos (Carnegie Mus.)

1 cf , Benevides (do.)

1 cf 4 9 , Santarem (do.)

6 cf 1 9 , Rio Tapajoz, left bank (do.)

1 cf , Rio Tapajoz, Miritituba (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 189

There is no appreciable difference between Brazil and Guiana
specimens. As usual, birds from upper Amazonia are notably larger,
and the name isidori Deville based on a bird from Sarayacu, north-
eastern Peru is available, in spite of the fact that the type is a mela-
noid. Curiously enough a female from Mt. Duida is inseparable from
the type of penardi Bangs and Barbour from Costa Rica.

Family BUCCONIDAE

347. Bucco capensis Linnseus

Type locality: Guiana

Para (Wallace, Snethlage, Stone); Castanhal (Stone); Peixe-Boi (Hellmayr)

1 cf , Rio Tapajoz, Caxiricatuba

1 9 , Obidos (Carnegie Mus.)

1 d"2 9, Santarem (do.)

2 o" 1 9, Rio Tapajoz, Villa Braga, Apacy (do.)

348. NOTHARCUS MACRORHYNCHUS PARAENSIS SaSSl

Type locality: Para

Para (Layard, Natterer, Wallace, Snethlage, Stone); Cajutuba (Natterer);
Rio Capim (Goeldi); Rio Tocantins (Sassi)

3 d1 2 9, Santarem (Carnegie Mus.)

These birds have notably bigger bills than those from the left bank
of the Rio Tapajoz, agreeing with the measurements of Para birds.

349. NOTHARCUS MACRORHYNCHUS HYPERRHYNCHUS (Sclater)

Type locality: upper Amazonia

1 cf 1 9 , Rio Tapajoz, Itaituba (Carnegie Mus.)

Apparently the upper Amazonian representative of macrorhynchus
of Guiana, which reaches the Amazon only at Manaos, of which
giganteus Pelzeln from Marabitanas, Rio Negro, is presumably a
synonym.

350. Notharcus ordii Cassin

Type locality: Venezuela
Rio Cussary (fide Snethlage)

190 bulletin: museum of comparative zoology

351. Notharcus tectus tectus (Boddaert)

Type locality: Cayenne

Para (Layard, Natterer, Wallace, Stone); Igarape-Assu and San Antonio do
Prata (Hellmayr) ; Santarem (Chapman and Riker) ; Rio Guama, Marajo
Island, Monte Alegre, Obidos, Rio Jamunda, Rio Tocantins (Snethlage)
7 cf 2 9 , Obidos (Carnegie Mus.)
2 d* 2 9 , Benevides (do.)
6 d1 5 9 , Santarem (do.)
2 cf 1 9 , Rio Tapajoz, Apagy, Itaituba, Villa Braga (do.)

Birds from the south bank of the Amazon differ from Cayenne topo-
types in having minutely longer bills, which are also a little broader
at the base.

352. NOTHARCHUS MACRODACTYLUS MACRODACTYLUS (Spix)

Type locality: Fonte Boa, Rio Solimoes, by subs, desig.
1 cT, Obidos (Carnegie Mus.)

353. Nystactes tamatia tamatia Gmelin

Type locality: Cayenne

Amapa, Arumanduba, Monte Alegre, Erere, Rio Maecuru, Obidos (Snethlage)
1 cf 1 9 , Obidos (Carnegie Mus.)

354. Nystactes tamatia subsp.

1 cf 2 9 , Rio Tapajoz, left bank, Villa Braga

Surprisingly close to typical tamatia in coloration and consequently
sharply distinct from hypnaleus; resembling the latter, however, in the
shorter and slenderer bill, a marked character of that race in good
series.

Presumably Hellmayr's record of typical tamatia from the right bank
of the Rio Madeira belongs here also. A series from the Rio Purus has
the pale throat of pulmentum Sclater, but is not heavily spotted. Very
fine series from the Solimoes presumably represent yulmentum. There
is the possibility that these birds might represent interior Cherrie and
Reichenberger.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 191

355. Nystactes tamatia hypnaleus Cabanis and Heine

Type locality: Para-
Para (Natterer); Ipitinga and Rio Acara (Hellmayr) Rio Capim (Wallace);
Santarem (Chapman and Riker, Pinto); Rio Tocantins and Rio Tapajoz
(Snethlage)

1 cf 3 9 , Para, Val-de-Caes

2 9 , Rio Tapajoz, Tauary
12 cf 7 9 , Santarem (Carnegie Mus.)

The four Para birds are topotypes and are by no means as distinct in
color from a good series of typical tamatia, as Hellmayr's comments
would lead one to infer (cf. Novit. Zool., 1910, p. 391), as three have
the black apical breast spots minutely larger than the great majority
of our tamatia series. The 21 birds from the Rio Tapajoz are obviously
more heavily marked with black below and very distinct.

356. Nystalus maculatus maculatus (Gmelin)

Type locality: Brazil

Marajo Island and Rio Tapajoz (Snethlage) ; Santarem (Chapman and Riker) ;
Caviana Island (Brodkorb)

9 cf 9 9 , Rio Tapajoz, various localities

9 cf 7 9 , Santarem (Carnegie Mus.)

This species is really at the extreme northern limit of its range on the
south bank of the Amazon. It is quite variable, there being two other
races in Brazil.

357. Nystalus striolatus (Pelzeln)

Type locality: Engenho do Cap Gama

S. Antonio do Prata (Hellmayr and Snethlage); Rio Inhangapy (Stone); Rio
Guama (Stone, Snethlage)

At present this species has a curiously interrupted distribution, the
Para records being quite removed fronl the balance of the bird's known
range in upper Amazonia.

358. Malacoptila fusca (Gmelin)

Type locality: Cayenne
Obidos (Snethlage, Pinto)

6 cf 2 9 , Obidos (Carnegie Mus.)

192 bulletin: museum of comparative zoology

359. Malacoptila rufa brunnescens Zimmer

Type locality : Caxiricatuba, Rio Tapajoz

Santarem (Chapman and Riker) ; various localities from the Rio Madeira to
the right bank of the Rio Tapajoz (Zimmer)

2 cf 6 9 , Rio Tapajoz, various localities

2 cf 3 9 , Santarem (Carnegie Mus.)

10 cf 9 9 , Rio Tapajoz (do.)

360. Malacoptila rufa subspecies

Para (Natterer, Wallace); S. Antonio do Prata (Hellmayr); Rio Inhangapy
(Stone); Rio Capim (Goeldi); Peixe-Boi (Hellmayr; also numerous
localities near Para, the Rio Tocantins and the Rio Xingu (Snethlage)

For the latest comment on this species cf. Zimmer, Amer. Mus.
Novit., no. 500, 1931, pp. 3-7. He there advances reasons for supposing
that more than one form of the species may inhabit the region between
the Tapajoz and Para.

361. Monasa atra (Boddaert)

Type locality : Cayenne

Maraca, Cunany, Rio Jary, Obidos, Rio Jamunda (Snethlage); Obidos (Hell-
mayr)

1 cf 2 9 , near Obidos

362. Monasa morphceus morphceus (Hahn and Kiist)

Type locality: Brazil

1 9 , Para, Val-de-Caes

1 cf , Rio Acara, Acara

. 8 cf 6 9 , Rio Tapajoz, various localities

2 9 , Benevides (Carnegie Mus.)

11 cf 9 9, Santarem (do.)

6 cf 13 9 , Rio Tapajoz (do.)

Numerous records from every part of our area south of the Amazon,
except Mexiana and Marajo Islands.

We are quite unable to recognize rikeri Ridgway from Diamantina
and Santarem. Whether the malar apex is black or white is an individ-
ual variation in not only the series listed above, but in another modern
one from Bahia. The same character has been shown to break down in
certain alleged Panama species.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 193

363. MONASA NIGRIFRONS NIGRIFRONS (Spix)

Type locality: Rio Solimoes

3 cf 1 9 , Rio Amazonas, near Obidos
8 cf 1 9 , Rio Tapajoz, various localities
12 cf 7 9 , Santarem (Carnegie Mus.)
7 9 , Rio Tapajoz (do.)
2 cf 2 9 , Obidos (do.)

Numerous records throughout our area.

364. Nonnula rubecula simplex Todd, 1937

Type locality: Villa Braga, Rio Tapajoz

1 cf , the type, examined (Carnegie Mus.)

Replaced by cineracea Sclater on the Rio Madeira.

365. Chelidoptera tenebrosa tenebrosa (Pallas)

Type locality: Surinam

2 cf , Pard, Val-de-Caes
1 cf , Rio Acara, Acara

3 cf 2 9 , Rio Tapajoz, various localities
1 cf 1 9 , Benevides (Carnegie Mus.)

7 cf , Rio Tapajoz, Apacy (do.)

Numerous records throughout.

We agree with Hellmayr that birds from the south bank of the
Amazon are inseparable from a Surinam series and show no approach
to the radically larger and paler brasiliensis of eastern and southern
Brazil. Wing measurements of males of true tenebrosa are 101-108, of
brasiliensis, 114-116. Properly sexed modern series show that females
are decidedly larger than males, a Surinam series, 105-111 mm. There
is also a remarkable change in bill length and bill proportions with
maturity. Younger birds have much shorter and broader bills, and old
adults have not only longer bills, but a long subulate tip or terminal
half, entirely lacking in younger stages. Two males from the upper
Amazon in northeastern Peru and eastern Ecuador are tenebrosa in
color, but are as usual larger, 108-113 mm. Should this size difference
prove constant in series, the bird of upper Amazonia should be de-
scribed. There is one very remarkable specimen in the museum collec-
tion from Alagoas, Maceio, on the coast 60 miles south of Pernambuco,
collected by Newton Dexter. This bird is unsexed, but the wing
measures 123 mm. Even assuming that it must be a female, it is gi-
gantic, and strongly indicates the need of further material from this
neglected corner of Brazil.

194 bulletin: museum of comparative zoology

Family CAPITONIDAE

366. Capito brunneipectus Chapman

Type locality: Villa Braga, left bank of Rio Tapajoz

1 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

First known from four specimens collected by Snethlage at the type
locality, and sent to the American Museum for determination (cf.
Chapman, Amer. Mus. Novit., no. 2, 1921, p. 1). It marks the eastern-
most point reached by the genus. This species and C. dayi Cherrie of
the Rio Madeira represent the upper Amazonian C. auratus Dumont,
a variable species recently monographed by Chapman (Amer. Mus.
Novit. no. 335). In particular the relationships of C. dayi to birds from
the Rio Madeira, which Hellmayr (1910, p. 395) reported provisionally
as C. auratus intermedins Berlepsch and Hartert, and which are pos-
sibly A. auratus insperatus Cherrie, should be investigated.

367. Capito nicer (P.L.S. Midler)

Type locality: Cayenne

Obidos and Rio Jamunda, Faro (Snethlage)

Family RAMPHASTIDAE

368. Ramphastos toco toco (P.L.S. Miiller)

Type locality: Cayenne

Mexiana Island (Wallace, Hagmann); Marajo Island and Monte Alegre
(Snethlage); Pataua (Pinto)

2^2 9, Para, Val-de-Caes

3 9 , Santarem (Carnegie Mus.)

Lower Amazon birds do not differ from a Cayenne series in the
Carnegie Museum. In life toco has the bill uniform orange except for
the black areas, the ridge of both mandible and maxilla crimson. In
dried museum skins the bill fades to a uniform dull yellow, the ridges
crimson or orange. Two specimens recently killed on the Rio Iguassu,
Paraguay, are remarkable in having the maxilla largely crimson in-
stead of orange, orange only for a narrow black band just forward of
the black base; the crimson color increases in intensity and depth to-
wards the tip of the bill. Birds from Sao Paulo are strikingly distinct
in having the pure white throat untinged with yellow and constitute a

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 195

recognizable race, albogularis Cassin. Some birds from Sao Paulo and
others from Goyaz are identical with Paraguayan birds in coloration
of the bill, while the Paraguayan birds approach albogularis in other
respects. Specimens from Bolivia in the Carnegie Museum do not
show the bill characters of the Paraguayan birds and are inseparable
from toco.

309. Ramphastos tucanus tucanus Linnaeus

Type locality: Cayenne

Amapd and Rio Jamunda1 (Snethlage); Para (Natterer, Stone, Wallace);
Igarape-assu, Peixe-Boi, and Ipitinga (Hellmayr); Rio Capim (Goeldi);
Providencia and Sta. Isabel (Snethlage). Birds from the Rio Xingu
(Zimmer), and the Rio Tocantins (Snethlage) approach the next race.
Obidos, Murutucu, Par£ region (Pinto)
1 9 , Rio Acard, Acard
1 d\ Obidos (Carnegie Mus.)

370. Ramphastos tucanus oblitus Griscom & Greenway, 1937

Type Locality: Rio Tapajoz, Tauary
Santarem (Pinto)

1 cf (the type, M. C. Z.)

3 c? 1 9 , Santarem (Carnegie Mus.)

This well marked intermediate form will presumably range at least
as far east as the left bank of the Rio Xingu. It must be remembered
that Para specimens, while referable to typical tucanus, approach this
race in having deeper colored under tailcoverts. The distinctness of
oblitus is best evident when Cayenne topotypes of tucanus are used in
comparison.

371. Ramphastos tucanus cuvieri Wagler

Type locality: Borba, Rio Madeira, by subsequent designation
Itaituba, left bank of Rio Tapajoz (Snethlage)
1 9 hum., Boim, Rio Tapajoz.

372. Ramphastos vitellinus vitellinus Lichtenstein

Type locality: Cayenne

Cunany, Obidos, Rio Jamunda1 (Snethlage) ; Obidos (Pinto)
1 9 , Obidos (Carnegie Mus.)

19G bulletin: museum of comparative zoology

373. Ramphastos vitellinus ariel Vigors

Type locality: Rio de Janeiro

Numerous records from south side of the Amazon, except the islands
2 d1 1 9 , Rio Acara, Acara

1 cf 2 9,1? Rio Tapajoz various localities, both banks.

2 cf 3 9 , Benevides (Carnegie Mus.)
1 cf , Santarem (do.)

1 cf 2 9 , Rio Tapajoz (do.)

This toucan is clearly a mere subspecies of the Guiana vitellinus,
which it represents on the south side of the Amazon southward to
eastern Brazil. It is specifically distinct from culminatus in having the
ridge of the culmen black, concolor with the side of the bill, and occurs
together with the culminatus-osculans group and the tucanus group in
one or another part of their ranges. While we appreciate Zimmer's
comments on the variations in the color of throat and upper tail
coverts, we feel that the two facts mentioned above should have weight,
until evidence to the contrary materializes.

374. Pteroglossus aracari aracari (Linnaeus)

Type locality: Northeastern Brazil, ex Marcgrave

Numerous records from Marajo Island and Para to the right bank of the

Madeira

1 9 , Rio Acara, Acara

2 cf 1 9 , Santarem (Carnegie Mus.)

2 cf 2 9 , Rio Tapajoz, left bank (do.)

3 cf 4 9 , do. , right bank (do.)

A fair series from Bahia and Rio de Janeiro differ only in averaging
minutely paler yellow below. In Sao Paulo we find the very distinct
vergens Griscom and Greenway, which in color characters approaches
castanotis australis Cassin so remarkably. Were it not for the fact that
these two birds occur together at Valparaiso, Sao Paulo, we should
have been inclined to think that castanotis, mariae, azarae, and aracari
were conspeeific.

375. Pteroglossus aracari atricollis (P.L.S. Miiller)

Type locality: Brazil in error; Cayenne by Bangs and Penard, 1918
Maraca, Monte Alegre, Obidos (Snethlage)

9 cf 2 9 , north bank of Amazon near Obidos

2o13 9 , Obidos (Carnegie Mus.)

A fine series of Cayenne topotypes in the Carnegie Museum shows
that birds from the north bank of the Amazon are atricollis. This race

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 197

differs from typical aracari in having a broad black culminal stripe,
and the breast is sulphur to orange yellow instead of lemon yellow.

We have here a further illustration of the rashness of assuming that
the "Guianas" (all three colonies) are necessarily a homogeneous unit.
Eight specimens from Paramaribo, Surinam, before us are immediately
separable from atricollis in having lemon yellow breasts as in typical
aracari, and the thighs more olive, less ochraceous brown. Two birds
from eastern Venezuela agree with these Surinam birds. There is every
possibility that this subspecies, by pure chance, will have to be known
as roraimae Brabourne Chubb, a name originally proposed on purely
nomenclatural but invalid grounds!

There would appear to be another misunderstanding of nomencla-
ture. Pinto (1938, p. 329, footnote) has claimed that wiedii Sturm,
1847, is an earlier name for vergens Griscom and Greenway. In the
first place, birds from Wied's collecting area, from Bahia south to
Rio de Janeiro are typical aracari; hence Wied cannot have collected
vergens a southern race from Sao Paulo. In the second place, the
plate and description of wiedii Sturm is an excellent characterization
of true aracari. Gould (Monog. Toucans, 2nd ed., 1854) took up
%viedii Sturm. His plate and description show clearly that the differ-
ences between aracari and wiedii are exactly the differences between
Guiana and Brazil specimens of aracari. In other words, Gould and
Sturm thought of aracari as a Guiana bird, now aracari atricollis, and
wiedii, the east Brazil bird, is a synonym of true aracari, originally
based on a Marcgrave specimen ex northeast Brazil. There are two
races of this species in Brazil.

376. Pteroglossus bitorquatus bitorquatus Vigors

Type locality : none given : we suggest Para

Para (Natterer and Wallace); S. Antonio do Prata and Ipitinga (Hellmayr);

Rio Capim (Goeldi); Ourem (W. A. Schulz); Providencia and Benevides

(Snethlage); Utinga (Pinto)
1 9 , Rio Acara, Acara

377. Pteroglossus bitorquatus reichenowi Snethlage

Type locality: Monte Alegre, north bank of Amazon, apparently in error; this

locality questioned in Aves do Amazonicas
Rio Tocantins, Cameta and Rio Jamauchim, Sta. Helena (Snethlage); San-

tarem (Chapman and Riker); Santarem (Pinto)
9 d" 8 9 , Santarem (Carnegie Mus.)

198 bulletin: museum of comparative zoology

An obvious representative of bitorquatus, differing only in the absence
of the narrow yellow pectoral band. A specimen in the M.C.Z. with no
original label is labelled "Marajo Island," a locality which cannot pos-
sibly be correct, as the genus is unknown on the open campos of this
island.

378. Pteroglossus bitorquatus sturmi Natterer

Type locality: Borba, Rio Madeira

1 9 , Rio Tapajoz, Miritituba (Carnegie Mus.)

1 <? 1 9 , (do.) , Apacy (do.)

2 d" 1 9 , (do.) , Villa Braga (do.)

379. Pteroglossus viridis inscriptus Swainson

Type locality: Guiana in error

Para (Layard, Natterer, Stone, Wallace) ; Ipitinga (Hellmayr) ; Rio Inhangapy
(Stone); Santarem (Chapman and Riker, Hellmayr); Providencia,
Benevides, St. Antonio do Prata, Rio Guamd, Rio Moju; Rio Tocantins,
Rio Tapajoz (Snethlage)

2 d" 2 9 , Benevides Carnegie Mus.)

8 cf 5 9 , Santarem (do.)

1 d\ Rio Tapajoz, Villa Braga (do.)

Replaced by P. humboldti Wagler from the Rio Madeira westward.

380. Pteroglossus viridis viridis (Linnaeus)

Type locality : Cayenne

Maracd,, Obidos, Rio Jamundd (Snethlage); Obidos (Pinto).
6o",2 9 , Obidos (Carnegie Mus.)

The southern limit of this species is apparently the north bank of the
Amazon. It is replaced by didymus Sclater in far upper Amazonia.
This race and humboldti bridge the gap between viridis and inscriptus.

381. Selenidera maculirostris gouldi (Natterer)

Type locality: Para

Para (Layard, Natterer, Stone, Wallace); Igarape-Assu and Ipitinga (Hell-
mayr); Rio Capim (Goeldi); Providencia, Rio Moju (Snethlage).
1 cf 1 9 , Pard, Val-de-Caes and Bosque
1 9 , Rio Acara, Acara

Birds from the Rio Tocantins (Snethlage) may be this form or the
next.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 199

382. Selenidera maculirostris hellmayri Griscom & Greenway,

1937

Type locality: Rio Tapajoz, Boim
Santarem, Itaituba (Pinto)

1 cf 1 9 , Rio Tapajoz, Boim and Tauary.
3 9 , Santarem (Carnegie Mus.)

2cf2 9, Rio Tapajoz, left bank (do.)

383. Selenidera culik (Wagler)

Type locality: Cayenne

Obidos (Hellmayr), Obidos (Pinto)

6 cf 2 9 , Obidos (Carnegie Mus.)

Family PICIDAE

384. Colaptes campestris chrysosternus (Swainson)

Type locality: the Sertao of Bahia
Monte Alegre (Snethlage)

Birds from this isolated colony on the north bank of the Amazon
should be compared with typical material from much further south.

385. Piculus chrysochloros paraensis (Snethlage)

Type locality: Para

Murutucu (Hagmann, Snethlage); Para (Snethlage)
1 9 , Rio Tapajoz, Caxiricatuba

1 9 , Santarem (Carnegie Mus.)

Recorded west to the right bank of the Rio Tapajoz, but very few
specimens exist. The closely allied race capistratus (Malherbe) has
been found at Manaos, and should be sought on the north bank of the
Amazon in our area.

386. Piculus chrysochloros hypochryseus Todd, 1937

Type locality: Arima, Rio Purus, Brazil

2 cf 1 9 , Rio Tapajoz, Villa Braga and Miritituba (Carnegie Mus.)

387. Piculus flavigula flavigula (Boddaert)

Type Locality: Cayenne

Rio Jary, Rio Jamunda (Snethlage); Obidos (Hellmayr)

3 cf, Obidos (Carnegie Mus.)

200 bulletin: museum of comparative zoology

388. Piculus flavigula magnus (Cherrie and Reichenberger)

Type locality: Monte Christo, Matto Grosso

Numerous records throughout the area south of the Amazon by all collectors;
absent on Mexiana and Marajo Islands

1 9 , Para, Bosque

1 c? 2 9 , Rio Tapajoz, Tauary and Caxiricatuba

1 o\ Benevides (Carnegie Mus.)

2 <? 1 9 , Santarem (do.)

2 d\ 1 9 , Rio Tapajoz, Villa Braga (do.)

A puzzling series of intermediates, which we have compared with a
good series of true flavigula from Guiana and the type, and two other
recorded specimens of magnus in the American Museum of Natural
History. One Para bird is inseparable from true flavigula. The Rio
Tapajoz birds agree perfectly in longer wing with magnus, but have
slightly smaller and weaker bills. All birds seen from south of the
Amazon differ from true flavigula in having the olive green of the
underparts a slightly paler shade.

389. Chrysoptilus melanochloros mariae Hargitt

Type locality: Chamicuros, east Peru in error; actually Marajo Island
Marajo Island (Hagmann and Snethlage)

This race of the very variable malanoehloros of eastern and southern
Brazil is otherwise known only from a small island off the coast of
Maranhao.

390. Chrysoptilus punctigula punctigula (Boddaert)

Type locality: Cayenne

Amapa, Monte Alegre, Paituna, Rio Jamunda (Snethlage); Lago Cuipeua
(Pinto)

3 cf 2 9, Obidos (Carnegie Mus.)

The Obidos birds agree perfectly with a fine series from Cayenne and
Surinam.

391. Chrysoptilus punctigula pallidior
Griseom & Greenway, 1937

Type locality: Amazon River, Lago Grande

Santarem (Chapman and Riker, Pinto) ; Lago Grande (Pinto)

1 d1 5 9 , Lago Grande

7 <? 7 9 , Santarem (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 201

The few records of this species from our area have always been called
guttatus (Spix), a subspecies which is still credited with too extensive a
range for so variable a bird. Hellmayr has shown that Spix's type
agrees with a Natterer specimen from Manaos. This subspecies, un-
known to us from topotypes, differs from punctigula only in those re-
spects in which pallidior and punctigula are alike.

392. Leuconerpes candidus (Otto)

Type locality: Cayenne errore

Mexiana Island (Hagmann, Snethlage) ; Marajo Island, Monte Alegre, Paituna
(Snethlage) ; Lago Cuipeua (Pinto)

3 cf 4 9 , near Obidos

1 cf 1 9 , Obidos (Carnegie Mus.)

3 d" 3 9, Santarem (do.)

393. Tripsurus cruentatus cruentatus (Boddaert)

Type locality: Cayenne

Para region (numerous records); Rio Tocantins (Pinto); Santarem and Rio
Tapajoz (all collectors)

1 cf 1 9 , Rio Tapajoz, Tauary

3 d\ Benevides (Carnegie Mus.)
1 <?, Santarem (do.)

394. Tripsurus cruentatus extensus Todd, 1937

Type locality: Arima, Rio Purus

3o,2 9, Rio Tapajoz, left bank, Villa Braga, Apagy, Miritituba,
(Carnegie Mus.)

Birds from the Rio Tapajoz are intermediates between these two
subspecies, but those from the left bank are on the whole nearer

extensus.

395. Tripsurus rubRifrons (Spix)

Type locality: Para,

Para region, numerous records (all collectors)
1 cf, Benevides (Carnegie Mus.)

Mr. Todd (1937, pp. 250-251) has recently commented on the
curious relationships between cruentatus and rubrifrons. The former
of these supposed species is common and wide ranging; the latter is

202 bulletin: museum of comparative zoology

restricted to the Guianas and the Para region. In the Guianas, rubri-
frons is definitely commoner than cruentatus. As regards external
differences, rubrifrons might be described as a cruentatus which has lost
the white postocular stripe and the yellow nuchal collar. Now the
collections in the Carnegie Museum show clearly various types of
intermediates. Some have no nuchal collar, others have no super-
ciliary or it is incomplete. We are obviously dealing with hybrids be-
tween two species, or rubrifrons is a color phase of cruentatus, localized
in a portion only of the latter's range. Mr. Todd inclines to the
hybrid theory largely on the ground that the characters of cruentatus
are so constant over the greater part of its extensive range. Our own
idea is that rubrifrons is an imaginary species, a mere color phase of
cruentatus, which is losing its white postocular and yellow nuchal collar
in the northeastern portion of its range. The chief arguments against
the hybrid theory are that there is no region where only rubrifrons
occurs, and its characters are purely negative.

396. Veniliornis passerinus passerinus (Linnaeus)

Type locality: Cayenne

Cussary, Amapa, Monte Alegre, Rio Maecuru, Obidos (Snethlage)
1 d\ Obidos (Carnegie Mus.)

397. Veniliornis passerinus subspecies

Mexiana and Marajo Islands (Snethlage, Hellmayr)
1 9 , Santarem
5 cf 2 9 , do. (Carnegie Mus.)

Apparently no adequate series from the south bank of the Amazon
has ever been reported on. Our birds are quite different from a great
series from Cayenne in the Carnegie Museum in being a more golden
brown, both above and below. They appear to be intermediate be-
tween passerinus and taenionotus (Reichenbach) in size and color.
Our material of this latter form and transfluvialis Hellmayr is inade-
quate ; both occur in Piauhy and the latter in Maranhao, very close to
our area, where Marajo Island birds might belong. Snethlage records
taenionotus and passerinus as distinct species and occurring together in
the same place. The characters formerly used to separate these
alleged "species" were, however, chiefly matters of individual varia-
tion.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 203

398. Veniliornis affinis cassini (Malherbe)

Type locality: Cayenne, by subsequent designation
Rio Jary, Monte Alegre, Obidos, Rio Jamunda (Snethlage)
2 cf 4 9 , Obidos (Carnegie Mus.)

399. Veniliornis affinis ruficeps (Spix)

Type locality: Amazon River

Para (Layard, Natterer, Stone); Peixe-Boi and Ipitinga (Hellmayr); Rio Mur-
aiteua (Stone); Rio Capim (Goeldi); Santarem (Chapman and Riker,
Hellmayr); Marajo Island, Rio Tocantins, Rio Iriri and Rio Tapajoz
(Snethlage)

13 cf 7 9 , Rio Tapajoz, various localities east bank

2 cf 1 9 , Benevides (Carnegie Mus.)
7 d1 6 9 , Santarem (do.)

3 o71 1 9 , Rio Tapajoz, both banks (do.)

Typical affinis occurs in southern and eastern Brazil. On the Rio
Madeira westward, we encounter hacmatostigma (Malherbe), which is
a composite of several subspecies.

400. Celeus flavescens ochraceus (Spix)

Type locality: Amazon River

Santarem (Chapman and Riker) ; Cussary, Monte Alegre (Snethlage) ; Obidos
(Hellmayr); Marajo Island (Pinto)

2 d" 1 9 , near Obidos

5 d" 1 9 , Rio Tapajoz, various localities

3 c? 3 9 , Obidos (Carnegie Mus.)
8 d1 5 9 , Santarem (do.)

401. Celeus elegans elegans (P.L.S. Miiller)

Type locality: Cayenne

Obidos (Hellmayr); Cunany, Rio Jamunda (Snethlage); Obidos (Pinto).
2 d" 5 9 , Obidos (Carnegie Mus.)

204 bulletin: museum of comparative zoology

402. Celeus jumana jumana (Spix)

Type locality: in sylvis Amazonum

Para (Layard, Wallace, Stone, Natterer); Muria (Natterer); Rio Capim
(Goeldi); Utinga, Ipitinga, Igarape-assu (Hellmayr); S.Antonio do Prata,
Sta. Isabel, Rio Tocantins,Rio Tapajoz (Snethlage) ; Santarem (Hellmayr)
3 9 , Para, Val-de-Caes

2 J2 9, Rio Tapajoz, various localities

1 cf , Benevides (Carnegie Mus.)
5 cf 2 9 , Santarem (do.)

3 o"3 9 , Rio Tapajoz, Apacy (do.)

403. Celeus undatus multifasciatus (Malherbe)

Type locality: Brazil

Para (Natterer, Stone) ; Ipitinga (Hellmayr) ; Ananindeua, Maguary, S. An-
tonio do Prata, Rio Tocantins (Snethlage)

2 cf 2 9 , Benevides (Carnegie Mus.)

404. Celeus grammicus subcervinus Todd, 1937

Type locality: Rio Tapajoz, Villa Braga

1 o71 1 9 , Santarem (Carnegie Mus.)

2 cf 2 9 , Rio Tapajoz, Villa Braga (do.)

This species is recorded from Cussary (Snethlage), but there is no
telling what subspecies it may prove to be ; possibly typical grammicus.

405. Crocomorphus flavus flavus (P.L.S. Muller)

Type locality: Cayenne

Amapd, Rio Jamunda, Faro (Snethlage)

406. Crocomorphus flavus inornatus Cherrie

Type locality: Santarem, Rio Tapajoz

Para (Natterer, Stone); numerous records near city (Snethlage); Marajo

Island (Snethlage); Cussary (Snethlage); Santarem (Chapman and

Riker)

1 9 , Para, Val-de-Caes

3 cf 4 9 , Rio Tapajoz, various localities.

2 d\ Benevides (Carnegie Mus.)

4 cf1 5 9 , Santarem (do.)

2 cf 2 9, Rio Tapajoz, Villa Braga (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 205

Our series shows that this race is phenomenally variable. Six speci-
mens only have no rufous on the outer webs of the secondaries and
wing-coverts. The balance have more or less, and are indistinguishable
from Cayenne topotypes. Two only are devoid of the extensive white
spotting on the upper wing. Several are inseparable from true flavus
in this respect, and all the remainder have some.

407. Cerchneipicus torquatus (Boddaert)

Type locality: Cayenne

1 d\ Obidos (Carnegie Mus.)

408. Cerchneipicus tinnunculus angustus
Griscom & Greenway, 1937

Type locality: Rio Tapajoz, Caxiricatuba

Rio Moju, Cussary, Rio Tapajoz, Boim (Snethlage)

1 o71, Rio Tapajoz, Caxiricatuba (M.C.Z.)

3 cT 1 9 , Santarem (Carnegie Mus.)

409. Scapaneus rubricollis rubricollis (Boddaert)

Type locality: Cayenne

1 <? 2 9 , Obidos (Carnegie Mus.)

410. Scapaneus rubricollis trachelopyrus (Malherbe)

Type locality: Peru

Numerous records throughout our area on the south side of the Amazon.

2 9 , Para, Val-de-Caes

10 cf 1 9 , Rio Tapajoz, various localities
1 9 , Benevides (Carnegie Mus.)
1 cf , Santarem (do.)

In default of topo typical Peruvian material, we can only follow
Hellmayr in referring our series here, on the ground that the outer
webs of the primaries are extensively rufous. Brazilian birds should
prove separable, as they are apparently smaller than the two Peruvian
birds Hellmayr measured.

206 bulletin: museum of comparative zoology

411. SCAPANEUS MELANOLEUCOS MELANOLEUCOS (Gmelin)

Type locality: Surinam

Pard (Natterer); Mexiana Island (Hagmann); Marajo Island and north bank
of Amazon (Snethlage) ; Santarem (Chapman and Riker) ; Caviana Island
' (Brodkorb)

2 9 , north bank of Amazon near Obidos

2 cf, Rio Tapajoz, Santarem and Pinhy
1 d71 , Santarem (Carnegie Mus.)

1 d1, Rio Tapajoz, Villa Braga (do.)

412. Ceophloeus lineatus lineatus (Linnaeus)

Type locality: Cayenne

Recorded by all collectors throughout the area

1 9 , Para\ Val-de-Caes

1 d" 3 9, Rio Tapajoz, east bank

1 cT, Obidos (Carnegie Mus.)

413. PlCUMNUS CIRRHATUS MACCONNELLI Sharpe

Type locality: British Guiana

Marajo Island (Hellmayr); Santarem (Chapman and Riker); Arumanduba,
Monte Alegre, Rio Tocantins (Snethlage)

3 o1 5 9 , north bank of Amazon River near Obidos.

1 cf, Rio Tapajoz, Santarem
1 9 , Obidos (Carnegie Mus.)
13 d" 16 9 , Santarem (do.)

These birds are very different from true cirrhatus, of which we have
an excellent series, in just the respects Hellmayr says characterize
macconnelli, which he records from Para (cf. Novit. Zool., 1913,
p. 349). P. amazonicus Snethlage 1906 {nee buffoni amazonicus Sneth-
lage, 1914) is a synonym.

414. Picumnus varze^e Snethlage

Type locality: Obidos

Obidos and Rio Jamunda, Faro (Snethlage)

24 c? 27 9 , Obidos (Carnegie Mus.)

A very distinct species, put next to cirrhatus by Madame Snethlage.
It has nothing to do with this species, but the key and description in
the Cat. Aves Amazonicas are so poor that no one would imagine how
distinct it is.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 207

415. Picumnus guttifer pallidus Snethlage

Type locality: Quatipuru, northeast of Para

Only known from the type locality and obviously connecting guttifer
and spilogastcr.

416. Picumnus exilis buffoni Lafresnaye

Picumnus buffoni meridionalis Domaniewski: New name for P. b. amazonicus
Snethlage, 1914, preoccupied by P. amazonicus Snethlage, 1906. (cf.
Ann. Zool. Mus. Polon., 4, 1925, p. 296).

Rio Jary and Obidos (Snethlage)

2 cf 2 9 , Obidos (Carnegie Mus.)

Inseparable from true buffoni of Cayenne, of which we have seen a
great series of topotypes.

417. Picumnus aurifrons transfasciatus Hellmayr & Gyldenstolpe

Type locality: Marai, east bank of Rio Tapajoz. (cf. Arkiv Zool., 29 B,

no. 6, Jan. 1937, p. 1)
Rio Curua and Rio Tacantins (Snethlage)

16 cf 9 9 , Rio Tapajoz; various localities east bank.

1 cf , Santarem (Carnegie Mus.)
3 cf 1 9 , Rio Tapajoz. (do.)

In this recent paper, typical aurifrons is restricted to the Rio
Madeira.

418. Picumnus borbae Pelzeln

Type locality: Borba, Rio Madeira, Brazil

Santarem (Hellmayr); Rio Jamauchim, Rio Tapajoz, Boim, Pinhel, Villa
Braga (Snethlage)

7 cf 2 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

3 cf, (do.), Itaituba (do.)
2 cf 2 9 , (do.), Apacy - (do.)

This species is only known from the localities cited above, and its
range is consequently included in that of aurifrons. It differs from that
species only in that the crown feathers of the male are tipped with red
instead of yellow.

In the Carnegie Museum series, all male aurifrons have faintly
barred backs, and all male borbae have uniform olive backs. We con-
sequently refer females with barred backs to aurifrons, and those with

208 bulletin: museum of comparative zoology

immaculate backs to borbae. We are reinforced in this theory by the
fact that two obviously immature birds, one of which is unquestion-
ably borbae, because the red crown, spots are coming in, have the backs
uniform.

Family DENDROCOLAPTIDAE

419. Dendrocolaptes certhia certhia (Boddaert)

Type locality: Cayenne

Rio Jary, Obidos (Snethlage) ; Rio Jamunda (Snethlage and Zimmer)
7 cf 4 9 , Obidos (Carnegie Museum)

420. Dendrocolaptes certhia concolor Pelzeln

Type locality: Villa Bella de Matto Grosso; Borba, Rio Madeira (Hellmayr, by
subsequent designation)
Various localities, west bank of Rio Tapajoz (Snethlage, Zimmer).
1 c? 1 ?, Rio Tapajoz, east bank (approaching medius)

5 cf 3 9 , Rio Tapajoz, west bank (Carnegie Mus.)

421. Dendrocolaptes certhia medius Todd

Type locality: Benevides, Para

Para (Wallace, Stone); Igarape-Assu, S. Antonio do Prata, Peixe-Boi (Hell-
mayr); Magoary (Steere); Providencia, Apehii, Rio Guama (Snethlage)

1 d", Rio Acara, Acara

1 cf 1 9 , Benevides (Carnegie Mus.)

6 C? 6 9 , Santarem (do.)

1 d1 , Rio Tapajoz, right bank (do.)

This species is well known from every part of the south bank of the
Amazon in our area. Zimmer has written at length of the remarkably
inconsistent manner in which intergradation takes place; (cf. Amer.
Mus. Novit., no 753, p. 2). Birds from the east bank of the Tapajoz
to the Rio Tocantins are three different types. Some as far east as the
Tocantins are concolor; a very few birds from the Tocantins are medius;
all others are variously intermediate.

The series in the Carnegie Museum supports Zimmer's comments
on the variability of this species. On the other hand, it does not sup-
port Hellmayr's reduction to synonymy of his own ridgwayi from
Santarem. Birds from Para to the Rio Acara are homogeneous, and
are less different from true certhia than is the Santarem series from

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 209

medius. Concolor differs from a Santarem series in averaging browner
and more uniform below. Three Santarem birds are the concolor type,
but all the others are easily separable from concolor, and strikingly dis-
tinct from the Para medius. On the basis of our material, therefore,
ridgwayi should be recognized.

There are, of course, numerous other races of this variable species,
which has a wide distribution northward and westward.

422. Dendrocolaptes hoffmansi Hellmayr

Type locality: Calama, Rio Madeira

Villa Braga, left bank of Rio Tapajoz (Snethlage), recorded in error as "concolor
ridgwayi")

3 d1 2 9, Rio Tapajoz, Villa Braga and Apacy (Carnegie Mus.)

Only known from the localities listed above.

423. Dendrocolaptes transfasciatus Todd

Type locality: Miritituba, Rio Tapajoz
2 cf, Santarem

2 d\ Santarem (Carnegie Museum)
1 o\ Miritituba (do.)

Apparently a very rare bird, known from the three original speci-
mens, the two here listed, and one from Santarem in the Museu
Paulista (Pinto).

These two species presumably represent the picumnus group on the
south bank of the Lower Amazon. We have here a fine illustration of
the disadvantages of the trinomial system. On the "Formenkreis"
theory, these two birds could be reduced to subspecies of picumnus.
The facts are however, that geographically, they are isolated from any
other race of that species, nor are intermediate specimens known.
Treated as races, they are obviously not homologous with the sub-
specific variation just discussed in the species certhia. Treated as
species, they are not homologous with the specific characters of certhia
as contrasted with picumnus or platyrostris. The limits of our nomen-
clature cannot bring out the exact facts.

424. Dendrocolaptes picumnus picumnus Lichtenstein

Type locality: Cayenne

Obidos, Rio Jamunda (Snethlage and Zimmer)
16 d* 7 9 , Obidos (Carnegie Mus.)

210 bulletin: museum of comparative zoology

425. Dendrexetastes rufigula rufigula (Lesson)

Type locality: Cayenne
Obidos (Hellmayr)

1 cf , Obidos (Carnegie Mus.)

426. Dendrexetastes rufigula paraensis Lorenz

Type locality: Marco de Lagoa, Para

An exceedingly rare genus and species. The type of the present race
is still unique. On the right bank of the Rio Madeira, we find moniliger
Zimmer. The bird will almost certainly be found in the intermediate
area.

427. Hylexetastes perrotii perrotii (Lafresnaye)

Type locality: Cayenne

Obidos and Rio Jamunda (Snethlage, Zimmer, Pinto)
5 cf 6 9 , Obidos (Carnegie Museum)

428. Hylexetastes perrotii uniformis Hellmayr

Type locality: Calama, Rio Madeira

Rio Tapajoz: right bank, Caxiricatuba; left bank, Igarape-Brabo, Igarape
Amorin; Rio Jamauchim (all Zimmer)

1 d71, Rio Tapajoz, Caxiricatuba

2 a71 2 9 , Santarem (Carnegie Mus.)

2 d" 2 9 , Rio Tapajoz, Villa Braga (do.)

429. Xiphocolaptes orenocensis berlepschi Snethlage

Type locality: Cachoeira, Rio Purus

Rio Tapajoz, Apacy; Colonia de Mojuy, Santarem (Hellmayr, in Carnegie
Museum)

2 cf, Santarem (Carnegie Mus.)
1 9, Rio Tapajoz, Apacy (do.)

Hellmayr and Zimmer differ as to whether orenocensis is specifically
distinct from promeropirhynckus or not.

The two birds from Santarem appear to be a distinct subspecies
from true berlepschi, of which we have 3 from the Rio Purus, and 2
from Tonantins, Rio Solimoes. We note, however, that one of the
latter is inseparable from the Santarem birds, so that better series are
obviously needed. A specimen from Manacapuru, north bank of the
Solimoes, is apparently ignotns Ridgway.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 211

430. Dendroplex picus picus (Gmelin)

Type locality: Cayenne

Recorded abundantly by all authors from every part of our area, including
Mara jo Island.

1 c? 19, north bank of Amazon near Obidos
11 <? 11 9 , Rio Tapajoz; various localities

1 d\ Obidos (Carnegie Mus.)
10 o" 2 9, Santarem (do.)
3 d1 19, Rio Tapajoz, left bank (do.)

The three birds from near Obidos are more rufeseent below than the
Rio Tapajoz and a great series from Cayenne and Surinam. They thus
agree with Hellmayr's comment on birds from the Rio Branco in ap-
proaching kienerii. Those from Villa Braga have the larger size of
kienerii, but are not more rufeseent.

431. Dendroplex necopinus Zimmer

Type locality: Muirapinina, Rio Negro

Rio Jamunda, Faro and Villa Bella Imperatriz (Zimmer)

We have nothing but admiration for the keenness displayed by Mr.
Zimmer in picking out the anomalous birds described as necopinus
from his fine series of picus. We have 30 specimens before us, repre-
senting kienerii from both banks of the Rio Solimoes (Manacapuru,
Caviana, Sao Paulo de Olivenca, islands in the river near Mana-
capuru), and the Rio Purus (Arima, Nova Olinda, Hyutanahan). Two
birds from the islands in the Solimoes stand out from this series, and
display every one of the characters of color and proportionate size
ascribed to necopinus. Moreover, one is obviously adult and one is
immature, thus endorsing Mr. Zimmer's contention that the characters
of necopinus cannot be ascribed to some stage of immaturity of picus.
They come, however, from so near the type locality of kienerii, that
the application of the name becomes, perhaps, doubtful. A third
specimen from the same locality is apparently an intermediate. The
bill and color characters are those of kienerii, but it has the propor-
tionately long tail of necopinus. A male adult from Villa Braga has
the ovate streaking of necopinus rather than the squamate effect of
picus; it is olivaceous on the mantle rather than the rufescence of
-picus. It has the proportionately shorter tail of picus. The bill is that
of picus. In the longer wing-tip, and brown rather than rufous wing-

212 bulletin: museum of comparative zoology

coverts, it resembles necopinus. It will be apparent, therefore, that
the shorter, slender bill of necopinus is its best character.

We heartily agree with Mr. Zimmer in "seriously questioning" the
distinctness of the genus Dendroplex.

432. Xiphorhynchus guttatus guttatoides (Lafresnaye)

Type locality: Rio Maraiion, Peru
Villa Bella Imperatriz (Zimmer)

433. XlPHORPHYNCHUS GUTTATUS EYTONI (Sclater)

Type locality: Rio Capim, Par&

Innumerable records throughout the balance of our area from Marajo Island
(Snethlage) west to the Rio Madeira.
2 9 , Rio Acara, Acara
10 d" 5 9 , Rio Tapajoz; various localities right bank.
7 d1 3 9 , Benevides (Carnegie Mus.)
14 c? 5 9 , Santarem (do.)

3 cf , Rio Tapajoz, both banks (do.)

We are by no means convinced that there is not another recognizable
subspecies along the south bank of the Amazon. Our Tapajoz series
differs strikingly from the Acara birds in being more richly ochraceous
below, and in having the feathers of the back with much broader, light
centers, giving a less streaked, more guttate effect. While we defer to
Zimmer's critique of the remarkably variable nature of this series,
we note that he lists no topotypes of eytoni in his enormous series, and
Hellmayr, who had a good series of eytoni, lists only two specimens
outside the Para region.

434. Xiphorhynchus guttatus polystictus (Salvin and Godman)

Type locality: Bartica Grove, British Guiana
Obidos (Hellmayr) ; Faro (Zimmer and Snethlage) ; Pataua (Pinto)
7 cf 3 9 , Obidos (Carnegie Mus.)

Zimmer has shown (Amer. Mus. Nov., no. 756, 1934) that this little
known "species" is merely an earlier name for the well known sororius
(Berlepsch and Hartert). It is interesting to see how the various races
meet and intergrade along the extreme western limits of our area.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 213

435. XlPHORPHYNCHUS OCELLATUS OCELLATUS (Spix)

Type locality: corrected to Rio Madeira

Left bank of the Rio Tapajoz (Snethlage) ; Villa Bella Imperatriz, Faro, Rio
Xingii and Rio Tocantins (Zimmer)

12 cf 5 9 , Rio Tapajoz, Apacy and Villa Braga (Carnegie Mus.)

436. XlPHORHYNCHUS PARDALOTUS (Vieillot)

Type locality: Cayenne

Rio Jary, Obidos, Rio Jamunda (Snethlage); also right bank of Rio Tapajoz
(Zimmer)

9 c? 3 9 , Obidos (Carnegie Mus.)

This species occurs with ocellatus on the north bank of the Amazon,
and occurs on the south bank only where ocellatus is apparently lack-
ing.

437. Xiphorhynchus spixii spixii (Lesson)

Type locality: Para

Abundant from Para to the right bank of the Tapajoz
2d,2 9, Para, Bosque

4 c" 4 9 , Rio Tapajoz, right bank

8 cf 4 9 , Benevides (Carnegie Mus.)

5 cf 3 9 , Santarem (do.)

9 & 3 9 , Rio Tapajoz, right bank (do.)

This great series shows that fraterculus Ridgway is untenable.

438. Xiphorhynchus spixii elegans (Pelzeln)

Type locality: Engenho do Gama, Matto Grosso

Rio Tapajoz, left bank, Villa Braga (Snethlage) and Itaituba (Hellmayr).
7 cf 4 9 , Rio Tapajoz, left bank (Carnegie Mus.)

A bird of very limited distribution ; otherwise known only along the
Rio Madeira.

439. Xiphorhynchus obsoletus obsoletus (Lichtenstein)

Type locality: "province of Para"

Recorded from the north bank (Snethlage), and the south bank from the Rio
Tocantins westward (Chapman and Riker, Snethlage, Zimmer)
1 d\ Obidos

3 a" 1 9 , Para, Val-de-Caes

5 cf 3 9 , Rio Tapajoz; various localities.

1 cf 1 9 , Obidos (Carnegie Mus.)

11 d* 3 9, Santarem ' (do.)

4 cf 2 9 , Rio Tapajoz, both banks (do.)

214 bulletin: museum of comparative zoology

Apparently never definitely reported previously near Para. Our
specimens agree with three from Obidos in being a warmer, more
rufeseent brown than the Tapajoz series. Zimmer has shown that cer-
tain specimens from the north bank approach notatus (Eyton).

440. Lepidocolaptes albolineatus albolineatus (Lafresnaye)

Type locality: Cayenne
Rio Jary, Obidos, Rio Jamunda (Snethlage)
9 c? 4 9 , Obidos (Carnegie Mus.)

This is the bird better known as puncticeps (Sclater and Salvin).

441. Lepidocolaptes albolineatus layardi (Sclater)

Type locality: Para

Para (Layard, Snethlage, Stone, Zimmer); Peixe-Boi (Hellmayr); Rio Guama

(Snethlage); Rio Tocantins (Snethlage, Zimmer); Santarem (Chapman

and Riker), Rio Tapajoz, right bank, (Zimmer)

1 d\ Rio Tapajoz, Caxiricatuba

2 d1 3 9 , Benevides (Carnegie Mus.)
1 d\ Santarem (do.)

1 c\ Rio Tapajoz, Miritituba (do.)

442. Lepidocolaptes albolineatus madeirae (Chapman)

Type locality: Porto Velho, Rio Maderia
Rio Tapajoz, left bank (Snethlage, Hellmayr, Zimmer)
1 cf , Rio Tapajoz, Villa Braga (Carnegie Mus.)

Hellmayr and Zimmer disagree on the line of demarcation between
these two races, the former referring specimens from the right bank
of the Tapajoz to madeirae. Hellmayr also regards both these birds as
races of fuscicapillus, which he regards as specifically distinct.

443. Lepidocolaptes angustirostris coronatus (Lesson)

Type locality: Piauhy

Marajo Island (Hellmayr); Monte Alegre (Snethlage); Santarem (Allen;
Pinto)

1 cf 2 9 , Santarem

1 cf 1 9 , Santarem (Carnegie Mus.)

This species is more southern, and is primarily characteristic of
campo country. There is only one female on record from the north bank
of the Amazon. A series should be carefully compared.

griscom and greenway: birds of lower amazonia 215

444. Campylorhamphus trochilirostris snethlage.*:

Zimmer, 1934

Type locality: Serra de Parintins, Villa Bella Imperatriz. Also recorded from
mouth of Rio Andira and Rio Jamundd, Faro (Zimmer) ; Faro and Monte
Alegre (Snethlage)

445. Campylorhamphus procurvoides procurvoides (Lefresnaye)

Type locality: Cayenne

Rio Jary, Rio Jamunda (Snethlage); Faro (Zimmer)

1 d\ Obidos (Carnegie Mus).

446. Campylorhamphus procurvoides probatus Zimmer. 1934

Type locality: Igarape Auara, Rio Madeira

Villa Bella Imperatriz, left bank of Rio Tapajoz (Zimmer) and also Snethlage
for the latter locality

4 cf 3 9 , Rio Tapajoz, Itaituba, Villa Braga, Apacy (Carnegie Mus.)

447. Campylorhamphus procurvoides multostriatus

(Snethlage)

Type locality: Arumatheua, Rio Tocantins

Also east bank of Rio Tapajoz and the Rio Xingu (Zimmer); Aveiro (Pinto)

2 o* 2 9 , Rio Tapajoz, Caxiricatuba (August, 1932) and Pinhy
(May and June, 1933)

7 cf 2 9 , Santarem (Carnegie Mus.)
2c?2 9, Rio Tapajoz, Miritituba (do.)

Zimmer's careful monograph of this group (Amer. Mus. Novit., no.
728, 1934) has been followed here. Its most important departure from
Hellmayr's treatment is the discovery of the race probatus, connecting
the little known multostriatus Snethlage with the procurvoides group.

448. Nasica longirostris longirostris (Vieillot)

Type locality: Obidos, by subsequent designation

Obidos (Hellmayr); Maracd, Monte Alegre, Rio Maecuru, Rio Jamundd
(Snethlage); Marajo Island (Brodkorb)

2^59, Boca do Igarape-Piaba, near Obidos
1 9 , Obidos (Carnegie Mus.)

216 bulletin: museum of comparative zoology

449. Nasica longirostris australis Griscom and Greenway, 1937

Type locality: Santarem, Brazil

Santarem (Chapman and Riker, Hellmayr); Rio Tocantins and Rio Tapajoz
(Snethlage) ; Pataua and Santarem (Pinto)

3 d" 2 9,1 imm. ?, Rio Tapajoz, Santarem and Caxiricatuba.

1 9 , Rio Tapajoz, Apagy, (Carnegie Mus.)

5 cf 5 9, Santarem (do.)

The last word has yet to be written on subspecific variation in this
Woodhewer. Our description of australis was based on the assumption
that birds from the south bank of the Amazon were separable from
those of the north bank, and the material later examined in the
Carnegie Museum confirmed it. Mr. Zimmer, however, assures us that
a larger series in the American Museum does not bear out the charac-
ters alleged.

However this may be, there are unquestionably two subspecies.
Nearly a century ago Lesson described albicollis from French Guiana,
but even Hellmayr never saw a specimen from that country. Five
specimens from French Guiana in the Carnegie Museum are sub-
specifically distinct from australis at a mere glance. It now turns out
most unfortunate that we should have restricted typical longirostris to
Obidos on the north bank of the Amazon. The facts are that -this
population is intermediate as usual. We do not for a moment favor the
recognition of three subspecies. The material in Cambridge, New
York, and Pittsburgh should be combined to determine whether
australis is a synonym of longirostris, or whether albicollis is a synonym
of longirostris. Nine times out of ten in birds having a similar range,
the best "break" is the Amazon River.

450. Glyphorhynchus spirurus spirurus (Veillot)

Type locality: Cayenne

Amapa' and Rio Jamunda, Faro (Snethlage); Faro (Zimmer); Obidos (Hell-
mayr)

3 d\ Obidos (Carnegie Mus.)

451. Glyphorhynchus spirurus inornatus Zimmer, 1934

Type locality: Villa Bella Imperatriz
Left bank of Rio Tapajoz (Zimmer)

2 cf 1 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

This recently described form ranges west to the Rio Madeira.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 217

452. Glyphorhynchus spirurus cuneatus (Lichtenstein)

Type locality: Bahia, Brazil

Abundantly recorded around Para, (all collectors) ; Rio Tocantins, Rio Xingu
and Rio Tapajoz, east bank (Snethlage and Zimmer)
7 <? 6 9 1 ? , vicinity of Para
2 cf 1 9 , Rio Acara, Acara

1 9 , Rio Tapajoz, Pataua

5 o1 2 9, Benevides (Carnegie Mus.)

6 c? 1 9 , Santarem (do.)

2 c? 1 9 , Rio Tapajoz, right bank (do.)

453. Sittasomus griseicapilltjs amazonus Lafresnaye

Type locality: Peruvian Amazons

Left bank of Rio Tapajoz (Snethlage and Zimmer) ; Villa Bella Imperatriz and
Rio Xingu (Zimmer); Rio Tocantins (Snethlage)?

11 d" 4 9, Rio Tapajoz, left bank (Carnegie Mus.)

454. Sittasomus griseicapillus axillaris Zimmer, 1934

Type locality: near Faro, Rio Jamunda, Brazil
Also right bank of Rio Tapajoz (Zimmer)

3 d1 1 9 , Rio Tapajoz, Pinhy and Caxiricatuba
6 d* 1 9 , Santarem (Carnegie Mus.)

2 d\ Rio Tapajoz, Miritituba (do.)

2 d" 2 9 , Obidos (do.)

The species is recorded from Obidos by Hellmayr. In view of the
curious distribution of the races we cannot be absolutely certain that
this record belongs here. We cannot distinguish them locally.

455. Deconychura longicauda longicauda (Pelzeln)

Type locality: Manaos, Brazil

Rio Jamunda, Faro (Zimmer); Obidos (Carnegie Mus.)
2 d" 1 9 , Obidos (Carnegie Mus.)

456. Deconychura longicauda pallida Zimmer, 1934

Type locality: Rio Purus, Brazil

Villa Bella Imperatriz, Rio Tapajoz, left bank; Rio Iriri and Rio Xingu

(Zimmer); Rio Iriri (Snethlage); Peixe-Boi (Hellmayr); Providencia

(Snethlage); Benevides (Zimmer)

1 d\ Rio Tapajoz, Pataua

2 d\ Benevides (Carnegie Mus.)

1 d\ Rio Tapajoz, Miritituba (do.)

2 d\ (do.), Villa Braga (do.)

218 bulletin: museum of comparative zoology

457. Deconychura stictolaema stictolaema (Pelzeln)

Type locality: Borba, Rio Madeira

Villa Bella Imperatriz and Rio Tocantins (Zimmer)

1 9 , Rio Tapajoz, Pataua

1 d\ Santarem (Carnegie Mus.)

1 d71, Rio Tapajoz, Villa Braga (do.)

1 d* , Rio Tapajoz, Miritituba (do.)

458. Deconychura stictolaema clarior Zimmer, 1929

Type locality: Pied Saut, Oyapok, French Guiana
Rio Jamunda, Faro (Zimmer) ; Igarape Aniba (Pinto)
1 cf 1 9 , Obidos (Carnegie Mus.)

459. Dendrocincla merula merula (Lichtenstein)

Type locality: Cayenne

Rio Jamunda, Faro (Zimmer)

1 c? 1 9 , Obidos (Carnegie Mus.)

460. Dendrocincla merula badia Zimmer, 1934

Type locality : Pedral, Rio Tocantins

Igarape-Assu, Peixe-Boi (Hellmayr); Rio Guama (Snethlage); Rio Tocantins,
right bank (Zimmer)

2 9 , Rio Acara, Acara

461. Dendrocincla merula castanoptera Ridgway

Type locality: Diamantina, Para, Brazil

Santarem (Chapman and Riker); Rio Tapajoz, right bank (Zimmer)

8 <f 5 9 , Rio Tapajoz, various localities, right bank

7 cf 3 9 , Santarem (Carnegie Mus.)

5 cf 2 9 , Rio Tapajoz, Miritituba (do.)

This race presumably ranges eastward to the left bank of the Rio
Tocantins.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 219

462. Dendrocincla meriila olivascens Zimmer, 1934

Type locality: Villa Bella Imperatriz, Amazon River left bank of the Rio
Tapajoz (Snethlage and Zimmer)

1 c? 4 9 , Rio Tapajoz, left bank (Carnegie Mus.)

463. Dendrocincla fuliginosa fuliginosa (Vieillot)

Type locality: Cayenne

Obidos and Rio Jamunda (Snethlage) ; Faro (Zimmer)
9o"3 9, Obidos (Carnegie Mus.)

464. Dendrocincla fuliginosa rufo-olivacea Ridgway

Type locality: Diamantina, Rio Tapajoz

Vicinity of Para (numerous collectors); Castanhal (Stone); Rio Acara (Hell-

mayr); Rio Tocantins (Snethlage and Zimmer); Rio Xingu (Zimmer),

Rio Tapajoz, right bank (Zimmer)

2 d1 3 9 , Benevides (Carnegie Mus.)

2 C?, Santarem (do.)

3 d* 4 9 , Rio Tapajoz, Miritituba (do.)

465. Dendrocincla fuliginosa atrirostris
(Lafresnaye and D'Orbigny)

Type locality: Guarayos, Bolivia

Left bank of Rio Tapajoz, (Snethlage, Zimmer); Villa Bella Imperatriz
(Zimmer)

4 6" 2 9 , Rio Tapajoz, left bank (Carnegie Mus.)

It will be noted that Zimmer 's arrangement of the species and races
of this genus is radically different from that of Hellmayr. We follow it
here, as the material at his command was imcomparably superior.

Family FURNARIIDAE

466. Furnarius minor Pelzeln

Type locality: Rio Madeira

Santarem (Chapman and Riker, Hellmayr) ; Monte Alegre, Rio Maecuru, Rio
Jamunda (Snethlage)

4 c? 9 9, north bank of Amazon, near Obidos

1 cf, Obidos (Carnegie Mus.)

2 cf 19, Rio Tapajoz

11 cf 12 9, Santarem (Carnegie Mus.)
1 cT 19, Rio Tapajoz, right bank (do.)

220 bulletin: museum of comparative zoology

Two birds from upper Amazonian Ecuador are apparently much
burner below than lower Amazonian series of comparable plumage.
This is especially marked on the belly, which is not so extensively
white in sharp contrast with the chest.

467. Furnarius figulus pileatus Sclater and Salvin

Type locality: Santarem, Brazil

Santarem (Chapman and Riker); Rio Iriri, Arumanduba, Monte Alegre, Rio
Maecuru, Rio Jamunda (Snethlage)

6 c? 4 9 , north bank of the Amazon near Obidos

1 9 , Obidos (Carnegie Mus.)
1 cf, Rio Tapajoz, Santarem

7 cf 5 9 , Santarem (Carnegie Mus.)

These two species of Furnarius are characteristic of the Amazon
River, while leucopus has representative races both north and south
of the Amazon, but not in the valley itself.

468. Synallaxis albescens subspecies?

Mexiana Island (Hagman); Marajo Island, Arumanduba (Snethlage)

Series from Mexiana and Marajo Islands have yet to be properly
compared. Hellmayr's reference of the Mexiana bird to albigularis was
based on the examination of one immature female. At that time
albigularis was supposed to range from the Rio Napo, Ecuador through-
out the Amazon Valley. The species is primarily found in open
"campo" or semi-arid scrub country, and is probably nearly as local
in our area, as the few records would indicate. The population con-
sidered will probably prove to be one more of the already numerous
local subspecies.

469. Synallaxis albescens inaequalis Zimmer, 1935.

Type locality: Villa Bella Imperatriz

1 d" 1 9 , Lago Grande, south bank of the Amazon
1 o71, Santarem (Carnegie Mus.)

These specimens are provisionally referred to this recently described
form, which ranges between the Madeira and Tapajoz rivers, "crossing
to the north bank in the same general region."

10 c? 9

2 d1 1

4 c? 1

3c?4

1 cf 1

3 cf 3

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 221

470. Synallaxis gujanensis gujanensis (Gmelin)

Type locality: Cayenne

Recorded abundantly throughout our area from Marajo Island (Brodkorb) to
the left bank of the Rio Tapajoz

1 d\ near Obidos

2 cf 3 9 , Para, Val-de-Caes
Rio Tapajoz, right bank
Obidos (Carnegie Mus.)
Benevides (do.)
Santarem (do.)
Rio Tapajoz, right bank (do.)

(do.) , left bank (do.)

The typical subspecies ranges south along the coast to Maranhao,
and is replaced by inomata Pelzeln from the Rio Madeira westward.
A certain percentage of birds from the Rio Tapajoz area are minutely
hoarier than Para birds, but there is no necessity for a third subspecies.

471. Synallaxis propinqua Pelzeln

Type locality: Rio Madeira, below mouth of Rio Marcy
Rio Tocantins, Baiao (Zimmer)

An interesting range extension of this very distinct upper Amazonian
species, which is as yet unknown between the type locality and the
Rio Tocantins.

472. Synallaxis rutilans rutilans Temminck

Type locality: Cameta, Rio Tocantins

Rio Tocantins, left bank, Rio Xingu, Rio Iriri, Rio Jamauchim, Rio Tapajoz,
right bank (Snethlage); Santarem (Chapman and Riker, Hellmayr);
numerous localities in same range (Zimmer)
15 cf 11 9,1?, Rio Tapajoz, right bank
15 c? 4 9, Santarem (Carnegie Mus.)
1 c?1 2 9, Rio Tapajoz, right bank (do.)

473. Synallaxis rutilans dissors Zimmer, 1935

Type locality: Manaos, Brazil
Obidos (Snethlage and Zimmer); Faro (Zimmer)
1 cf 3 9 , Obidos (Carnegie Mus.)

222 bulletin: museum of compakative zoology

474. Synallaxis rutilans amazonica Hellmayr

Type locality: Itaituba, left bank of Rio Tapajoz

Rio Tapajoz, left bank, Boim and Villa Braga (Hellmayr); Villa Bella Inapera-
triz (Zimmer)

19 1?, Rio Tapajoz, Pinhel

28 <? 15 9, (do.), left bank (Carnegie Mus.)

475. Synallaxis rutilans omissa Hartert

Type locality: Para

Recorded abundantly by all collectors from the vicinity of Para to the right
bank of the Tocantins

2o,5 9,2?, Para, Val-de-Caes

8 cf 4 9 , Benevides (Carnegie Mus.)

This series raises a very interesting problem. Hartert originally
described omissa as a distinct species, based on males of the cinereous
type, the chestnut reduced to the wings only. Later collections near
Para reported on by Hellmayr yielded females of the chestnut type,
which were barely separable from true rutilans in minor differences of
color, shade and tone. Additional males of the cinereous type proved
to have a variable number of feathers on breast and back. Dr. Hell-
mayr was obviously correct in reducing omissa to a race of rutilans.

The series before us slightly alters the interpretation of the facts,
as it becomes apparent that the sex difference breaks down. We have
a pair of adults in the chestnut phase, and the balance of the series
consists of cinereous birds of both sexes. A study of this series shows
exactly the sexual and age variations to be expected in Synallaxis.
Some are obviously immature birds in the freckling below, in the
shorter, less pointed and less stiff tail feathers. It is also these birds
that have the maximum amount of chestnut feathers on the abdomen.
It will be apparent,, therefore, that some birds are always chestnut and
other birds are hatched cinereous. We are dealing, therefore, not
with sexual dimorphism, but a color phase.

476. Certhiaxis cinnamomea cinnamomea (Gmelin)

Type locality: Cayenne

Arumanduba, Erere, (Snethlage); Mexiana Island (Wallace; Hagman); Marajo

Island (Muller); Rio Tocantins, Baiao (Zimmer), and probably Snethlage

also; Caviana Island (Brodkorb)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 223

477. Certhiaxis cinnamomea pallida Zimmer, 1935

Type locality: Igarape

Cacao Pereira, Rio Negro, Rio Jamunda' (Snethlage and Zimmer) ; Rio Tapajoz
(Zimmer); Santarem (Chapman and Riker)

4 6", Rio Tapajoz, Pinhel (west bank)

2 d\ Obidos

2 cf 1 9 , (do.) (Carnegie Mus.)

3 cf 1 9 , Santarem (do.)

478. Certhiaxis mustelina (Sclater)

Type locality: Rio Madeira, Brazil

Santarem (Chapman and Riker) ; Monte Alegre (Snethlage and Zimmer)

2 cf 1 9 , north bank of Amazon near Obidos

3o,29, (do.) (Carnegie Mus.)

1 cf , Santarem (do.)

479. Cranioleuca vulpina vulpina (Pelzeln)

Type locality: Matto Grosso, Brazil
Rio Tocantins and Rio Tapajoz (Snethlage)
1 d" 1 9 , Rio Tapajoz, Santarem
13 a" 5 9 , Santarem (Carnegie Mus.)

1 9 , Rio Tapajoz, Miritituba (do.)

480. Cranioleuca vulpina alopecias (Pelzeln)

Type locality: Rio Branco, Brazil
Monte Alegre, Rio Maecuru (Snethlage)

3 c? 3 9 , near Obidos

1 d" 3 9 , Obidos (Carnegie Mus.)

481. Cranioleuca mulleri (Hellmayr)

Type locality: Mexiana Island

Also Monte Alegre, Obidos, Rio Jamunda (Snethlage)
1 9 , north bank near Obidos

3 9, Obidos (Carnegie Mus.)

4 d" 2 9 , Santarem (do.)

224 bulletin: museum of comparative zoology

482. Cranioleuca gutturata (Lafresnaye and D'Orbigny)

Type locality: Yuracares, Bolivia

Rio Tocantins, Rio Tapajoz (Snethlage)

3 d\ Rio Tapajoz, Miritituba (Carnegie Mus.)

A comparatively little known bird, with a remarkably wide but
scattered distribution.

483. Thripophaga fusciceps obidensis Todd

Type locality : islands near Obidos, Brazil

2 d" 4 9 , Obidos (Carnegie Mus.)

Known only from the type series of six specimens in the Carnegie
Museum. The genus is so rare that other forms will undoubtedly be
discovered in the intervening area. Typical fusciceps from Bolivia is
reported from southeastern Peru and eastern Ecuador, but these
specimens do not agree with the type, according to Hellmayr.

484. Berlepschia rikeri (Ridgway)

Type locality: Diamantina, Santarem, Brazil

Para (Snethlage); Santarem (Chapman and Riker, Pinto)

1 cf, Para, Val-de-Caes

1 9 , Rio Acara, Acara

485. Hyloctistes subulatus subulatus (Spix)

Type locality: Rio Amazonas

3 c? 2 9 , Rio Tapajoz, west bank, Ville Braga (Carnegie Mus.)

An extension of range eastward from the Rio Madeira.

486. Ancistrops strigilatus cognitus
Griscom and Greenway, 1937

Type locality: Tauary, Rio Tapajoz

Santarem and Rio Tapajoz, both banks (Griscom and Greenway)

1 9 , Rio Tapajoz, Tauary

2 cT 1 9, (do.), Miritituba (Carnegie Mus.)
2 c? 1 9 , (do.), Villa Braga (do.)

1 d1, Santarem (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 225

487. Philydor pyrrhodes (Cabanis)

Type locality: coast of British Guiana

Para (Snethlage); Rio Capim (Goeldi, Stone); Rio Tocantins, Obidos (Sneth-
lage); Santarem (Hellmayr, Pinto); Utinga, Rio Tocantins, Rio Tapajoz
(Zimmer)

2 d71 2 9 , Rio Tapajoz, east bank
1 9 , Rio Acara, Acara

1 d\ Obidos (Carnegie Mus.)

2 cf , Santarem (do.)

3 d", Rio Tapajoz, both banks (do.)

488. Philydor erythropterus diluvialis
Griscom and Greenway, 1937

Type locality: Caxiricatuba, right bank of Rio Tapajoz, Para, Brazil.
Rio Tapajoz, both banks (Griscom and Greenway)

1 d\ Rio Tapajoz, Caxiricatuba

3 cf (do.), Villa Braga (Carnegie Mus.)

As mentioned in the original description, this subspecies is very
distinct from large series from upper Amazonia (east Ecuador, Rio
Solimoes, Rio Purus), but whether these last represent true erythrop-
terus, based on "Bogota" collections, is problematical.

489. Philydor ruficaudatus (Lafresnaye and D'Orbigny)

Type locality: Yuracares, Bolivia

S. Antonio do Prata, Rio Guama\ Rio Tocantins, Rio Jary (Snethlage)

1 ?, Rio Tapajoz, Tauary

2 cf 1 9 , Obidos (Carnegie Mus.)

2 d\ Rio Tapajoz, Miritituba, and Villa Braga (do.)

This widely ranging species has proved to be very constant geo-
graphically. Snethlage's records should be critically examined in
view of the confusion with the next species which has prevailed in the
past. We are quite aware that Zimmer (Am. Mus. Novit., 785, March,
1935,) has published a valuable critique on the relations between
P. ruficaudatus and P. erythrocercus, and that he lists no specimens of
P. ruficaudatus from our area in Brazil. The specimens we record
above are, therefore, a notable range extension, and its occurrence in a
region where P. erythrocercus is abundant further endorses the concept

226 bulletin: museum of comparative zoology

of P. ruficaudatus as a distinct species. The Carnegie Museum also
possesses specimens from the Rio Purus, thus partly filling the great
hiatus in this species' range.

490. Philydor erythrocercus erythrocercus (Pelzeln)

Type locality: Manaos, Brazil
Obidos (Snethlage); Faro (Zimmer)

1 d" 1 9 , Obidos (Carnegie Mus.)

491. Philydor erythrocercus lyra Cherrie

Type locality: Rio Roosevelt, Matto Grosso

Para (Wallace, Para); Peixe-Boi, Igarapc-Assu, S. Antonio do Prata (Hell-
mayr); Santarem (Chapman and Riker); Sta. Isabel, Rio Guama, Rio
Moju, Rio Tocantins, Rio Xingu, Rio Curua, Rio Tapajoz, Rio Jamauchim
(Snethlage); large series, Para to Villa Bella Imperatriz (Zimmer)
5 d" 3 9 , Para, Val-de-Caes

15 d1 5 9 1 ?, Rio Tapajoz, various localities, east bank.
32 d71 11 9, Benevides, Rio Tapajoz, Santarem (Carnegie Mus.)

492. Automolus infuscatus cervicalis (Sclater)

Type locality: Bartica Grove, British Guiana

Rio Jary (Snethlage); Obidos (Hellmayr); Faro (Zimmer)

2 d1 2 9 , Obidos (Carnegie Mus.)

493. Automolus infuscatus paraensis Hartert

Type locality: Benevides, near Pard

Near city of Para and Rio Acara (numerous records, all collectors) ; Rio Tocan-
tins, Rio Iriri, Rio Tapajoz, Rio Jamauchim (Snethlage); Rio Tapajoz to
Para (Zimmer)

1 cf 1 9 , Pard, Bosque
1 ?, Rio Tapajoz, various localities east bank

, Benevides (Carnegie Mus.)

, Santarem (do.)

, Rio Tapajoz, both banks (do.)

494. Automolus ochrolaemus auricularis Zimmer

Type locality: Caxiricatuba, Rio Tapajoz

Rio Tapajoz (Snethlage and Zimmer) ; Villa Bella Imperatriz (Zimmer)

3 d1 2 9 , from the type locality.

2 d1 1 9 , Santarem (Carnegie Mus.)

17 d1 4 9 , Rio Tapajoz, both banks (do.)

9^8

5 d1 4

11 d1 5

2 d1 3

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 227

495. Automolus ochrolaemus turdinus (Pelzeln)

Type locality: Manaos, Brazil

Obidos, Rio Jamunda (Faro) on the north bank (Snethlage and Zimmer)
2 cf 1 9 , Obidos (Carnegie Mus.)

496. Automolus rufipileatus rufipileatus (Pelzeln)

Type locality: Pard

Para (Natterer); Rio Guama (Stone); Rio Tocantins, Baiao (Snethlage);
Rio Tapajoz (in Carnegie Museum, fide Hellmayr)

2 cf 1 9 , Santarem (Carnegie Mus.)

1 d* 1 9 , Rio Tapajoz, Miritituba (do.)

A rare and little known species, the typical race reported west to the
Rio Purus.

497. Xenops minutus genibarbis Illiger

Type locality: Cameta, Rio Tocantins

Para (Stone); S. Antonio do Prata (Hellmayr); numerous localities near Para
(Snethlage); Rio Tocantins, Rio Xingu, Rio Jamauchim, Rio Tapajoz
(Snethlage) ; Santarem (Hellmayr) ; numerous localities, Para to Villa
Bella Imperatriz (Zimmer)
1 d\ Para, Bosque

1 C? 1 ?, Rio Acara, Acara

12 cf 4 9 , Rio Tapajoz, various localities, east bank

13 cf 8 9 , Benevides, Santarem, Rio Tapajoz, both banks, (Car-

negie Mus.)

49S. Xenops minutus ruficaudus (Vieillot)

Type locality: Cayenne

Obidos (Snethlage); Faro (Zimmer)

3 c? 1 9 , Obidos (Carnegie Mus.)

499. Xenops tenuirostris tenuirostris Pelzeln

Type locality: Rio Madeira

Rio Tapajoz, Apacy, Itaituba (Hellmayr); left bank of Rio Tapajoz (Zimmer)

2 cf 1 9 , Rio Tapajoz, Apacy, Itaituba (Carnegie Mus.)

228 bulletin: museum of comparative zoology

500. Xenops rutilans purusianus Todd

Type locality: Hyutanahan, Rio Purus, Brazil

Left bank of Rio Tapajoz, Igarape Amorin (Zimmer)

It is interesting to note that minutus is very definitely the dominant
species in our area.

501. Sclerurus mexicanus macconnelli Chubb

Type locality: Ituribisci, British Guiana

Rio Capim (Wallace) ; Peixe-Boi (Hellmayr) ; Rio Tapajoz (Snethlage and Zim-
mer)

1 9 , Para, Bosque

3 cf, Rio Tapajoz, east bank

3 d\ Obidos (Carnegie Mus.)

1 o" 2 9, Santarem (do.)

3 cf 2 9 , Rio Tapajoz, both banks (do.)

502. Sclerurus rufigularis rufigularis Pelzeln

Type locality: Borba, Rio Madeira

Para, Rio Guama (Stone) ; S. Antonio do Prata (Hellmayr) ; various localities
near Para and Rio Tocantins (Snethlage)

1 o71 1 9 , Para, Bosque

2 <? 2 9 , Benevides (Carnegie Mus.)

503. Sclerurus rufigularis fulvigularis Todd

Type locality: Tamanoir, French Guiana
Obidos (Snethlage)

3 9 , Obidos (Carnegie Mus.)

504. Sclerurus caudacutus insignis Zimmer, 1934
Type locality: Faro, Rio Jamunda, Brazil

Only known from the type locality.

505. Sclerurus caudacutus pallidus Zimmer, 1934

Type locality: Villa Bella Imperatriz

Igarape-Assu (Hellmayr); Rio Capim (Wallace); Peixe-Boi, Rio Tocantins,
Rio Jamauchim (Snethlage) ; also Rio Tocantins and Rio Tapajoz (Zim-
mer)

1 c? 2 9 , Rio Tapajoz, east bank

1 o71 3 9 , Benevides (Carnegie Mus.)

2 d1 4 9 , Santarem (do.)

2 cf 4 9 , Rio Tapajoz, both banks (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 229

Family FORMICARIIDAE

506. Cymbilaimus lineatus lineatus (Leach)

Type locality: Berbice, British Guiana

Rio Jary (Snethlage) ; Rio Jamunda, Faro (Snethlage and Zimmer)
2 d" 2 9 , Obidos (Carnegie Mus.)

507. Cymbilaimus lineatus intermedius (Hartert and Goodson)

Type locality: Rio Madeira, Brazil

Rio Tocantins, Rio Xingu, Rio Tapajoz (numerous localities, Snethlage and
Zimmer)

15 c? 2 9 , Rio Tapajoz, various localities, east bank
5 cf 7 9 , Santarem (Carnegie Mus.)

4 d1 2 9 , Rio Tapajoz, both banks (do.)

It is surprising that this Ant-Shrike should be lacking from the
Para region.

508. Frederickena viridis (Vieillot)

Type locality: Cayenne

10 d1 4 9 , Obidos (Carnegie Mus.)

It is curious that this Ant-Shrike should as yet be unreported from
Brazil north of the Amazon.

509. Taraba major semifasciatus (Cabanis)

Type locality: Para

Para, (Layard, Natterer, Stone, Wallace); Rio Guamd, Castanhal (Stone);
Igarape-Assu, S. Antonio do Prata, Peixe-Boi (Hellmayr) ; Quati-puru,
Rio Guama, Rio Moju, Rio Tapajoz, Arumanduba, Monte Alegre, Obidos,
Rio Jamunda (Snethlage); Rio Tocantins, Rio Xingu, Villa Bella Impera-
triz (Zimmer) ; Santarem (Hellmayr, Chapman and Riker)

1 cf 1 9 , Rio Tapajoz, east bank

2 9 , near Obidos

2 d\ Obidos (Carnegie Mus.)
1 cf 1 9 , Benevides (do.)

5 cf 2 9 , Santarem (do.)

3 cf 1 9 , Rio Tapajoz (do.)

230 bulletin: museum of comparative zoology

510. Sakesphorus luctuosus luctuosus (Liechtenstein)

Type locality: Cameta, Rio Tocantins

Rio Tocantins, Rio Xingu, Rio Iriri, Rio Tapajoz, Rio Jamauchim, Aruman-
duba, Monte Alegre, Rio Maecuru, Obidos, Rio Jamunda (all Snethlage
under the genus Myrmelastes) ; Santarem (Chapman and Riker, Hell-
mayr)

3 cf 1 9 , Para, Val-de-Caes

2 d" 2 9 , Obidos (Carnegie Mus.)

9 d" 9 9 , Santarem (do.)

2 cf 2 9 , Rio Tapajoz, both banks (do.)

4 d1 5 9 , Rio Tapajoz, various localities, east bank
2 cf 1 9 , near Obidos

It seems surprising that this characteristic lower Amazonian species
should have escaped detection previously in the Para region. The typi-
cal form ranges west to the Rio Madeira and is represented in the
interior of Goyaz by the little known araguayae Hellmayr. S. hagmanni
Mir.-Ribeiro would appear to be a synonym of the typical form.

511. Thamnophilus doliatus doliatus (Linnaeus)

Type locality: Surinam

Marajo Island (Sclater and Salvin, Allen, Snethlage, Hellmayr; Castanhal,
near Para, Amapa, Monte Alegre (Snethlage); Caviana Island (Brodkorb)

512. Thamnophilus doliatus difficilis Hellmayr

Type locality: Rio Claro, Goyaz, Brazil
Tocantins River (fide Hellmayr, 1929)

513. Thamnophilus doliatus signatus Zimmer, 1933

Type locality : Santarem, Brazil

Santarem, (Chapman and Riker, Pinto); Rio Tapajoz, Villa Bella Imperatriz
Rio Jamunda (Zimmer) ; Obidos (Pelzeln)

1 cf 2 9 , Rio Tapajoz, Santarem

4 cf 5 9 , Santarem (Carnegie Mus.)

6 d" 9 9 , Obidos (Carnegie Mus.)

This race is little more than a variable series of intermediates.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 231

514. Thamnophilus palliatus palliatus (Lichtenstein)

Type locality: Bahia, Brazil

Para (Layard, Stone, Pinto) ; S. Antonio do Prata (Hellmayr) ; Quati-puru,

Braganca, Rio Guama, Rio Moju, Rio Tocantins, Rio Tapajoz (Snethlage

and Zimmer)

4 cf 3 9 , Benevides (Carnegie Mus.)

Zimmer combines this species with the upper Amazonian temiipunc-
tatus, reviving puncticcps Sclater as the connecting form, which ranges
from the Rio Madeira to northeastern Peru (cf. Amer. Mus. Novit.,
no. 646, pp. 9-15).

515. Thamnophilus nigrocinereus nigrocinereus Scalter

Type locality: Rio Tocantins

Mexiana Island (Wallace, Hellmayr, Hagmann); Ilha das Oncas, Marajo
Island, Arumanduba, Monte Alegre (Snethlage)

9(^29, Para, Val-de-Caes

1 cf 1 9 , Rio Tocantins, Cameta

516. Thamnophilus nigrocinereus huberi Snethlage

Type locality: Ilha de Goyana, Rio Tapajoz
Santarem (Zimmer)

10 cf 4 9 , type locality (Carnegie Mus.)
1 cf , Rio Tapajoz, Itaituba (do.)
9 cf 14 9 , Santarem, (do.)

The Santarem series is easily separable from the topotypes. Males
are minutely paler and greyish on the abdomen, while females are a
noticeably less rich rufescent brown below.

517. Thamnophilus aethiops punctuliger Pelzeln

Type locality: Borba, Rio Madeira

Villa Bella Imperatriz, Rio Jamunda, Faro and Rio Tapajoz, west bank
(Zimmer)

11 cf 13 9, Rio Tapajoz, Villa Braga (Carnegie Mus.)

232 bulletin: museum of comparative zoology

518. Thamnophilus aethiops atriceps Todd

Type locality: Miritituba, Rio Tapajoz

Rio Tapajoz, right or east bank (Todd and Zimmer)

3 c? 2 9 , type locality

6 c? 3 9 , Rio Tapajoz, type series (Carnegie Mus.)

4 d" 3 9 , Santarem (do.)

519. Thamnophilus aethiops incertus Pelzeln

Type locality: Para

Numerous records by all collectors from the Para region to the right bank of the
Rio Tocantins; see especially Snethlage and Zimmer

4 cf 2 9 , Para, Bosque, and Val-de-Caes

1 9 , Rio Acara, Acara

13 cf 9 9 , Benevides (Carnegie Mus.)

The Acara female is remarkably distinct in having the belly and
abdomen whitish in striking contrast with the tawny chest.

520. Thamnophilus schistaceus inornatus Ridgway

Type locality: Diamantina, Santarem, Brazil

Rio Tapajoz, left bank (Snethlage and Zimmer); the right bank (Zimmer);
Rio Tocantins and Rio Xingii (Zimmer); Santarem (Chapman and
Riker)

16 cf 16 9 , Rio Tapajoz, east bank
10 d 8 9, Santarem (Carnegie Mus.)
10 cf 8 9, Rio Tapajoz, both banks

Recorded by Snethlage as Dsyithamnus schistaceus and squamosus.
Birds from the left bank of the Rio Tapajoz have often been referred
to typical schistaceus D'Orbigny.

521. Thamnophilus murinus cayennensis Todd

Type locality: Pied Saut, French Guiana
Obidos, Rio Jamunda, Faro (Snethlage)

11 d 9 9, Obidos (Carnegie Mus.)

This species does not occur on the south side of the Amazon in our
rea, but canipennis Todd crosses the river on the left bank of the Rio
Madeira.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 233

522. Thamnophilus punctatus punctatus (Shaw)

Type locality: Cayenne

Rio Jary, Rio Maecuru, Monte Alegre, Obidos, Rio Jamunda (Snethlage) and
also Marajo Island according to Snethlage; numerous additional speci-
mens from the north bank recorded by Zimmer
6^2 9, Obidos (Carnegie Mus.)

523. Thamnophilus punctatus saturatus Todd

Type locality: Villa Braga, left bank, Rio Tapajoz
Both banks of the Rio Tapajoz (Zimmer)

1 9 , Rio Tapajoz, left bank, Pinhel

7 cf 1 9 , Rio Tapajoz, right bank, Santarem, etc. (Carnegie Mus.)

524. Thamnophilus punctatus stictocephalus Pelzeln

Type locality: Sao Vincente, Matto Grosso

Right bank of Rio Xingii and left bank, Rio Tocantins (Zimmer)

1 c? from Val-de Caes is apparently this race in having the crown
feathers extensively white basally, but we have no comparable mate-
rial. For notes on the identity of this form, cf. Zimmer, 1933, p. 11.

525. Thamnophilus amazonicus amazonicus Sclater

Type locality: Rio Javary, upper Amazon

Villa Bella Imperatriz (Zimmer); Rio Tapajoz, left bank (Snethlage)
3 cf 6 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

526. Thamnophilus amazonicus obscurus Zimmer

Type locality: Tauary, Rio Tapajoz

Rio Tapajoz, right bank and Rio Xingu, right bank (Zimmer)
12 cf 15 9 , Rio Tapajoz, east bank

1 cf 19, Santarem (Carnegie Mus.)

5 cf 2 9, Rio Tapajoz, Miritituba

234 bulletin: museum of comparative zoology

527. Thamnophilus amazonicus paraensis Todd

Type locality: Benevides, Para

Innumerable records near Para (all collectors) west to the Rio Tocantins
(Snethlage and Zimmer); on the north bank, Rio Jary (Snethlage) and
Rio Jamunda, Faro (Zimmer)
4 cf 1 9 , Para, Val-de-Caes
6 cf 4 9 , Rio Acara, Acara

In spite of Zimmer's helpful critique on the distinguishing characters
of males of this species and punctatus, we would just as soon refer
some of the males listed above to punctatus, were it not for the fact
that that species is not recorded from the vicinity of Para.

528. Pygiptila stellaris stellaris (Spix)

Type locality: province of Para,

Para, (Spix, Stone); Utinga, S. Antonio do Prata, Ipitinga (Hellmayr); Rio
Tocantins, Rio Xingii, Rio Iriri, Rio Curua, Rio Tapajoz, Rio Jamau-
chim (Snethlage) ; large series from Utinga to Villa Bella Imperatriz (Zim-
mer)

1 9 , Rio Acara, Acara

6 cf 3 9 , Rio Tapajoz, east bank

9 cf , Santarem (Carnegie Mus.)

22 cf , 8 9 , Rio Tapajoz, both banks (do.)

529. Dysithamnus mentalis emiliae Hellmayr

Type locality: S. Antonio do Prata, Brazil

Rio Capim (Goeldi); Quati-puru (Hellmayr); Peixe-Boi, Rio Guama, Rio
Tocantins (Snethlage)

530. Dysithamnus ardesiacus obidensis Snethlage

Type locality: Obidos, Brazil

Rio Jary and Obidos (Snethlage) ; Rio Jamunda, Faro (Zimmer)
14 a71 3 9 , Obidos (Carnegie Mus.)

531. Dysithamnus ardesiacus saturninus (Pelzeln)

Type locality: Borba, Rio Madeira

Rio Tapajoz, left bank (Snethlage and Zimmer); Villa Bella Imperatriz
(Zimmer)

16 cf 4 9 , Rio Tapajoz, Villa Braga and Apacy (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 235

The interrupted distribution of this genus in Lower Amazonia is of
interest. In particular the species mentalis is lacking between the Rio
Tocantins and northern Matto Grosso.

532. Thamnomanes caesius hoffmansi Hellmayr

Type locality: S. Antonio do Prata, Para

Para, Castanhal, Rio Inhangapy (Stone); Igarape-Assu and Peixe-Boi (Hell-
mayr); numerous localities near Para to Rio Tocantins (Snethlage); Rio
Xingu (Zimmer)

2 cf 1 9 , Para, Bosque
2 c? 1 9 , Rio Acara, Acara

533. Thamnomanes caesius persimilis Hellmayr

Type locality: Teffe, Rio Solimoes, Brazil

Rio Tapajoz, left bank (Hellmayr, Snethlage, Zimmer); birds variously inter-
mediate are recorded from the right bank of the Rio Tapajoz, Rio Jamau-
chim and Santarem (Hellmayr, Snethlage and Zimmer)
18 c? 15 9 , Rio Tapajoz, various localities, east bank
2 c? 7 9, Santarem (Carnegie Mus.)
5 c? 12 9 , Rio Tapajoz, both banks

This is much the largest series studied from this intermediate area.
The females in particular are clearly nearer persimilis than hoffmansi.

534. Thamnomanes caesius glaucus Cabanis

Type locality: Cayenne

Rio Jary and Obidos (Snethlage) ; Rio Jamunda, Faro and Obidos (Zimmer)
10 c? 8 9 , Obidos (Carnegie Mus.)

The great series in the Carnegie Museum do not endorse Hellmayr's
treatment of this group. Seven males and a female from Sao Paulo de
Olivenca, Rio Solimoes, are absolutely inseparable from Cayenne topo-
types, indicating that this form ranges much further west in northern
Brazil than the Rio Negro. Collected at the same place and on the
same day is a typical female of schistogynas Hellmayr, surely a curious
state of affairs, as this is also a notable range extension. Judging by
Zimmer's comments on his Peruvian material and his difficulties with
it, it begins to look as if schistogynus were specifically distinct.

236 bulletin: museum of comparative zoology

535. Myrmotherula brachyura brachyura (Hermann)

Type locality: Cayenne

Rio Jary and Obidos (Snethlage) ; Rio Tocantins, Rio Tapajoz and Rio Jamau-
chim (Snethlage); Rio Jamunda, Faro (Zimmer)
1 d71, Rio Tapajoz, Pinhy
10 d" 6 9 , Obidos (Carnegie Mus.)
9 c? 5 9 , Santarem (do.)
18 cf 9 9 , Rio Tapajoz, both banks (do.)

536. Myrmotherula sclateri Snethlage

Type locality: Boim, Rio Tapajoz, Brazil

Large series from the right bank of the same river (Zimmer) and others in the
Carnegie Museum

1 <?, Rio Tapajoz, Caxiricatuba

1 9 , Santarem (Carnegie Mus.)

1 d> 1 9 , Rio Tapajoz, Villa Braga and Aveiros (do.)

There are 7 c? 3 9 , from the Rio Purus (Carnegie Mus.), a notable
range extension.

537. Myrmotherula surinamensis surinamensis (Gmelin)

Type locality: Surinam

Obidos (Pinto); Rio Jamunda, Faro (Snethlage)

1 9 , Obidos (Carnegie Mus.)

538. Myrmotherula surinamensis multostriata Sclater

Type locality: Ucayali River, east Peru

Peixe-Boi and S. Antonio do Prata (Hellmayr) ; numerous, localities from Rio
Guama to Rio Tapajoz, (Snethlage) ; Villa Bella Imperatriz (Zimmer)

2 cf 3 9 , Rio Tapajoz, both banks (Carnegie Mus.)

539. Myrmotherula klagesi Todd

Type locality: Santarem, Brazil

Also islands in the Amazon near Obidos (Todd)

1 9 ad., near Obidos, Boca do IgarapeVPiaba

4 d\ Obidos (Carnegie Mus.)

3 c? 2 9 , Santarem (do.)

This "species" connects the last with cherriei Berlepsch and Hartert
of the upper Orinoco.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 237

540. Myrmotherula guttata (Vieillot)

Type locality: Cayenne

Rio Jary and Obidos (Snethlage)

16 cf 6 9 , Obidos (Carnegie Mus.)

541. Myrmotherula hauxwelli clarior Zimmer

Type locality: Rio Andira, west of the Rio Tapajoz, Brazil
Rio Tapajoz, both banks and Rio Jamauchim (Snethlage) ; same localities and
Villa Bella Imperatriz (Zimmer)

1 cf, Rio Tapajoz, Caxiricatuba

9 cT1 6 9 , Santarem (Carnegie Mus.)

4 c? 5 9 , Rio Tapajoz, both banks (do.)

542. Myrmotherula hauxwelli hellmayri Snethlage

Type locality; near Para, Brazil

Peixe-Boi, Igarape-Assu, S. Antonio do Prata (Hellmayr); Rio Capim (Wal-
lace) ; numerous localities, Para (Stone) to Rio Tocantins (Snethlage) ;
Rio Xingi (Zimmer)
3 9 , Para, Bosque
1 9 , Rio Acara, Acara
7 cf 7 9 , Benevides (Carnegie Mus.)

543. Myrmotherula gutturalis Salvin and Godman

Type locality: Bartica Grove, British Guiana
Rio Jary and Obidos (Snethlage); Faro (Zimmer)
6 d* 5 9 , Obidos (Carnegie Mus.)

This species represents erythrura of upper Amazonia.

544. Myrmotherula leucophthalma phaeonota Todd

Type locality: Villa Braga, left bank, Rio Tapajoz

Numerous localities, left bank Rio Tapajoz (Snethlage, Todd, Zimmer)

1 9 , Rio Tapajoz, Pinhel

12 cf 5 9 , Rio Tapajoz, left bank (Carnegie Mus.)

This Ant-Wren is sometimes considered to be a race of haematonota.

238 bulletin: museum of comparative zoology

545. Myrmotherula leucophthalma sordida Todd

Type locality: Colonia do Mojuy, Santarem, Brazil

Marajo Island, Rio Tocantins, Rio Xingu, Rio Jamauchim (Snethlage)
6 cf 8 9 , Santarem (Carnegie Mus.)
11 cf 14 9, Rio Tapajoz, east bank (M.C.Z.)
3(?79, (do.), (do.) (Carnegie Mus.)

546. Myrmotherula ornata hoffmannsi Hellmayr

Type locality: Itaituba, Rio Tapajoz, Brazil

Rio Tocantins to Rio Tapajoz (Snethlage, Zimmer, Pinto)

5 d" 4 9 , Rio Tapajoz, east bank

5 cf 5 9 , Santarem (Carnegie Mus.)

7 cf 9 9 , Rio Tapajoz, both banks

547. Myrmotherula axillaris axillaris (Vieillot)

Type locality: Cayenne

S. Antonio do Prata (Hellmayr) ; Rio Capim (Wallace) ; Peixe-Boi and Ipitinga
(Hellmayr); Santarem (Chapman and Riker; Hellmayr); Obidos (Hell-
mayr, Pinto) ; great series from entire area (Snethlage and Zimmer)

1 cf , Rio Acara, Acara

16 cf 6 9 , Rio Tapajoz, east bank

2 cf 2 9 , Obidos (Carnegie Mus.)

3 cf, Benevides (do.)
15 cf 6 9 , Santarem (do.)

13 cf 6 9, Rio Tapajoz, both banks (do.)

548. Myrmotherula longipennis longipennis Pelzeln

Type locality: Marabitanas, Rio Negro

San Antonio de Cachoeira, Rio Jary (Snethlage)

549. Myrmotherula longipennis ochrogyna Todd

Type locality: Villa Braga, left bank, Rio Tapajoz

Villa Bella Imperatriz (Zimmer); left bank of the Tapajoz (Snethlage and
Zimmer)

11 cf 5 9, type locality (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 239

550. Myrmotherula longipennis paraensis (Todd)

Type locality: Benevides, Para

Igarape-Assu, S. Antonio do Prata, Peixe-Boi (Hellmayr); Rio Capim (Wal-
lace); numerous localities Rio Guama and Para region to right bank of
Rio Tapajoz (Snethlage and Zimmer, Stone)
1 d\ Para, Bosque
3 cf 3 9 , Rio Tapajoz, east bank
8 cf 5 9 , Benevides (Carnegie Mus.)
8 d1 1 9 , Rio Tapajoz, east bank (do.)

551. Myrmotherula iheringi iheringi Snethlage

Type locality: Boim, left bank of Rio Tapajoz
Boim and Villa Braga (Snethlage)

1 d\ Villa Braga (ex Carnegie Mus.)

8 cT 4 9 , Villa Braga (do.)

This little known species is closely related to minor Salvadori of
southeastern Brazil and garbei Ihering of the Rio Jurua. It is other-
wise known only from the Rio Madeira and northern Matto Grosso.

552. Myrmotherula menetriesii cinereiventris Sclater & Salvin

Type locality: Cayenne

Rio Jary and Obidos (Snethlage) ; Rio Jamunda, Faro (Zimmer)
1 d1 2 9 , Obidos (Carnegie Mus.)

553. Myrmotherula menetriesii omissa Todd

Type locality: Benevides, Para

Igarape-Assu, S. Antonio do Prata, Peize-Boi (Hellmayr) ; Santarem (Chapman
and Riker); numerous localities, Rio Guama to Rio Jamauchim (Sneth-
lage); Rio Tapajoz, Tauary (Zimmer)

1 9 , Rio Acara, Acara

9 d1 6 9 , Rio Tapajoz, east bank

3 d1 2 9 , (do.) (do.) (Carnegie Mus.)
6 d1 2 9, Benevides (do.)

2 d" 1 9 , Santarem (do.)

554. Myrmotherula menetriesii berlepschi Hellmayr

Type locality: Salto do Girao, Rio Madeira

left bank of Rio Tapajoz (Snethlage and Hellmayr)

13 d1 5 9 , Rio Tapajoz, west bank (Carnegie Mus.)

240 bulletin: museum of comparative zoology

555. Myrmotherula assimils Pelzeln

Type locality: below Barcellos, Rio Negro

Santarem (Ihering); Rio Jamunda, Faro (Snethlage, Zimmer); Villa Bella
Imperatriz (Zimmer)

1 cf 1 9 , near Obidos

1 c? 1 9 , Obidos (Carnegie Mus.)

18 cf 13 9, Santarem (do.)

This rare species has a curious distribution, suggesting incomplete
knowledge. It is well represented from the Rio Negro and the Rio
Madeira, and turns up in northeastern Peru.

556. Dichrozona cincta zononota Ridgway

Type locality: Diamantina, Santarem, Brazil

Diamantina (Chapman and Riker); Villa Bella Imperatriz and Rio Tapajoz
(Zimmer)

1 d71 2 9 , Rio Tapajoz, Pinhy and Pataua

2 o\ Santarem (Carnegie Mus.)
15 c? 6 9, Rio Tapajoz (do.)

A rare antbird with a widely scattered distribution in upper Ama-
zonia.

557. Herpsilochmus sticturus sticturus Salvin

Type locality: Bartica Grove, British Guiana
2 cf, Obidos (Carnegie Mus.)
Previously unrecorded from Brazil.

558. Herpsilochmus rufimarginatus frater Sclater and Salvin

Type locality: Sarayacu, east Ecuador
Peixe-Boi and Marajo Island (Snethlage)

1 d\ Rio Tapajoz, Caxiricatuba

2 d1 1 9 , Benevides (Carnegie Mus.)

2 9 , Santarem (do.)

In default of topotypical material, we can only follow Hellmayr in
referring lower Amazonian specimens to frater. The bird is still very
rare in collections.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 241

559. Microrhopias quixensis bicolor (Pelzeln)

Type locality : Rio Madeira

Itaituba (Hellmayr); left bank of Rio Tapajoz (Snethlage and Zimmer);
Villa Bella Imperatriz (Zimmer)

29 <? 22 9 , Rio Tapajoz, left bank (Carnegie Mus.)

560. Microrhopias quixensis emiliae Chapman

Type locality : Alta Mira, Rio Xingu

Rio Tocantins (Chapman) ; Rio Tapajoz, right bank (Zimmer)

2 cf 3 9 , Santarem (Carnegie Mus.)

4 d1 3 9 , Rio Tapajoz, right bank (do.)

561. Microrhopias quixensis microsticta (Berlepseh)

Type locality: Rio Approuague, French Guiana
Rio Jary and Arumanduba (Snethlage)

562. Formicivora grisea grisea (Boddaert)

Type locality : Cayenne

Para (Layard, Natterer, Stone); Castanhal (Stone); S. Antonio do Prata
(Hellmayr); Rio Capim (Goeldi); Santarem (Chapman and Riker; Hell-
mayr, Pinto) ; numerous localities, Para, Rio Guama and Marajo Island to
left bank of Rio Tocantins, Amapa, and Monte Alegre on north bank
(Snethlage)

4 o* 2 9 , Para, Val-de-Caes

8 cf 10 9 , Rio Tapajoz, east bank
7 d1 3 9 , Benevides (Carnegie Mus.)
7 cT 9 9, Santarem (do.)

5 cf 4 9 , Rio Tapajoz (do.)

563. Formicivora rufa chapmani Cherrie

Type locality: Altar do Chao, Rio Tapajoz

Santarem (Chapman and Riker; Pinto); Rio Acara, Monte Alegre, Serra de
Erere, Rio Maecuru (Snethlage)

5 d> 4 9 , Rio Tapajoz, Santarem
12 c? 7 9 , Santarem (Carnegie Mus.)

Birds from the north bank of the Amazon in the Museu Goeldi may
represent an undescribed form.

242 bulletin: museum of comparative zoology

564. Drymophila devillei subochracea Chapman

Type locality: Rio Curua, lower Rio Xingu
Known only from the 9 type

565. Terenura spodioptila elaeopteryx Leverkuhn

Type locality: Cayenne

S. Antonio da Cachoeira, Rio Jary (Snethlage); Rio Jamunda, Faro (Zimmer)
2 d71 1 9 , Obidos (Carnegie Mus.)

566. Terenura spodioptila meridionalis Snethlage

Type locality: Villa Braga, Rio Tapajoz
Rio Tapajoz, left bank (Zimmer)

2 d" 2 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

567. Cercomacra cinerascens immaculata Chubb

Type locality: Supenaam, British Guiana
Rio Jamunda, Faro (Zimmer)

13 c? 4 9 , Obidos (Carnegie Mus.)

568. Cercomacra cinerascens iterata Zimmer

Type locality: Caxiricatuba, Rio Tapajoz

Igarape-Assu, S. Antonio do Prata, Santarem (Hellmayr); Rio Guama, Rio
Capim, Rio Tocantins, Rio Jamauchim, Rio Tapajoz left bank (Sneth-
lage) ; Rio Tocantins and Rio Tapajoz (Zimmer)
2 d\ Rio Tapajoz west bank, Pinhel
6 a71 2 9 , Rio Tapajoz east bank, various localities

2 d", Benevides (Carnegie Mus.)

3 d" 4 9 , Santarem (do.)

17 d1 8 9 , Rio Tapajoz, both banks (do.)

569. Cercomacra tyrannina laeta Todd

Type locality: Benevides, Para

Para (Layard, Natterer, Stone); Igarape-Assu, S. Antonio do Prata, Peixe-
Boi, Ipitinga, Obidos (Hellmayr); numerous localities, Rio Guama, Para
region to Rio Tocantins, and whole of north bank to Rio Jamunda (Sneth-
lage)

6 d71 5 9 , Para, Bosque, and Val-de-Caes
8 d" 11 9, Obidos (Carnegie Mus.)
11 d71 4 9, Benevides (do.)

GRISCOM AND GREEN WAY: BIRDS OF LOWER AMAZONIA 243

570. Cercomacra nigrescens (Cabanis and Heine)

Type locality: Cayenne

Rio Jamunda, Faro (Snethlage)

1 d\ Obidos (Carnegie Mus.)

571. Cercomacra nigrescens approximans Pelzeln

Type locality: Engenho do Gama, Matto Grosso

Santarem (Hellmayr); Rio Tocantins, Rio Tapajoz, left bank (Snethlage);
Villa Bella Imperatriz and other localities (Zimmer)

10 cf 8 9 , Rio Tapajoz, east bank

12 cT 7 9, (do.), both banks (Carnegie Mus.)

10 d" 5 9, Santarem (do.)

572. Pyriglena leucoptera leuconota (Spix)

Type locality: Para, Brazil

Para (Spix, Layard, Natterer, Wallace, Stone); Souza, Igarap6-Assu, Peixe-

Boi, S. Antonio do Prata (Hellmayr); Murutucu (Munich Museum);

numerous localities, Rio Guama, Para region to Rio Tocantins and Rio

Curua, (Snethlage) ; Utinga (Pinto)

7 d" 4 9 , Pard, Val-de-Caes

1 9 , Rio Acard, Acara
3 d\ Rio Tocantins, Cametd,
12 d" 10 9 , Benevides (Carnegie Mus.)

573. Pyriglena leucoptera similis Zimmer, 1931

Type locality: Caxiricatuba, Rio Tapajoz

2 d* 5 9 , Rio Tapajoz, Santarem

3 d1 1 9 , (do.), both banks (Carnegie Mus.)
9 d" 3 9 , Santarem (do.)

574. Myrmoborus letjcophrys subsp.

Itaituba (Hellmayr); Rio Tocantins and Rio Jamauchim (Snethlage and
Zimmer)

8 d1 3 9 , Rio Tapajoz, both banks (Carnegie Mus.)

Zimmer's study of this species (Amer. Mus. Novit., no. 545, 1932'
pp. 1-5) lists a "subsp. ?" of which he had been able to examine only
2 specimens from the Rio Tocantins and 1 from the Rio Jamauchim.
His griseigula was based on a series from the left bank of the Rio

244 bulletin: museum of comparative zoology

Madeira. The adult male is closest to angustirostris, merely averaging
a little darker on the belly. The females, however, differ in just the
respects ascribed by Zimmer to griscigula, but the under tail-coverts
are almost pure white instead of buffy brown. We have here a rather
poorly characterized intermediate, which Mr. Todd does not care to
describe. With no material of our own we are unable to do so either,
but somebody will most certainly do so in the future.

575. Myrmoborus leucophrys angustirostris (Cabanis)

Type locality: coastal forestsof British Guiana, Cunany, Rio Jary, Rio Maecuru

(Snethlage)

576. Myrmoborus lugubris lugubris (Cabanis)

Type locality: Para, Brazil

Santarem (Chapman and Riker) ; Monte Alegre, Obidos, Rio Jamunda (Sneth-
lage) ; Rio Tocantins, Rio Xingu, Villa Bella Imperatriz (Zimmer)
1 d\ Para, Val-de-Caes

6 cf 4 9 , north bank of Amazon near Obidos
1 d", Rio Tapajoz, Santarem

7 d% Obidos (Carnegie Mus.)
16 cf 13 9, Santarem (do.)

577. Myrmoborus myotherinus ochrol.ema (Hellmayr)

Type locality: Itaituba, left bank, Rio Tapajoz

Rio Tocantins, Rio Xingii, Rio Jamauchim, Rio Tapajoz (Snethlage and
Zimmer)

4 c\ Para, Val-de-Caes

8 cf 4 9 , Rio Tapajoz, east bank

1 9 , Rio Tapajoz, west bank (Pinhel)
19 cf 12 9 , Santarem (Carnegie Mus.)
15 cf 9 9 , Rio Tapajoz, both banks (do.)

We cannot find that this species has been reported east of the Rio
Tocantins. The four Para males are minutely paler below, but unfor-
tunately we have no females.

578. Hypocnemis cantator cantator (Boddaert)

Type locality: Cayenne

Obidos, (Hellmayr, Snethlage, Zimmer, Pinto); Rio Jamunda, Faro (Sneth-
lage and Zimmer)

2 cf 1 9 , Obidos (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 245

579. Hypocnemis cantator affinis Zimmer, 1932

Type locality: Baiao, Rio Tocantins

Rio Tocantins and Rio Xingu, right bank (Snethlage, Zimmer)

580. Hypocnemis cantator striata (Spix)

Type locality: Santarem

Rio Jamauchim, Rio Tapajoz, both banks (Snethlage and Zimmer)

12 c? 11 9 , 2 ? , Rio Tapajoz, both banks

7 d 3 9 , Santarem (Carnegie Mus.)

9 cf 5 9 , Rio Tapajoz, both banks (do.)

581. Hypocnemis cantator implicata Zimmer, 1932

Type locality: near Borba, Rio Madeira
Villa Bella Imperatriz (Zimmer)

582. Hypocnemis hypoxantha ochraceiventris Chapman

Type locality: Alta Mira, Rio Xingii
Rio Tapajoz, east bank, (Carnegie Mus. and Zimmer)
5 d 4 9 , Rio Tapajoz, east bank
15 d 8 9 , Rio Tapajoz, (do.), (Carnegie Mus.)

583. Hypocnemoides melanopogon melanopogon (Scalter)

Type locality: Guiana

North bank, Rio Jary, Arumanduba, Obidos, Rio Jamunda (Snethlage);
Mexiana Island (Wallace, Hellmayr); Rio Tocantins (Snethlage, Zimmer);
Rio Xingii, Rio Tapajoz, Villa Bella Imperatriz (Ziinmer)
3 d , Para, Val-de-Caes
2 d 1 9 , near Obidos
1 d 2 9 , Obidos (Carnegie Mus.)
5 d 4 9 , Santarem (do.)

Snethlage and Hellmayr have both commented on the peculiarly
overlapping ranges of melanopogon and maculicauda. Recent collec-
tions show that there is little or nothing in this ; the Para specimens of
melanopogon listed above now show that species to have a continuous
range on the south bank of the Amazon.

246 bulletin: museum of comparative zoology

584. Hypocnemoides maculicauda (Pelzeln)

Type locality : Matta Grosso

Rio Capim (Goeldi); Pard, S. Antonio do Prata, Rio Acard, Rio Iriri, Rio
Jamauchim, Rio Tapajoz (Snethlage)

2 d" 1 9 , Rio Tapajoz, east bank

7 cf 8 9 , (do.) both banks (Carnegie Mus.)

It will be noted that both species are now known from the Para
region and the east bank of the Tapajoz.

585. Percnostola rufifrons rufifrons (Gmelin)

Type locality: Cayenne

San Antonio do Cachoeira, Rio Jary, Obidos (Snethlage, Hellmayr, Zimmer,
Pinto)

10 d" 9 9 , Obidos (Carnegie Mus.)

586. Percnostola rufifrons subcristata Hellmayr

Type locality: Manaos

Rio Jamunda, Faro, (Snethlage, Zimmer)

587. Sclateria naevia naevia (Gmelin)

Type locality: Surinam

Para, Peixe-Boi, Rio Acard (Hellmayr); Pard, S. Antonio do Prata, Ilha das
Oncas, Rio Acard (Snethlage); Pard (Stone)

The series from Obidos are perfect intermediates between this and
the next.

588. Sclateria naevia toddi Hellmayr

Type locality: Santarem, Brazil

1 9 , Rio Tapajoz, Tauary

3 cf 2 9 , Obidos (Carnegie Mus.)
5 o" 5 9, Santarem (do.)

1 d\ Rio Tapajoz, Miritituba (do.)

589. Schistocichla leucostigma leucostigma (Pelzeln)

Type locality: Manaos
Obidos (Hellmayr, 1929)

1 d", Obidos (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 247

590. Schistocichla leucostigma rufifacies Hellmayr

Type locality: Apacy, Rio Tapajoz, Brazil

Arumatheua, Rio Tocantins (Snethlage); Santarem and Villa Braga (Hell-
mayr) ; Villa Imperatriz and Serra do Parintins (Zimmer)

2 o" 1 9 , Rio Tapajoz, east bank

1 9 , Santarem (Carnegie Mus.)

3 cf 2 9 , Rio Tapajoz, left bank, including type

591. Myrmeciza longipes griseipectus Berlepsch & Hartert

Type locality: Caicara, Venezuela

Obidos (Hellmayr, Pinto); Monte Alegre, Erere, Rio Maecuru (Snethlage)

1 d" imm., near Obidos

24 d" 11 9, Obidos (Carnegie Mus.)

592. Myrmeciza ferruginea ferruginea (P.L.S. Miiller)

Type locality: Cayenne

Cunany, Rio Jary, Obidos, Rio Jamunda (Snethlage); Obidos (Hellmayr,
Pinto)

3 o71 1 9 Obidos (Carnegie Mus.)

593. Myrmeciza ferruginea eluta (Todd)

Type locality: Villa Braga, Rio Tapajoz

9 o71 6 9 , Rio Tapajoz, left bank (Carnegie Mus.)

Only known from the type locality, where first found by Snethlage,
to the right bank of the Rio Madeira.

594. Myrmeciza atrothorax atrothorax (Boddaert)

Type locality: Cayenne

Rio Jamunda, Faro (Snethlage); Marajo Island (Snethlage)

The subspecies of the Marajo Island bird still remains to be deter-
mined.

595. Myrmeciza atrothorax melanura (Menetries)

Type locality: Cuyaba, Matto Grosso

Igarape Brabo, left bank, Rio Tapajoz, 1 d71 (Zimmer)

248 bulletin: museum of comparative zoology

596. Myrmeciza atrothorax stictothorax (Todd)

Type locality: Apacy, Rio Tapajoz

1 c? 1 9 , Rio Tapajoz, Apacy (Carnegie Mus.)

Only known from the two original specimens. As Zimmer very justly
remarks, more material from the Tapajoz is needed.

597. Myrmeciza hemimelaena pallens Berlepsch and Hellmayr

Type locality: Villa Bella de Matto Grosso

Rio Xingu, Cussary, Tamucury, Rio Tapajoz (Snethlage)

16 d1 10 9 , Rio Tapajoz, east bank
5 cf 5 9, (do.) both banks (Carnegie Mus.)

12 cf 11 9, Santarem (do.)

598. (Myrmeciza dubia Snethlage

Type locality: Rio Iriri (tributary of the Xingu)

Snethlage, in her original description (1925), places this species in
the genus Myrmeciza with a good deal of doubt. She seems to have
been under the impression that Drymophila Swainson is a synonym
of Myrmeciza G. R. Gray. The type, which we suppose is in the
Goeldi Museum at Para, should be examined.

599. Formicarius colma colma Boddaert

Type locality: Cayenne

Rio Jary and Obidos (Snethlage) ; Rio Jamunda, Faro (Zimmer)

2 cf 3 9 , Obidos (Carnegie Mus.)

600. Formicarius colma amazonicus Hellmayr

Type locality: Borba, Rio Madeira

Para (Natterer, Snethlage); Igarape-assu, S. Antonio do Prata, Peixe-Boi,
Ipitinga, (Hellmayr); Benevides, Sta. Isabel, Rio Tocantins, Rio Iriri,
Rio Jamauchim, Rio Tapajoz, (Snethlage); Rio Tocantins, Rio Tapajoz,
Villa Bella Imperatriz (Zimmer)

4 c? 2, Rio Tapajoz, both banks
13 cf 5 9 , Santarem (Carnegie Mus.)
2 9 , Rio Tapajoz, left bank (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 249

601. Formicarius analis analis (Lafresnaye and D'Orbigny)

Type locality: Yuracares, Bolivia

Para (Wallace, Stone) ; Rio Guama (Stone) ; Peixe-Boi (Hellmayr) ; Provi-
dencia, Benevides, Maguary, Rio Guama, Rio Acara, Rio Tocantins, Rio
Tapajoz (Snethlage); Villa Bella Imperatriz (Zimmer)
4 cf , Para, Bosque, and Val-de-Caes
1 cf , Rio Acara, Acara
1 cf , Rio Tapajoz, Caxiricatuba
8 cf 3 9 , Benevides (Carnegie Mus.)

6 cf 4 9 , Santarem (do.)

3 cf , Rio Tapajoz, both banks

602. Formicarius analis crissalis (Cabanis)

Type locality: Roraima, British Guiana

Monte Alegre (Snethlage), the identification requiring confirmation

603. Chamaeza nobilis fulvipectus Todd

Type locality: Colonia do Mojuy, Santarem, Brazil
The type in Carnegie Museum examined.

604. Pithys albifrons albifrons (Linnaeus)

Type locality: Cayenne

Obidos, Rio Jamunda, Faro (Snethlage and Zimmer)

4 cf 2 9 , Obidos (Carnegie Mus.)

605. Gymnopithys rufigula rufigula (Boddaert)

Type locality: Cayenne

Obidos and Rio Jamunda, Faro (Snethlage)

7 cf 4 9 , Obidos (Carnegie Mus.)

606. Rhegmatorhina gymnops Ridgway

Type locality: Diamantina, near Santarem, Brazil

Santarem (Chapman and Riker); Rio Curua, Rio Tapajoz, Rio Jamauchim
(Snethlage); Rio Tapajoz (Pinto)

1 cf 1 9 , Miritituba and Colonia de Mojuy, Rio Tapajoz

2 cf 4 9 1 ?, (do.) (Carnegie Mus.)

3 cf 3 9 , Santarem (Carnegie Mus.)

250 bulletin: museum of comparative zoology

607. Rhegmatorhina berlepschi (Snethlage)

Type locality: Villa Braga, left bank, Rio Tapajoz
Boim (Snethlage)

lo"l 9, Villa Braga

10 d" 10 9 , (do.) (Carnegie Mus.)

608. Hylophylax naevia theresae (Des Murs)

Type locality: Rio Javari, northeast Peru
Villa Braga, left bank of Rio Tapajoz (Hellmayr)
8 d" 9 9 , Rio Tapajoz, left bank (do.)

Recorded by Snethlage as the next race. This locality marks the
easternmost point in the range of theresae.

609. Hylophylax naevia ochracea (Berlepsch)

Type locality: Tucunare, Rio Jamauchim

Rio Tocantins, Rio Xingu, Rio Tapajoz, right bank (Snethlage)
5 d\ Rio Tapajoz, right bank

14 J1 6 9 , (do.) (do.) (Carnegie Mus.)
1 9 , Santarem (do.)

610. Hylophylax punctulata subochracea Zimmer, 1934

Type locality: Limoal, left bank, Rio Tapajoz

Rio Curua (Snethlage, fide Hellmayr) ; Rio Tapajoz, both banks and Rio Xingu,
Tapara (Zimmer)

5 d\ Rio Tapajoz, both banks (Carnegie Mus.)

3 d71 1 9 , Santarem (do.)

611. Hylophylax poecilonota poecilonota (Cabanis)

Type locality: Cayenne

Obidos, (Snethlage); Rio Jamunda, Faro (Zimmer)
8 C? 2 9 , Obidos (Carnegie Mus.)

612. Hylophylax poecilonota nigrigula (Snethlage)

Type locality: Boim, Rio Tapajoz

Villa Bella Imperatriz and Rio Tapajoz, both banks (Zimmer)
9 cf 7 9 , Rio Tapajoz, east bank

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 251

613. Hylophylax poecilonota vidua (Hellmayr)

Type locality: Igarape-Assu, Rio Acara, Pard

Pard, Peixe-Boi, S. Antonio do Prata (Hellmayr); numerous localities, Rio
Guamd and Pard (Stone) to Rio Tocantins (Snethlage); Rio Xingu and
elsewhere (Zimmer) ; Rio Capim, Rio Inhangapy (Stone)

5 c? 7 9 , Pard, Bosque and Val-de-Caes
2 cf, Rio Acara, Acara

10 c? 13 9 , Benevides (Carnegie Mus.)

614. Phlegopsis nigro-maculata bowmani Ridgway

Type locality: Diamantina, Santarem, Brazil

Santarem (Chapman and Riker); Rio Curud, Rio Jamauchim, Rio Tapajoz

(both banks) (Snethlage); Villa Bella Imperatriz and both banks of Rio

Tapajoz (Zimmer)

6 d1 3 9 , Rio Tapajoz, east bank
15 d* 9 9 , Santarem (Carnegie Mus.)

9 cT 4 9 , Rio Tapajoz, west bank (do.)

615. Phlegopsis nigro-maculata confinis Zimmer

Type locality: Tapara, Rio Xingu

Only known from the banks of the Rio Xingu (Zimmer)

Snethlage, 1926, comments on this subspecies on p. 55, and predicts
that the bird between the Xingu and Rio Tocantins will prove separ-
able.

616. Phlegopsis nigro-maculata paraensis Hellmayr

Type locality: Para

Para (Natterer, Wallace, Stone); Murutucu, Igarape-Assu (Hellmayr); Rio
Capim (Goeldi) ; Rio Guamd, vicinity of Para and Rio Tocantins (Sneth-
lage) ; Rio Guama to Rio Tocantins (Zimmer) ; Murutucu (Pinto)
6 cf 6 9 , Benevides (Carnegie Mus.)

617. Phlegopsis borbae Hellmayr

Type locality: Borba, Rio Madeira

Villa Braga, left bank of Rio Tapajoz (Hellmayr), 1 d1 ad. in Carnegie Museum,
examined by us

252 bulletin: museum of comparative zoology

618. Myrmornis torquata (Boddaert)

Type locality: Cayenne

Peixe-Boi, Rio Tocantins, Cussary and Rio Tapajoz (Snethlage); Santarem
(Allen) ; Aveiro (Pinto)

1 9 , Rio Tapajoz, Caxiricatuba

2 d> 2 9 , Obidos (Carnegie Mus.)
1 d" 1 9 , Benevides (do.)

4 c? 4 9 , Santarem (do.)

4 cf 4 9 , Rio Tapajoz, both banks (do.)

619. Myrmothera campanisona campanisona (Hermann)

Type locality: Cayenne

Obidos (Snethlage); Rio Jamunda, Faro (Zimmer)

1 ?, near Obidos

7 cf 4 9 , Obidos (Carnegie Mus.)

620. Myrmothera campanisona subcanescens Todd

Type locality: Colonia de Mojuy, Santarem, Rio Jamauchim (Snethlage);
both banks of Rio Tapajoz (Todd); Aveiro (Pinto)
1 c? 6 9 , Rio Tapajoz, east bank
13 c? , 3 9, (do.), both banks (Carnegie Mus.)
6 d1 1 9 , Santarem (do.)

621. Grallaria varia distincta Todd

Type locality: Villa Braga, Rio Tapajoz

1 d\ Santarem

4 d>, Rio Tapajoz, Villa Braga, Apagy, and Itaituba (Carnegie Mus.)

This rare bird is otherwise reported only from Calama, Rio Madeira.

622. Grallaria berlepschi Hellmayr

Type locality: Engenho do Gama, western Matto Grosso
Cussary (Snethlage) ; Santarem (Todd, in Carnegie Museum)

2 cf 2 9 , Rio Tapajoz, east bank
20 d" 5 9 , Santarem

1 671, Rio Tapajoz, Miritituba (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 253

623. Grallaria macularia macularia (Temminck)

Type locality: Cayenne

1 9 , Obidos (Carnegie Mus.)

The first definite record for Brazil.

624. Grallaria macularia paraensis Snethlage

Type locality: Ourem, Rio Guama, Para, Brazil

Also from Rio Acara and Rio Jamauchim (Snethlage)

1 cf, Miritituba, Rio Tapajoz (ex Carnegie Mus.)

3 cf 2 9 , Santarem (Carnegie Mus.)

3 cf, Rio Tapajoz, Miritituba and Villa Braga (do.)

The specimens from Villa Braga have distinct shaft streaks as in
typical macularia.

Family CONOPOPHAGIDAE

625. Conopophaga aurita aurita (Gmelin)

Type locality: Cayenne

Rio Jamunda, Faro (Zimmer)

626. Conopophaga aurita pallida Snethlage

Type locality: Cameta, left bank of Rio Tocantins
Only the three original specimens are recorded.

627. Conopophaga aurita snethlage.e Berlepsch

Type locality: Tucunarc. Rio Jamauchim

Cussary and type locality (Snethlage); Rio Tapajoz, both banks (Zimmer,
Pinto)

9 cf , Rio Tapajoz, various localities, east bank

4 cf 2 9 , Santarem (Carnegie Mus.)

12 cT 10 9 , Rio Tapajoz, both banks (do.)

254 bulletin: museum of comparative zoology

628. Conopophaga melanogaster Menetries

Type locality: Rio Maderia

Rio Tocantins and Rio Tapajoz, left bank (Snethlage)
1 cf , Rio Tapajoz, Boim

1 cf 1 9 , Rio Tapajoz, Villa Braga (ex Carnegie Mus.)
14 cf 6 9 , Rio Tapajoz, Apagy, Villa Braga, Itaituba (Carnegie
Mus.)

629. Conopophaga roberti Hellmayr

Type locality: Igarape-Assu, Para, Brazil

Peixe-Boi, Ipitinga, S. Antonio do Prata (Hellmayr); numerous localities near

Para to Rio Tocantins (Snethlage); Para and Castanhal (Stone); Prata

(Pinto)

1 cf 2 9 , Para, Bosque

2 cf 1 9 , Rio 'Acara, Acara

9 cf 3 9 , Benevides (Carnegie Mus.)

630. Corythopis torquata anthoides (Pucheran)

Type locality: Cayenne

Para (Wallace, Stone); Castanhal (Stone); Igarape-Assu, S. Antonio do Prata
(Hellmayr); Providencia, Ananindeua, Obidos (Snethlage)
1 9 , Rio Acara, Acara
3 cf , Rio Tapajoz, Pinhy, Caxiricatuba
9 cf 2 9 , Obidos (Carnegie Mus.)

1 cf , Benevides (do.)

2 cf , Santarem (do.)

7 cf 2 9 , Rio Tapajoz, both banks (do.)

Family RHINOCRYPTIDAE
631. Liosceles thoracicus thoracicus (Sclater)

Type locality: left bank of Rio Madeira

Villa Bella Imperatriz (Zimmer); Rio Tapajoz, Villa Braga (Snethlage)

1 cf 1 9 , Villa Braga, left bank, Rio Tapajoz (ex Carnegie Museum)
13 cf 4 9 , (do.) Carnegie Mus.)

This rare bird has only recently been found east of the Rio Madeira.
Zimmer has shown that Peruvian records should be transferred to
erithacus, leaving true thoracicus with an exceedingly restricted range.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 255

Family COTINGIDAE

632. Phoenicircus carnifex (Linnaeus)

Type locality: Surinam

Para, (Natterer, Wallace) ; Ipitinga (Hellmayr) ; Rio Guama, Para region and
Rio Tocantins (Snethlage); Santarem and Villa Braga, Rio Tapajoz
(Hellmayr, in Carnegie Museum)

2 c? 2 9 , Rio Tapajoz, Caxiricatuba
1 9 , Obidos (Carnegie Mus.)

1 9 , Benevides (do.)
5 c? 4 9 , Santarem (do.)

1 cf 1 9 , Rio Tapajoz, Villa Braga (do.)

633. Phoenicircus nigricollis Swainson

Type locality: Barcellos, Rio Negro, Brazil

Villa Braga and Miritituba, Rio Tapajoz (Hellmayr) in Carnegie Museum;
1 cf , 1 9 , examined by us

2 9 , Rio Tapajoz, Tauary and Caxiricatuba

These two species apparently occur together on the Rio Tapajoz,
carnifex primarily northern and eastern, nigricollis upper Amazonian.

634. Cotinga cotinga (Linnaeus)

Type locality: "Brazil"

Para (Natterer, Wallace) ; Providencia (Snethlage) ; Santarem (Chapman and
Riker) ; series from vicinity of Para in Museo Paulista (Cunha Vieiro)

635. Cotinga cayana (Linmeus)

Type locality : Cayenne

Para (Natterer, Stone); Peixe-Boi and Ipitinga (Hellmayr); Rio Inhangapy

(Stone); Para region, Rio Acara, Rio Tocantins, Rio Tapajoz, (Snethlage);

Santarem (Chapman and Riker)

1 cf 3 9 , Obidos (Carnegie Mus.)

2 9 , Benevides (do.)

636. Xipholena punicea (Pallas)

Type locality: Surinam

Rio Jamundd, Faro (Snethlage)

2 cf, Obidos (Carnegie Mus.)

256 bulletin: museum of comparative zoology

637. Xipholena lamellipennis lamellipennis (Lafresnaye)

Type locality; Para, by subsequent designation

Para (Natterer, Wallace); Peixe-Boi, Ipitinga, Igarape-Assu (Hellmayr);
Pinheiro (Stone) ; Para region and Rio Tocantins (Snethlage)

4 d" 3 9 , Para, Val-de-Caes

4 J 2 9, Benevides (Carnegie Mus.)

638. Xipholena lamellipennis pallidior Griscom & Greenway

Rio Tapajoz, Boim (Snethlage)

4 oM 9 , Rio Tapajoz, Pinhy, and Santarem

1 d", Santarem (Carnegie Mus.)

2 cf 1 9 , Rio Tapajoz, both banks (do.)

The two species of the genus in our area have a somewhat unusual
distribution, as punicca the "northern" one, crosses the Amazon west
of our area and occurs along the Rio Madeira.

639. Iodopleura isabellae Parzudaki

Type locality: Rio Nigro, "in Venezuela"

Para (Layard); Marajo Island (Hellmayr); Providencia and Rio Tocantins
(Snethlage); Pinheiro (Stone); Obidos and Murutucu (Cunha Vieiro)
2 cf 1 9 1?, Benevides (Carnegie Mus.)

A little known bird, which reappears on the Rio Solimoes and Rio
Purus, and thence extends to eastern Colombia, Ecuador, and Peru.
Its relationship to other "species" in the genus still remains to be de-
termined.

640. Attila spadiceus spadiceus (Gmelin)

Type locality: Cayenne

Benevides, Providencia, Rio Tocantins, Rio Tapajoz, Cussary, Monte Alegre,
Obidos (Snethlage); Utinga (Beebe); Rio Tapajoz, numerous localities
(Hellmayr, series in Carnegie Museum); Santarem (Ridgway; Chapman
and Riker)

1 cf , Para, Bosque

2 cf 2 9 , Rio Tapajoz, east bank

2 9 , Benevides (Carnegie Mus.)
14 cf 8 9 , Santarem (do.)
4 d" 2 9 , Rio Tapajoz (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 257

641. Attila bolivianus nattereri Hellmayr

Type locality: Borba, Rio Madeira

Para (Natterer) ; Monte Alegre and Rio Jamunda, Faro (Snethlage)

2 cf 1 9 , near Obidos
10 c? 2 9 , Obidos (Carnegie Mus.)

6 d" 6 9, Santarem (do.)

642. Attila cinnamomeus cinnamomeus (Gmelin)

Type locality : Cayenne

Ipitinga (Hellmayr); Para (Layard); Mexiana Island (Spix); Rio Inhangapy
(Stone); Amapa (Goeldi); Itaituba (Hellmayr); Para region, Marajo
Island, Cussary, Rio Tapajoz, Rio Jary, Arumanduba, Monte Alegre,
Rio Jamunda (Snethlage); Caviana Island (Brodkorb)

1 d\ Rio Tapajoz, Pinhy

2 d\ Obidos (Carnegie Mus.)

3 d" 2 9, Santarem (do.)

3 d\ Rio Tapajoz, left bank (do.)

643. Casiornis rufa (Vieillot)

Type locality : Paraguay
Monte Alegre (Snethlage)

A campo species which is lacking in the Amazonian forests. Birds
from this isolated locality on the north bank of the Amazon might
prove separable.

644. Casiornis fusca Sclater and Salvin

Type locality: Bahia

Rio Muraitua (Stone); Para, S. Antonio do Prata, Rio Tocantins, Rio Xingii,
Rio Tapajoz (Snethlage)

1 9 , Rio Tapajoz, Pinhy

1 d\ Benevides (Carnegie Mus.)

1 cf 2 9 , Santarem (do.)

258 bulletin: museum of comparative zoology

645. Laniocera hypopyrrha (Vieillot)

Type locality:

Peixe-Boi and S. Antonio do Prata (Hellmayr); Para (Stone); Para region,
Rio Tocantins, Rio Tapajoz, Rio Jary, Obidos (Snethlage) ; large series
from both banks of the Amazon (Zimmer)
3 c? 2 9 , Rio Tapajoz, east bank
9 d\ Obidos (Carnegie Mus.)
1 d" , Benevides (do.)
3 cf, Santarem (do.)
7 d\ Rio Tapajoz, both banks (do.)

646. Rhytipterna simplex frederict (Bangs and Penard)

Type locality: Paramaribo, Surinam

Para (Natterer); S. Antonio do Prata, Igarape-Assu (Hellmayr); Santarem
(Chapman and Riker; Hellmayr); Para region, Rio Tocantins, Rio Tapa-
joz (Snethlage) ; Rio Jary, Obidos (Snethlage) ; Faro (Zimmer)
1 d\ Para, Bosque
1 c? 5 9 , Rio Tapajoz, east bank
6 d" 3 9 , Obidos (Carnegie Mus.)

2 9 , Benevides (do.)
5 cf 6 9 , Santarem (do.)

Our material brings out an interesting point in this connection. Mr.
Zimmer has recently (1936) described the bird from the south bank as
intermedia on very fine series. There would not appear, however, to be
justifiable grounds for regarding our own as inadequate. Only four
birds north of the Amazon out of thirty are appreciably paler and
greyer, while only five from the south bank are noticeably yellower on
the abdomen. Mr. Zimmer assigns all Amazonian birds to intermedia,
but 12 specimens before us from the Rio Purus and Rio Solimoes appear
instantly separable from Lower Amazon birds in being darker below.
Thus we would not have described the birds from the south bank,
and would most certainly have separated the others(!) were it not
for the fact that Zimmer's material does not endorse it. If the two
series were combined, the sensible course would probably be to follow
Hellmayr in calling everything frederici.

647. Rhytipterna immunda (Sclater and Salvin)

Type locality: Oyapock, Cayenne, probably erroneous
1 cT from Santarem (Zimmer)

An exceedingly rare species, otherwise known from the Cassiquiare
and the middle Rio Negro.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 259

648. Lipaugus cineraceus (Vieillot)

Type locality: Cayenne

Numerous records throughout our area by all collectors
1 cf, Rio Acara, Acara
18 cf 9 9 , Rio Tapajoz, east bank

1 cf 1 9 , Obidos (Carnegie Mus.)

2 cf, Benevides(do.)

5 cf 2 9 , Santarern (do.)

649. Pachyramphus viridis griseigularis Salvin & Godman

Type locality: Mt. Roraima, British Guiana
Marajo Island (Snethlage, 1926)

1 c? 1 9 , Rio Tapajoz, Pataua, June 26 and 27, a" breeding

This species is almost unknown from the lower Amazon. Our pair
are strikingly different from typical viridis, but agree minutely with
Hellmayr's comparative critique of griseigularis. Needless to say,
further material from the Amazon is required to settle the status of the
local form there.

650. Pachyramphus surinamus (Linna?us)

Type locality: Surinam
Obidos (Hellmayr)

1 <? , Obidos (Carnegie Mus.)

651. Pachyramphus rufus (Boddaert)

Type locality: Cayenne

Ubiquitous in our area, including Mexiana and Marajo Islands

2 cf , near Obidos

4 c? 3 9 , Rio Tapajos, east bank

2 c? 4 9 , Obidos (Carnegie Mus.)

2 cf 3 9 , Santarern (do.)

1 cf 2 9 , Rio Tapajoz, left bank Xdo.)

652. Pachyramphus castaneus subsp.

Rio Tocantins, Baiao, 1 c? (Zimmer)

1 <?, Rio Tapajoz, Caxiricatuba

For the status of these two specimens cf . Zimmer, Amer. Mus. Novit.,
no. 894, 1936, p. 8.

260 bulletin: museum of comparative zoology

653. Pachyramphus castaneus amazonum Zimmer

Type locality: Rosarinho, Rio Madeira, Brazil

Rio Jamunda, Faro, Obidos, Monte Alegre (Snethlage) ; Monte Alegre (Zimmer)
1 cf, Obidos
17 c? 15 9, (do.) (Carnegie Mus.)
3o"2 9, Santarem (do.)

654. Pachyramphus polychropterus tristis (Kaup)

Type locality: Cayenne

Mexiana Island (Sclater and Salvin) ; Marajo Island, and numerous localities
north bank of Amazon from Obidos eastward (Snethlage and Zimmer)

2 d1, near Obidos

6 c? 4 9 , Obidos (Carnegie Mus.)

655. Pachyramphus polychopterus niger (Spix)

Type locality: Fonteboa, Amazonas
Villa Bella Imperatriz and Faro (Zimmer)

656. Pachyramphus polychopterus polychopterus (Vieillot)

Type locality: Bahia, Brazil

? Para region (Sclater and Snethlage) ; Rio Tocantins (Snethlage and Zimmer)
to the left bank of the Rio Tapajoz (Zimmer)
3 <? 4 9 , Rio Tapajoz, east bank

1 9 , Benevides (Carnegie Mus.)
9 c? 7 9 , Santarem (do.)

An exceedingly variable series, one male from the Tapajoz very grey,
others approaching niger, while most of the birds from the north bank
are the blackest of all.(!) The three races listed here are a very unsatis-
factory assemblage of variable intermediates in most of the range cur-
rently assigned to them, and are distinguishable only in very large
series in the proportion of darker versus lighter birds.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 261

657. Pachyramphus marginatus nanus Bangs and Penard

Type locality: Xeberos, Peruvian Amazon

Santarem (Chapman and Riker); Para (Stone); Utinga (Beebe); numerous
localities Rio Guama, Para region and all rivers on the south bank to the
Rio Tapajoz, and various north bank localities (Snethlage)
1 9 , Rio Tapajoz, Santarem
7 cf 9 9 , Obidos (Carnegie Mus.)

1 cf 2 9 , Benevides (do.)

2 cf 5 9 , Santarem (do.)

12 cf 11 9 , Rio Tapajoz, both banks (do.)

658. Platypsaris rufus rufus (Vieillot)

Type locality: Paraguay

Mocajuba, Rio Tocantins; Para and Marajo Island (Zimmer)

The last two birds originally recorded by Snethlage as minor.

659. Platypsaris minor (Lesson)

Type locality: Cayenne

Para (Natterer, and Stone) ; Santarem (Chapman and Riker) ; Ipitinga (Hell-
mayr); Para region, Rio Tocantins and Rio Tapajoz (Snethlage); Obidos,
Rio Tapajoz and Santarem (Hellmayr)
1 cf 1 9 , Obidos (Carnegie Mus.)
6 cf 4 9 , Santarem (do.)

1 9 , Rio Tapajoz, Villa Braga (do.)

660. Tityra cayana cayana (Linnaeus)

Type locality: Cayenne

Para (Natterer, Wallace, Stone); Peixe-Boi, Ipitinga, S. Antonio do Prata,
Igarape-Assu (Hellmayr) ; Santarem (Chapman and Riker) ; Para region,
Rio Tocantins, and north bank localities (Snethlage); Marajo Island
(Brodkorb)

1 9 , Rio Acara, Acara
3 cf 3 9 , Obidos (Carnegie Mus.)

1 9 , Benevides (do.)

2 cf , Santarem (do.)

262 bulletin: museum of comparative zoology

661. Tityra semifasciata semifasciata (Spix)

Type locality: Para

Para (Spix, Cabanis); Mexiana Island (Hagmann); Rio Muria (Natterer);
Marajo Island, Rio Tocantins to Rio Tapajoz, and north bank localities
(Snethlage); numerous localities throughout (Zimmer)
1 9 , Rio Tapajoz, Tauary
2 d" 2 9 , Obidos (Carnegie Mus.)
4 cf 1 9 , Rio Tapajoz, both banks (do.)
11 a* 8 9, Santarem (do.)

It still remains to be determined how these two species divide the
territory between them in places where they are both recorded.

662. Tityra inquisitor pelzelni Salvin and Godman

Type locality: Matto Grosso, Brazil

Peixe-Boi, S. Antonio do Prata (Hellmayr); Santarem (Chapman and Riker,
Zimmer); Utinga (Beebe); Villa Bella Imperatriz (Zimmer)

1 9 , Santarem (Carnegie Mus.)

663. Tityra inquisitor erythrogenys (Selby)

Type locality: corrected to Cayenne

Rio Jamunda, Faro (Snethlage and Zimmer)

This rare species has undoubtedly been overlooked in our area. The
species should be sought on the left bank of the Rio Tapajoz, and such
specimens might assist in solving the status of the unique T. leucura
Pelzeln from the upper Rio Madeira, the characters of which suggest
individual aberration. The greyer breast and upper parts, however,
suggest the possibility of a local subspecies.

664. Haematoderus militaris (Shaw)

Type locality: Cayenne

Cameta, Rio Tocantins (Sclater and Salvin); Para (Natterer); Igarap6-Assu
(Hellmayr)

2 d" 2 9 , Obidos (Carnegie Mus.)

An exceedingly rare bird, particularly in our area. It remains to be
determined whether Amazonian specimens are the same as Guianian.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 263

665. Querula purpurata (Miiller)

Type locality : Cayenne

Ipitinga (Hellmayr); Rio Capim (Goeldi); Para (Descourtilz, Natterer);

Santarem (Chapman and Riker, Ihering) ; Castanhal (Stone) ; Rio Guama,

Para region, Cussary (Snethlage)

2 cf 2 9 , Para, Bosque, and Val-de-Caes

1 c? 2 9 , Rio Acara, Acara

1 cf , Obidos (Carnegie Mus.)

3 cf 3 9 , Benevides <do.)

2 cf 1 9, Santarem (do.)

666. Perissocephalus tricolor (Miiller)

Type locality: Cayenne

Rio Jary, Monte Alegre (Snethlage)

3 cf 1 9 , Obidos (Carnegie Mus.)

667. Gymnoderus foetidus (Linnaeus)

Type locality: Surinam

Santarem (Chapman); Mexiana Island (Hagmann); Caviana Island (Brod-
korb) ; Para, Quati-puru, Monte Alegre (Snethlage)
1 9 , Rio Tapajoz, Tauary
1 cf 2 9 , Santarem (Carnegie Mus.)
1 9 , Rio Tapajoz, Apacy (do.)

Family PIPRIDAE

668. Piprites chloris chlorion (Cabanis)

Type locality: Cayenne

Utinga (Beebe); Quati-puru, Rio Guama, Cussary, Rio Tapajoz, Rio Jary
(Snethlage)

1 cf 1 9 , Rio Tapajoz, Tauary

8 cf 2 9 , Obidos (Carnegie Mus.)

1 cf 1 9 , Santarem (do.)

3 cf 1 9 , Rio Tapajoz, Miritituba (do.)

1 cf , Rio Tapajoz, Villa Braga (do.)

Still a rare bird in most collections, and Madame Snethlage reported
a total of 7 specimens only from 7 localities. The species ranges widely,
but is apparently unrecorded in much of upper Amazonia as yet.

264 bulletin: museum of comparative zoology

669. Pipra aureola aureola (Linnaeus)

Type locality: Surinam

Mexiana Island (Hellmayr); Caviana Island (Brodkorb); Marajo Island, Rio
Moju, Maraca, Rio Jary, Arumanduba (?), Monte Alegre (?), Rio Mae-
curu (?), (Snethlage); Para to Rio Xingu and Rio Jary (Zimmer)

670. Pipra aureola flavicollis Sclater

Type locality: Manaos

Obidos and Rio Jamunda, Faro (Snethlage); Faro, Monte Alegre and Villa
Bella Imperatriz (Zimmer)

2 d", near Obidos

5o,2 9, Obidos (Carnegie Mus.)

671. Pipra aureola aurantiicollis Todd

Type locality: Santarem

Santarem (Chapman and Riker) ; Cussary, Tamucury (Snethlage) ; Rio Tapa-
joz, Santarem, Tamucury (Zimmer)

15 c71 7 9 , Santarem (Carnegie Mus.)

The races of this species have a somewhat peculiar and unusual
distribution in our area. The race flavicollis crosses the Amazon west-
ward and occurs on the Rio Madeira. On the other hand, typical
aureola occurs on the north bank and also the extreme eastern section
on the south bank of the Amazon. Between this region and the Rio
Madeira, the species is rare and local.

672. Pipra fasciicauda scarlatina Hellmayr

Type locality: Sao Paulo, Brazil

Rio Tocantins (Wallace, Snethlage and Zimmer) ; Itaituba, Rio Tapajoz (Hell-
mayr and Zimmer); Tapajoz, Jamauchim and Curua rivers (Snethlage)
5 d71 4 9 , Rio Tapajoz, both banks (Carnegie Mus.)

This species, closely related to the last and equally variable, is at
the extreme northern limit of its range on the south bank of the
Amazon, where it is apparently commoner than the last.

673. Pipra anomala Todd
Type locality: Santarem, Rio Tapajoz

This type is to date the only recorded specimen, and has been ex-
amined by us.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 265

674. Pipra iris iris Schinz

Type locality : Para, by subsequent designation

Ourem, Benevides, Igarape-Assu, S. Antonio do Prata, Peixe-Boi (Hellmayr) ;

Utinga, Santa Isabel and Castanhal (Stone) ; numerous localities near and

east of Pard, (Snethlage)

13 d* 4 9 , Benevides (Carnegie Mus.)

675. Pipra iris eucephala Todd

Type locality: Miritituba, Rio Tapajoz
Santarem, Colonia do Mojuy (Todd)

2 cf 1 9 , Rio Tapajoz, east bank

2 a" 2 9, (do.), Miritituba (Carnegie Mus.)

1 o" 3 9, Santarem (do.)

676. Pipra nattereri Sclater

Type locality: Borba, Rio Madeira

Left bank of Rio Tapajoz, Boim and Villa Braga (Snethlage)

26 d" 24 9 , Rio Tapajoz, Villa Braga, (Carnegie Mus.)

Snethlage also records this species from the Rio Jamauchim, but the
specimen is probably iris eucephala. These two species may prove to
be conspecific.

677. Pipra erythrocephala erythrocephala (Linnaeus)

Type locality: Surinam

Obidos (Hellmayr) ; north bank of Amazon, Rio Jary to Rio Jamunda (Sneth-
lage)

1 cf 2 9 , Obidos (Carnegie Mus.)

678. Pipra erythrocephala rubrocapilla Temminck

Type locality: Bahia

Bemfica, Marco de Legua, Nazare, Rio Capim, Peixe-Boi, Ipitinga, S. Antonio

do Prata (Hellmayr) ; Utinga (Beebe) ; various localities near Para (Stone) ;

throughout from Para to left bank of Rio Tapajoz (Snethlage) ; Santarem

(Chapman and Riker)
6 d\ Para, Bosque

2 d\ Rio Acara, Acara

26 d1 16 9 , Rio Tapajoz, east bank
16 d" 7 9, Benevides (Carnegie Mus.)
9 d1 6 9, Santarem (do.)
24 d" 11 9 , Rio Tapajoz (both banks) (do.)

266 bulletin: museum of compakative zoology

679. Pipra pipra pipra (Linnaeus)

Type locality: Surinam

Rio Jary, Obidos, Rio Jamunda (Snethlage) ; Faro (Zimmer)
2 c\ Obidos (Carnegie Mus.)

680. Pipra pipra separabilis Zimmer

Type locality: Tapara, Rio Xingu, Brazil

Para (Wallace, Layard) ; Benevides, Maguary, S. Antonio do Prata, Igarape-
Assu, Peixe-Boi, Ipitinga (Hellmayr); Rio Capim (Goeldi); Ipitinga
(Beebe); Castanhal (Stone); Para region and Rio Tocantins (Snethlage);
numerous localities, Para to Rio Tapajoz (Zimmer)
4 c? 2 9 , Para, Bosque
2 d* 1 9 , Rio Acara, Acara
4 <? 5 9 , Benevides (Carnegie Mus.)

681. Machaeropterus pyrocephalus pyrocephalus (Sclater)

Type locality: probably eastern Peru; Ucayali, Upper Amazons by Berlepsch

and Hartert.
Boim, Rio Tapajoz (Snethlage and Zimmer)

4 d* 3 9 , Rio Tapajoz, Santarem, Tauary & Caxiricatuba
31 d71 11 9, Santarem (Carnegie Mus.)

5 o71 19, Rio Tapajoz, Apacy (do.)

A rare species with a scattered distribution, suggesting that it still
remains to be discovered in intervening areas.

682. Ceratopipra cornuta (Spix)

Type locality : forests of Amazon River
Obidos (Bates)

683. Tyranneutes virescens (Pelzeln)

Type locality: Manaos, Brazil
Obidos, Rio Jamunda (Snethlage)

27 cT 7 9 , Obidos (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 267

684. Tyranneutes stolzmanni (Hellmayr)

Type locality: Marabitanas, Rio Negro

Peixe-Boi (Hellmayr); Castanhal (Stone); Para, Providencia, Ananindeua,
Rio Tocantins, Cussary, Rio Jamauchim, and Rio Tapajoz (Snethlage)

1 cf, Para, Bosque

15 c? 1 9 , Rio Tapajoz, various localities east bank.

6 cf 2 9 , Benevides (Carnegie Mus.)

13 cT 7 9 , Santarem (do.)

2 cf 1 9 1 ?, Rio Tapajoz (do.)

685. Chiroxiphia pareola pareola (Linnreus)

Type locality: Brazil and Cayenne

Para, S. Antonio do Prata, Ipitinga (Hellmayr); Santarem (Chapman and
Riker) ; Obidos (Hellmayr) ; Capim and Inhangapy (Stone) ; large series,
Para to right bank of the Tapajoz, Marajo Island, Monte Alegre (Sneth-
lage)

1 cT, Rio Acara, Acara
39 cf 15 9 , Rio Tapajoz, various localities, east bank.
1 c\ Obidos (Carnegie Mus.)
6 d" 1 9 , Benevides (do.)

8 <? 3 9 , Santarem (do.)

1 cf 1 9 , Rio Tapajoz, Aveiros (do.)

686. Chiroxiphia pareola regina Sclater

Type locality: Borba, Rio Madeira

West bank of Rio Tapajoz, Boim and Villa Braga (Snethlage and Hellmayr)
4 c? 6 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

687. Manacus manacus manacus (Linnpeus)

Type locality: Surinam

Monte Alegre, Obidos, Rio Jamunda (Snethlage); Faro and Monte Alegre
(Zimmer)

1 a* 2 9 , Obidos (Carnegie Mus.)

688. Manacus manacus purissimus Todd

Type locality: Benevides, Para, Brazil

Numerous records by all collectors from the Para region to the right bank of
the Rio Tocantins (Baiao, fide Zimmer)

3 c? 1 9 , Para, Val-de-Caes
1 d" 1 9 , Rio Acara, Acara

9 d1 4 9 , Benevides (Carnegie Mus.)

268 bulletin: museum of comparative zoology

689. Manacus manacus longibarbatus Zimmer

Type locality: Tapara, Rio Xingu, Brazil

Right bank of the Rio Xingu, and apparently also "Baiao", Rio Tocantins,
fide Zimmer

690. Manacus manacus purus Bangs

Type locality: Santarem, Rio Tapajoz

Numerous records by all collectors from the right bank of the Rio Tapajoz to
Villa Bella Imperatriz, and the right bank of the Rio Madeira

7 cf 7 9 , Rio Tapajoz, east bank

2 o"7 9, Santarem (Carnegie Mus.)

3 cf 2 9 , Rio Tapajoz (do.)

691. Neopipo ctnnamomea cinnamomea (Lawrence)

Type locality: "Upper Amazon"

1 cf, Rio Tapajoz, Villa Braga (Carnegie Mus.)

Previously unrecorded east of the left bank of the Rio Madeira

692. Schiffornis major Des Murs

»
Type locality: Sarayacu, Peru
Santarem (Hellmayr); Rio Jamundd, Faro (Snethlage)

1 d\ Santarem

3 d\ Obidos (Carnegie Mus.)

4 d* 2 9 , Santarem (do.)

We prefer Zimmer's more conservative course in not recognizing the
genus Massornis.

693. Schiffornis turdinus wallach (Sclater and Salvin)

Type locality: Para

Para (Wallace, Stone); S. Antonio do Prata (Hellmayr); Pard to Rio Xingu on

south bank, and north bank, Rio Jary to Rio Jamundd (Snethlage);

numerous localities throughout (Zimmer)

2 cf1, Para, Bosque

4 c? 2 9 , Rio Tapajoz, various localities, east bank

3 d\ Obidos (Carnegie Mus.)
6 d* 3 9 , Benevides (do.)

8 c? 3 9 , Santarem (do.)

4 d* 3 9 , Rio Tapajoz, east bank (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 269

694. Schiffornis turdinus amazonus (Sclater)

Type locality: Chamicuros, Peru

6 cf 2 9 , Rio Tapajoz, Villa Braga

The material in the Carnegie Museum makes it clear that a final
revision of the racial variation of this difficult species in upper Amazon-
ian Brazil must be postponed until someone can assemble material
from the entire area, without having to guess about the characters of
series in other institutions. Mr. Zimmer's recent revision (Amer. Mus.
Novit., no. 899, 1936, pp. 21-24) is, of course, quite the most authori-
tative, but he lacked Brazilian material west of the Rio Madeira. Mr.
Todd's intercedens was based on a comparison of his Purus series with
birds from east Ecuador, at that time passing as amazonus. Hellmayr
guessed shrewdly and Zimmer proved that intercedens Todd =
amazonus Sclater, and east Ecuador birds are a new race aeneus
Zimmer. Seventeen specimens from the Rio Purus before us, presum-
ably, therefore, representing amazonus, are strikingly distinct from
wallacii in just the characters ascribed. On the right bank of the
Tapajoz, occasional specimens show an approach to amazonus, but
the series before us from the left bank (Villa Braga) is clearly insepa-
rable from the Rio Purus series. We differ from Zimmer only in extend-
ing the range of amazonus eastward to the left bank of the Tapajoz.
Mr. Zimmer has already commented that birds seen by him from the
Rio Madeira "strongly suggest ainazonus."

695. Neopelma pallescens (Lafresnaye)

Type locality: Bahia

Rio Tapajoz, Santarem and Itaituba (Hellmayr); Serra de Paituna on north
bank (Snethlage)

5 d71 1 9 , Rio Tapajoz, Santarem
12 cf 2 9 , Santarem (Carnegie Mus.)

The four species of Neopelma have a general range in eastern South
America from the Guianas to southeastern Brazil. It is noteworthy
that only one occurs rarely and locally in the lower Amazon, and on
both banks at that. We have here an excellent illustration of a group
whose range must have become disrupted by the Amazonian sea, and
where the former gap in this range has not yet been filled in.

270 bulletin: museum of comparative zoology

696. Heterocercus linteatus (Strickland)

Type locality: upper branches of Amazon River

Santarem (Chapman and Riker) ; Rio Iriri, Rio Jamauchim and Rio Tapajoz
(Snethlage); various localities Rio Tapajoz (Zimmer)
14 (J1 15 9 , Rio Tapajoz, east bank

1 9 , Santarem (Carnegie Mus.)
19 c? 5 9, Rio Tapajoz, both banks (do.)

697. Heterocercus flavivertex Pelzeln

Type locality: Marabitanas, Rio Negro
Rio Jamundd, Faro (Zimmer)

Snethlage records //. linteatus from Monte Alegre on the north bank
of the Amazon. Either this locality is erroneous, or the specimen is

flavivertex.

Family TYRANNIDAE
698. Xolmis cinerea (Vieillot)

Type locality: South America

Marajo Island (Allen and Snethlage); Caviana Island (Brodkorb); Rio Iriri
(Snethlage)

A well known campos species, here at its northern limit.

699. Xolmis velata (Lichtenstein)

Type locality : Sao Paulo, Brazil

Mexiana Island (Wallace, Hagmann); Marajo Island (Allen, Hellmayr);
Erere, Rio Maecuru (Snethlage)

3 <? 1 9 , South bank of Amazon, Lago Grande

This species is also at its extreme northern limit in our area, occur-
ring only in unforested localities.

700. Colonia colonus colonus (Vieillot)

Type locality: Paraguay

1 <?, Arary, Marajo Island (Brodkorb, 1937)

It is certainly remarkable that the only Amazonian specimen of this
genus, although collected in 1871, was not recorded until 1937. The
subspecies must be regarded as tentative only. The typical race is not
otherwise known north, of Maranhao (Zimmer, 1937), while poecilonota
(Cabanis) is not reported outside of the Guianas.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 271

701. Knipolegus orenocensis xinguensis Berlepsch

Type locality: Santa Julia, Rio Iriri, Rio Xingu

This species, apparently strictly riparian, is still rare in collections.
The race xinguensis is still known only from the type collection, and
two males from the Rio Araguaya, Goyaz, which Hellmayr refers here
provisionally.

702. Knipolegus orenocensis sclateri Hellmayr

Type locality: Rio Madeira

2 cf imm., 2 9 , Rio Tapajoz, Pinhy and Caxiricatuba

An equally rare bird, so far reported only from the type locality and
Pebas, Peru. Our specimens are so dark and brown, that they cannot
possibly be referred to xinguensis.

703. Phaeotriccus pgecilocercus (Pelzeln)

Type locality: Rio Amajau, Rio Negro

"Lower Amazon" (type of Cnipolegus pusillus in Brit. Mus.); Rio Tapajoz,
Itaituba (Hellmayr); Rio Jamunda, Faro, Cussary, Monte Alegre, Rio
Iriri, Rio Tocantins (Snethlage) ; numerous localities east to the Tocantins
(Zimmer, 1937)

3 cf 1 9 , Rio Tapajoz, west bank, Pinhel

4 cf ad. 1 cf imm., Rio Tapajoz, east bank, Pinhy and Caxiricatuba.

Another strictly riparian species, apparently much commoner than
the Knipolegus. Both are recorded from the Orinoco as well as the
Amazon.

704. Fluvicola pica albiventer (Spix)

Type locality: "Brazil"

Mexiana Island (Wallace); Marajo Island (Snethlage); Caviana Island
(Brodkorb); Monte Alegre, Arumanduba, Rio Maecuru, Rio Iriri, Rio
Tapajoz (Snethlage); Santarem (Chapman and Riker)
4 d71 2 9 , north bank of Amazon near Obidos
4 cT 1 9 , Obidos (Carnegie Mus.)
2 cf 1 9 , Santarem (do.)

272 bulletin: museum of comparative zoology

705. Arundinicola leucocephala (Linnaeus)

Type locality: Surinam

Mexiana Island (Hagmann); Marajo Island (Hellmayr); Caviana Island
(Brodkorb); Peixe-Boi, Quati-puru, Arumanduba, Monte Alegre, Cussary,
(Snethlage) ; Santarem (Chapman and Riker)
3 d\ north bank of Amazon near Obidos
3 c? 2 9 , Obidos (Carnegie Mus.)
2 cf 2 9 , Santarem (do.)

706. Pyrocephalus rubinus rubinus (Boddaert)

Type locality: "Amazon River"; see Brodkorb, 1937
Monte Alegre, Rio Xingu, Rio Iriri, Rio Curua (Snethlage)
1 cf 1 9 , Santarem (Carnegie Mus.)

So far as we can see, typical rubinus, as currently restricted, still
requires further study accurately to delimit its range. Brodkorb's work
is certainly a step in the right direction.

707. Ochthornis littoralis (Pelzeln)

Type locality: Cachoeira, Rio Mamore, Rio Madeira; Rio Jamauchim (Sneth-
lage)

1 9 , Obidos (Carnegie Mus.)

2 d* 1 9 , Rio Tapajoz, Villa Braga (do.)

An upper Amazonian genus, only once reported east of the Rio
Madeira.

708. Muscivora tyrannus tyrannus (Linnaeus)

Type locality : Surinam

Pard region, Marajo and Mexiana Islands (numerous collectors); Caviana

Island (Brodkorb); Monte Alegre and Rio Jamunda, Faro (Snethlage);

Bio Tapajoz (various collectors) ; always reported as common where

found. Par&, Rio Tapajoz, various localities, and Villa Bella Imperatriz,

15 specimens definitely this subspecies (Zimmer)

Zimmer's monographic study of this species (Amer. Mus. Novit. no.
962, Nov. 1937) proves that typical tyrannus is migratory, and occurs
in our area from February to November. It is of course impossible to
allocate the old records for the species in lower Amazonia, which are all
given above.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 273

709. Muscivora tyrannus circumdatus Zimmer, 1937

Type locality: Tauary, Rio Tapajoz, Brazil

Apparently resident on the south bank of the Amazon from Villa Bella Im-
peratriz to the right bank of the Tapajoz (Zimmer)

13 cf 4 9 , Rio Tapajoz, various localities east bank (M.C.Z.)

710. Tyrannus albogularis Burmeister

Type locality: near Bahia and Pernambuco

Monte Alegre, 1 d" Aug. 9, 1908 (Snethlage); Santarem 1 d\ July 25, 1883,

(Chapman and Biker) ; Santarem, Rio Tapajoz and Villa Bella Imperatriz,

11 specimens, no dates (Zimmer)

1 d\ south bank of Amazon, Lago Grande, Sept. 5

2 d" 1 9 , Santarem, June 13, July 29, 1919 (Carnegie Mus.)

A characteristic campo bird of the interior of Brazil, apparently
rare in our area. The recorded dates of capture of the known speci-
mens, and the fact that Hellmayr has examined the bird from the
north bank and found it identical with Matto Grosso specimens, raises
the presumption that the species is partly migratory, and may occur
in our area as a winter visitant only.

711. Tyrannus melancholicus melancholicus Vieillot

Type locality: Paraguay.

Villa Bella Imperatriz (Zimmer)

This Kingbird is at least partly migratory, and most of the birds
from the south bank of the Amazon are unsatisfactory intermediates
(cf. Zimmer, Amer. Mus. Novit. no. 962).

712. Tyrannus melancholicus despotes (Lichtenstein)

Type locality: Bahia

Common throughout the area (all collectors)
1 d1 1 9 , Para, Bosque

Rio Tapajoz, various localities, east bank

Obidos (Carnegie Mus.)

Benevides (do.)

Santarem (do.)

Rio Tapajoz, both banks (do.)

23 & 6

1 tf 1

1 d1 2

3 d1 2

274 bulletin: museum of comparative zoology

713. Empidonomus varius varius (Vieillot)

Type locality: Paraguay

Rio Tapajoz, Aramanay and Igarape Brabo, probably migrants (Zimmer,
1937)

714. Empidonomus varius rufinus (Spix)

Type locality: Amazon River

Common throughout the area, but unrecorded from Marajo and Mexiana
Islands

2 cf , Para, Val-de-Caes

15 cf 3 9,1?, Rio Tapajoz, east bank

3 cf 2 9 , Benevides (Carnegie Mus.)
9 cf 5 9, Santarem (do.)

2 cf 1 9 , Rio Tapajoz, both banks (do.)

715. Empidonomus aurantio-atro-cristatus minor (Hellmayr)

Type locality: Sao Luiz, Maranhao, Brazil
Santarem 1 cf, (Snethlage); Santarem 2 cf, (Zimmer)

6 cf 5 9 , Rio Tapajoz, various localities, east bank

4 o71 1 9 , Santarem (Carnegie Mus.)

This species illustrates how spotty our knowledge is of the status of
many Amazonian birds. The good series collected by the Olallas
could not have been predicted by the previous absence of records.
There is apparently no reason why it should not occur in other parts
of our area, and it is possibly a winter visitant only.

716. Legatus leucophaius leucophaius (Vieillot)

Type locality: Cayenne

Para region (common); Marajo Island (Snethlage); Rio Tapajoz region
(common); Obidos (Snethlage)

2 cf , Para, Bosque and Val-de-Caes
14 cf 3 9 , Rio Tapajoz, Tauary and Pinhy
2 cf , Obidos (Carnegie Mus.)
2 cf 1 9 , Benevides (do.)
4 cf , Santarem (do.)

As is usual in a good series in this species, one or two are notably
larger than all the others.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 275

717. SlRYSTES SIBILATOR SUBCANESCENS Todd

Type locality: upper Rocana, northern Para, Brazil

Rio Jamunda, Faro, 1 9 , (Snethlage) ; Rio Tocantins, Baiao and Pedregal,
and Rio Tapajoz, Limoal (Zimmer)
2 cf 4 9 , Obidos (Carnegie Mus.)

The type locality is only a few miles from the French Guiana border.
Birds from the south bank of the Amazon might well be another sub-
species, as Zimmer had no authentic specimens of subcanescens.
This widely diffused and variable species is very rare in all parts of its
range, except for the typical form in southeastern Brazil.

718. Myiodynastes maculatus maculattjs (Muller)

Type locality: Cayenne

Cajutuba (Natterer); Mexiana Island (Hagmann); Caviana Island (Brod-

korb) ; various north bank localities (Snethlage) ; Monte Alegre, Faro, and

Villa Bella Imperatriz, (Zimmer)
1 cf, Lago Jauary near Obidos

1 cf 2 9 , Obidos (Carnegie Mus.)

2 cf 1 9 , Santarem (do.)

The apparent absence of this species from most of the south bank of
the Amazon is noteworthv.

719. Myiodynastes maculatus solitarius (Vieillot)

Type locality: Paraguay

Para, region (all authors); Rio Tocantins (Snethlage); Santarem (Chapman
and Riker, Snethlage) ; Para to Rio Tapajoz (Zimmer)

4 cf 10 9,1? Rio Tapajoz, various localities, east bank.

1 9, Obidos (Carnegie Mus.)

5 cf 2 9, Santarem (do.)

Apparently a winter visitor only in our area. Zimmer reports inter-
mediates from the Rio Xingu.

276 bulletin: museum of compakative zoology

720. Megarynchus pitangua pitangua (Linnaeus)

Type locality: eastern Brazil

Mexiana Island (Wallace); Rio Tocantins (Snethlage); Santarem (Chapman

and Riker); Cussary, Rio Tapajoz, Rio Jamunda (Snethlage)

1 cf 1 9 , near Obidos

4 d71 2 9 1 ? , Rio Tapajoz, various localities, east bank

2 a71 1 9 , Obidos (Carnegie Mus.)

5 d71 2 9 , Santarem (do.)

1 d" 3 9 , Rio Tapajoz, left bank (do.)

721. Conopias trivirgata berlepschi Snethlage

Type locality: Rio Jamunda, Faro, 3 o*

Faro, Rio Tapajoz, Caxiricatuba, Tauary, Igarape Amorin (Zimmer)

1 d71 1 9 , Boca do Igarap^-Piaba, near Obidos, March, 1933.

1 9 , Rio Tapajoz, Pinhy, June 30, 1930

2 c" 1 9 , islands near Obidos (Carnegie Mus.)

3 d\ Obidos (do.)
5 d71 1 9 , Santarem (do.)

3 c? 2 9 , Rio Tapajoz, left bank (do.)

Obviously more widely ranging and less rare than formerly sup-
posed.

722. Coryphotriccus parvus parvus (Pelzeln)

Type locality: Marabitanas, Rio Negro

5 d1 2 9 , Obidos (Carnegie Mus.)

The first record for this rare genus in Amazonia.

723. Myiozetetes cayanensis cayanensis (Linnaeus)

Type locality: Cayenne

Para region (abundant); Mexiana Island (Hellmayr); Santarem (Chapman and
Riker); Rio Moju, Arumanduba (Snethlage)
2 d1 3 9 , Para, Bosque
13 cf 8 9 , Rio Tapajoz, various localities
1 o71 2 9 , Obidos (Carnegie Mus.)

1 d\ Benevides (do.)
1 d71 2 9 , Rio Tapajoz, both banks (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 277

724. Myiozetetes similis similis (Spix)

Type locality: Amazon River; Rio Madeira as restricted by Zimmer.
Para (Layard, Hellmayr); north shore of Amazon to Rio Jamunda (Snethlage);
Villa Bella Imperatriz and Rio Tapajoz, Igarape Brabo (Zimmer)
1 cT , Rio Tapajoz, Santarem
1 9 , near Obidos
11 d" 5 9, Obidos (Carnegie Mus.)
8 d* 4 9 , Santarem (do.)

Para records may prove to be ■pallidiventris Pinto.

725. Tyrannopsis sulphurea (Spix)

Type locality: Brazil

Rio Muria (Natterer); Rio Inhangapy (Stone); Santarem (Chapman and
Riker); Para, Marajo Island, Amapa (Snethlage)

3 d\ Rio Acara, Acara

2 o71 3 9, Santarem (Carnegie Mus.)

726. PlTANGUS SULPHURATUS SULPHURATUS (Lillliseus)

Type locality: Cayenne
Abundant throghout the area (all collectors)
3 6" 2 9, Para, Val-de-Caes
14 d1 12 9 , Rio Tapajoz, various localities, east bank.
1 9, Obidos (Carnegie Mus.)
1 d\ Benevides (do.)

1 cf , Santarem (do.)

2 cf, Rio Tapajoz, left bank (do.)

While Hellmayr claims that maximHiani is a very unsatisfactory
race, the ample material in the Museum of Comparative Zoology does
not endorse this characterization. It" is true that the average color
differences are relatively slight, but there is a marked difference in
wing length. In this respect our Amazonian series agrees with Cayenne
material, the wings of all combined being nearly an inch shorter on the
average than Bahia examples of maximiliani. We consequently quite
definitely refer Para birds to sulphuratus, and do not at all subscribe to
Hellmayr's statement that they "might as well be referred to one race
as the other".

278 bulletin: museum of comparative zoology

727. Pitangus lictor lictor (Liehtenstein)

Type locality: Para

Common, and recorded throughout our area

2^3 9, near Obidos

5 d1 3 9 2 juv., Rio Tapajoz, various localities, east bank

2 d\ Obidos (Carnegie Mus.)

5 d1 4 9 , Santarem (do.)

1 d 1 9 , Rio Tapajoz, both banks (do.)

728. Myarchus tyrannulus bahiae Berlepsch and Leverkuhn

Type locality: Bahia

Rio Muria (Natterer); Santarem (Chapman and Riker); Monte Alegre
(Snethlage); numerous localities in lower Amazonia (Todd); Rio Xingu,
Rio Tapajoz, Rio Jamunda, Monte Alegre (Zimmer)

3 d" 1 9, Rio Tapajoz, various localities east bank

5 d 2 9 , Obidos (Carnegie Mus.)

9 d" 4 9, Santarem (do.)

1 d" 1 9 , Rio Tapajoz, left bank (do.)

729. Myiarchus swainsoni pelzelni Berlepsch

Type locality: Bahia

Mexiana Island (Hellmayr, Snethlage, Zimmer); Marajo Island (Snethlage);
Boim, Rio Tapajoz (Snethlage) ; Rio Xingu, Tapara (Zimmer)

1 d1 4 9 , Rio Tapajoz, east bank

1 d1 3 9 , Santarem (Carnegie Mus.)

730. Myiarchus swainsoni amazonus Zimmer, 1938

Type locality: Rio Jamunda, Faro, Brazil

Mexiana Island, (Hellmayr); Benevides, Santarem (Todd); Rio Tapajoz
(Zimmer)

3 d1 2 9 , Rio Tapajoz, east bank

1 9 , Benevides (Carnegie Mus.)
3 d\ Santarem (do.)

Typical swainsoni is to be expected in our area as a migrant.

GKISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 279

731. Myiarchus ferox ferox (Gmelin)

Type locality: Cayenne

Throughout our area (Todd, Proc. Biol. Soc. Wash., 35, 1922, p. 197, complete
list of localities) ; Pard, to Villa Bella Imperatriz (Zimmer)
12 cf 10 9 1 ? , Rio Tapajoz, east bank

10 c? 12 9 , Obidos, Benevides, Santarem, Rio Tapajos (Carnegie
Mus.)

732. Myiarchus tuberculifer clarus Zimmer, 1938

Type locality: Tapard, Rio Xingu, Brazil

Whole north bank of Amazon (Snethlage) ; Rio Tapajoz (Todd, loc. cit., p. 21 1) ;
Faro, Villa Bella Imperatriz (Zimmer)

2 cf, Rio Tapajoz, Tauary and Pinhy

5 d\ Obidos (Carnegie Mus.)

4 cf 1 9 , Santarem (do.)

2 c? 1 9 , Rio Tapajoz, left bank (do.)

733. Myiarchus tuberculifer subsp.

Pard, Igarape-Assu, 1 9 , (Zimmer)

1 9 , Benevides (Carnegie Mus.)

These specimens will probably prove to be tricolor Pelzeln, which is
restricted by Zimmer to the coast region of eastern Brazil.

734. Contopus cinereus surinamensis Penard and Penard

Type locality: Surinam

Mexiana Island (Sclater and Hellmayr); Marajo Island (Hellmayr)

735. Empidonax lawrencei bolivianus Allen

Type locality: Yungas, Bolivia

Mexiana Island (Hellmayr); Rio Curua "(Snethlage); Rio Xingu, Tapard
(Zimmer)

1 9 , Boca do Igarape-Piaba, near Obidos

1 cf, Obidos (Carnegie Mus.)

1 9 , Benevides (do.)

5 cf 3 9 , Santarem (do.)

1 cf, Rio Tapajoz, Miritituba (do.)

The Carnegie Museum has series of" this bird from the Rio Purus
and Rio Solimoes.

280 bulletin: museum of comparative zoology

736. Empidonax euleri euleri (Cabanis)

Type locality: Cantagallo, Rio de Janeiro

Rio Capim, Rio Muraiteua (Stone); Rio Tocantins (Snethlage); Rio Tapajoz,
Igarape Amorin (Zimmer)

1 9 , Rio Tapajoz, Pinhy

1 9 iram., Santarem (Carnegie Mus.)

Presumably a winter migrant to our area.

737. Cnemotriccus fuscatus fumosus (Berlepsch)

Type locality: Cayenne

Arumanduba, Igarape de Paituna, Obidos, Marajo Island (Snethlage) ; Mexi-
ana Island and Faro (Zimmer)

2 c? 1 9 , Obidos (Carnegie Mus.)

2 d" 3 9 , islands, Obidos (do.)

738. Cnemotriccus fuscatus fuscatior Chapman

Type locality: Rio Curaray, Ecuador

Rio Tapajoz, Goyana (Snethlage); Villa Bella Imperatriz to Rio Tocantins
(Zimmer)

3 cf 1 9 , Santarem (Carnegie Mus.)

3 cf 2 9 1 ? , Rio Tapajoz, Goyana Island and left bank (do.)

We are unable to endorse Hellmayr's treatment of this difficult
species in the Cat. Birds America, where he refers all birds south of the
Amazon to bimaculatus (Lafresnaye & D'Orbigny) from Yungas,
Bolivia, with the exception, of course, of typical fuscatus from south-
eastern Brazil. We have before us Bolivian topotypes of bimaculatus,
and it is clear that this pale form ranges through the campo country of
western Brazil north to the upper stretches of the Rio Purus and the
Rio Madeira. A series (7 spec.) from the Rio Solimoes is a very differ-
ent dark race, slightly darker and browner above, and much darker
and heavily flammulated with grayish olive on the chest, and yellowish
rather than whitish on the belly. This is fuscatior Chapman of south-
eastern Ecuador, and it is clear that birds from Amazonian Peru prob-
ably belong here also, judging by Hellmayr's comments. We are quite
unable to separate our lower Amazonian series from these birds. Pro-
ceeding northeastward, a fine series from Cayenne (21 specimens) repre-
sent fumosus (Berlepsch), which differs from bimaculatus in just the
characters ascribed by Hellmayr. We agree that birds from Obidos

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 281

belong here, as do a series from the islands in the Amazon between
Obidos and Santarem. All these birds are barely separable from fusea-
tior Chapman, which is a slightly darker bird on the average, although
many specimens are interchangeable. We have here a most unsatisfac-
tory distributional picture, undoubted fuseatior on the Rio Solimoes,
undoubted bimaculatus on the upper Purus and the Rio Madeira, while
from Villa Bella Imperatriz eastward occurs an admittedly variable
and dimorphic population, perhaps nearer fuseatior than fumosus. A
conservative position which chose to reduce fuseatior to the synonymy
of fumosus could certainly be defended. Hellmayr's treatment was
equally conservative, but modern series prove that birds from the south
bank of the Amazon in our area most certainly cannot be referred to
bimaeulatus.

739. Terenotriccus erythrurus erythrurus (Cabanis)

Type locality: Guiana

Rio Jary and Obidos (Snethlage) ; Rio Jamundd,, Faro (Zimmer)
2 <?, Obidos (Carnegie Mus.)

740. Terenotriccus erythrurus hellmayri Snethlage

Type locality: Para

The Para region (all collectors) to the Rio Tocantins (Zimmer)
2 cf 1 9 , Rio Acard, Acara

1 9 , Benevides (Carnegie Mus.)

741. Terenotriccus erythrurus amazonus Zimmer, 1939

Type locality: Igarape Amorin, Rio Tapajoz, Brazil

Santarem and Rio Tapajoz (all collectors) ; Villa Bella Imperatriz (Zimmer)

2 d", Rio Tapajoz, Caxiricatuba and Pinhy
1 cf, Santarem (Carnegie Mus.)

14 d> 2 9 , Rio Tapajoz, both banks (do.)

742. Mytobius barbatus barbatus (Gmelin)

Type locality: Cayenne

Rio Jary and Obidos (Snethlage) ; Rio Jamunda, Faro (Zimmer)

3 cf , Obidos (Carnegie Mus.)

282 bulletin: museum of comparative zoology

743. Myiobius barbatus insignis Zimmer, 1939

Type locality: Piquiatuba, Rio Tapajoz, Brazil

Common throughout from Para region to the Rio Tapajoz

3 cf 1 9 , Rio Tapajoz, various localities east bank
19, Benevides (Carnegie Mus.)

2 cf 2 9, Santarem (do.)

5 cf 3 9 , Rio Tapajoz, both banks (do.)

744. Myiobius atricaudus connectens Zimmer, 1939

Type locality: Mocajuba, Rio Tocantins, Brazil
Para, Prata, Rio Tocantins, Rio Tapajoz, Rio Jamauchim (Zimmer)
2 cf, Rio Tapajoz, Pinhy and Caxiricatuba

745. Myiophobus fasciatus flammiceps (Temminck)

Type locality: Rio de Janeiro

Para (Sclater); Maguary and Mexiana Island (Snethlage)

746. Onychorhynchus coronatus coronatus (Miiller)

Type locality: Cayenne

Common from Pard region to the Rio Tapajoz; Villa Bella Imperatriz and
Faro (Zimmer)

1 9 , Rio Acara, Acara
1 cf 1 9 , Rio Tapajoz, Caxiricatuba
1 cf 3 9 , Obidos (Carnegie Mus.)
1 cf 1 9 , Benevides (do.)
1 cf 5 9 , Santarem (do.)
1 cf 1 9 , Rio Tapajoz (do.)

747. Platyrinchus senex griseiceps Salvin

Type locality: Ourumee, British Guiana
Obidos (Snethlage)

12 cf 8 9 , Obidos (Carnegie Mus.)

748. Platyrinchus senex amazonicus Berlepsch

Type locality: Peixe-Boi, Para

Numerous records Para region to Rio Tapajoz (Snethlage)

1 cf 2 9 , Rio Tapajoz, Pinhy and Caxiricatuba

6 cf 3 9 , Benevides (Carnegie Mus.)

2 cf 1 9 , Santarem (do.)

4 cf 1 9 , Rio Tapajoz, both banks

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 283

749. Platyrinchus saturatus Salvin and Godman

Type locality: Merume Mts., British Guiana

Peixe-Boi (Hellmayr); Para (Stone); Rio Jamundd, Obidos, Rio Jary and
Para region (Snethlage); Faro, Obidos, Utinga (Zimmer)
3 d1 1 9 , Obidos (Carnegie Mus.)
3^3 9, Benevides (do.)
1 9 , Santarem (do.)
1 9 , Rio Tapajoz, Villa Braga (do.)

750. Platyrinchus coronatus coronattjs Sclater

Type locality: Rio Napo, Ecuador

Rio Curua, Rio Jamauchim (Snethlage) ; Rio Tapajoz, Caxiricatuba (Zimmer)

1 d* , Rio Tapajoz, Caxiricatuba

5 o" , (do.), Miritituba (Carnegie Mus.)

751. Platyrinchus coronatus gumia (Bangs and Penard)

Type locality: Paramaribo, Dutch Guiana
Rio Jary (Snethlage Hellmayr); Faro (Zimmer)

752. Tolmomyias sulphurescens mixtus Zimmer, 1939

Type locality: Baiao, Rio Tocantins, Brazil

Known only from the type locality and one station in Maranhao.

753. Tolmomyias sulphurescens insignis Zimmer, 1939

Type locality: Rosarinho, Rio Madeira, Brazil
Rio Jamunda, Faro, Castanhal (Zimmer)

754. Tolmomyias flavotectus assimilis (Pelzeln)

Type locality: Borba, Rio Madeira

Rio Tapajoz, Igarape Amorin, Igarape Brabo (Zimmer)
1 <?, Rio Tapajoz, Pataua

755. Tolmomyias flavotectus paraensis Zimmer

Type locality : Utinga, Para, Brazil

Utinga and Cameta, Rio Tocantins (Zimmer)

284 bulletin: museum of comparative zoology

756. TOLMOMYIAS FLAVOTECTUS EXAMINATUS (Chubb)

Type locality: Bartica Grove, British Guiana
Rio Jamunda, Faro, Castanhal (Zimmer)

Zimmer's study of this genus (Amer. Mus. Novit., no. 1045) proves
that Amazonian records of sulphurescens really include races of flavo-
tedus, a species previously unrecognized in South America east of the
Andes. Unfortunately the genus is quite scarce in our area, and the
few specimens examined by Zimmer inevitably leave large distribu-
tional gaps. It is of course impossible to allocate the earlier records
without reexamination of the specimens. Snethlage reports "assi mills
(Pelzeln) from the Para region to the Rio Tapajoz. Birds from the
Para region might be either sulphurescens mixtus or flavotedus paracusis.
The identity of birds from Santarem and the right bank of the Rio
Tapajoz in the Carnegie Museum is problematical as Zimmer has been
unable to examine a specimen of either species from there. No one
knows where the subspecies mixtus and insignis of sulphurescens inter-
grade on the south bank of the Amazon. Equal uncertainty exists with
the two races of flavotedus, and without comparative material, our one
specimen from Pataua might be paracusis. The situation is almost as
bad on the north bank. Faro is the only place from which both species
are reported. The genus is recorded from the Rio Jary (Snethlage)
and Obidos (Hellmayr). These birds might be sulphurescens insignis
or flavotedus examinatus. On the other hand it is entirely in accord
with the distribution of other widely ranging and variable species, that
specimens from Obidos (and points further east) might indeed prove
to be sulphurescens cherriei (Hartert and Goodson), as reported by
Hellmayr.

757. Tolmomyias poliocephalus sclateri (Hellmayr)

Type locality: Manaos, Brazil

North bank localities (Snethlage), Para region (numerous collectors) west to
Rio Tapajoz (Snethlage) ; Faro and Para to Villa Bella Imperatriz (Zim-
mer)

1 C? 2 9 , Para, Bosque

3 c? 4 9 1 ? Rio Tapajoz, various localities, east bank
16 cf 10 9, Obidos, Santarem, Rio Tapajoz both banks (Carnegie
Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 285

758. Tolmomyias flaviventris dissors Zimmer, 1939

Type locality: Faro, Rio Jamunda, Brazil

Marajo Island, Rio Tocantins and Rio Tapajoz (Snethlage); Rio Tocantins to
Villa Bella Imperatriz (Zimmer)

3 cf 2 9 2?, Rio Tapajoz, various localities, east bank.
8 c? 2 9 , Santarem (Carnegie Mus.)

It is certainly peculiar that this species is as yet undetected in the
Para region.

759. Tolmomyias flaviventris collingwoodi Chubb

Type locality: Trinidad

North bank localities (Snethlage); Monte Alegre (Zimmer)
1 9 , near Obidos
6 cT 1 9 , Obidos (Carnegie Mus.)

760. Rhynchocyclus olivaceus guianensis McConnell

Type locality: British Guiana

Throughout area from the Para region westward on both banks (Hellmayr),
Snethlage, Zimmer)

3 d1 4 9 , Rio Tapajoz, various localities, east bank.
16 d" 5 9 , Obidos, Benevides, Santarem, Rio Tapajoz (Carnegie

Mus.)

761. Ramphotrigon ruficatjda (Spix)

Type locality: Amazon River

Para (Wallace); Rio Capim (Goeldi); Santarem (Riker); Rio Jamunda, Obidos,
Rio Maecuru (Snethlage)

2 cf, Para, Bosque

4 cf 3 9 , Rio Tapajoz, various localities, east bank

3 d" 3 9 , Obidos (Carnegie Mus.)
2 cf 1 9 , Benevides (do.)

5 d" 6 9 , Santarem (do.)

2 d> 1 9 , Rio Tapajoz, left bank"

A fine series from the Solimoes (wing 73-80, average 75) is obviously
a different subspecies from a series from Cayenne of similar size. The
latter are paler and yellower below, less heavily flammulated wTith olive
green. The Santarem and Tapajoz birds are intermediate. The Para
birds appear to be a little larger (wing 80-81), but a bigger series is
required to confirm this. In color they resemble Rio Solimoes birds
exactly.

286 bulletin: museum of comparative zoology

762. Todirostrum chrysocrotaphum illigeri (Cabanis and Heine)

Type locality: Para

Para (Cabanis, Hellmayr, Snethlage, Stone); Rio Tocantins (Snethlage)

1 cf, Santarem (Carnegie Mus.)

1 c? 1 9, Rio Tapajoz, Itaituba and Villa Braga (do.)

The two birds from the left bank of the Rio Tapajoz1 differ from
illigeri in lacking the malar stripe, and have minutely less white on the
throat, thereby approaching chrysocrotaphum (subsp.?), which is
recorded from the Rio Madeira.

763. Todirostrum pictum Salvin

Type locality: Annai, British Guiana
Obidos (Snethlage)

3(?291?, Obidos (Carnegie Mus.)

This series is obviously a different subspecies from another series
from French Guiana, in being paler, less golden yellow below, much less
heavily spotted, and streaked with black. However, there is no surety
that French Guiana birds properly represent true pictum from British
Guiana. Moreover, guttatum Pelzeln from the Rio Negro is almost cer-
tainly conspecific, and may or may not be properly represented by
specimens from the Rio Solimoes.

764. Todirostrum cinereum cinereum (Linnseus)

Type locality: Surinam

Marajo Island (Hellmayr, Snethlage); Monte Alegre (Snethlage); Santarem
(Chapman and Riker)

2 cf 2 9 , Santarem

1 d" 2 9 , Obidos (Carnegie Mus.)

1 d\ Santarem (do.)

This common and widely ranging species has a scattered distribution
in our area, where the typical form is at its extreme southeastern
limit. There are other races in Brazil south of the Amazon. Marajo
Island birds might approach cearae.

'Since described as similis Zimmer, Amer. Mus. Novit., no. 1066, May 3, 1940, p. 3.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 287

765. Todirostrum maculatum maculatum (Desmarest)

Type locality: French Guiana

Common on the north shore of the Amazon, Marajo and Mexiana Islands, and
the Para region west to the Rio Tocantins

1 cf 1 9 , near Obidos

2 d* 1 9 , Benevides (Carnegie Mus.)
1 d 1 9 , Obidos (do.)

766. Todirostrum maculatum signatum Sclater and Salvin

Type locality: Nauta, Rio Maraflon, Peru
Common from the Rio Xingu to the Rio Tapajoz

13 d 4 9 1 ?, Rio Tapajoz, various localities, east bank

5 d 1 9 , Santarem (Carnegie Mus.)

3 d 1 9 , Rio Tapajoz, both banks (do.)

As Hellmayr has pointed out, birds from the Rio Tapajoz are transi-
tional between true maculatum of the Guianas and signatum of upper
Amazonia.1 We agree that they are nearer signatum. Our series is
notably shorter billed than a series from Surinam, averaging 2-3 mm.
shorter. The two birds from near Obidos are intermediate in colora-
tion, but have the longer bill of Surinam birds, so we refer them to typi-
cal maculatum.

767. Todirostrum latirostre senectum Griscom and Greenway,

1937.

Type locality: Obidos, Brazil

Santarem (Ihering), not given by Snethlage

1 o71 1 9 , north bank of Amazon, Boca do Igarape-Piaba near Obidos
5 d, Obidos (Carnegie Mus.)

5 d 2 Q , Santarem (do.)

Recent publications on this species have plunged this unfortunate
little Tody-Tyrant into systematic confusion. Typical latirostre comes
from Borba, Rio Madeira, and Dr. Hellmayr, the only student who
has seen Pelzeln's type, has always stated that Bolivian, Matto Grosso
(Chapada) and Rio Solimoes birds are typical, giving a distribution
by no means exceptional. On this basis typical latirostre is are latively
light colored and brownish headed bird, in spite of the fact that

1 Since described as diversum Zimmer, Airier. Mus. Novit., no. 1066, 1940, p. 6.

288 bulletin: museum of comparative zoology

Pelzeln described the type as "pileo plumbeo induto". There is no
question about the very distinct caniceps Chapman, a much darker
bird, with a dark grayish olive crown, and more yellow wash on the
abdomen. Only two specimens are known from the type locality in
eastern Colombia, and birds from various parts of Amazonian
Ecuador may or may not properly represent it.

In May 1937, we described senectum from Obidos on the north bank
of the Amazon, a considerable range extension of the species to the
east. We had the advantage of examining the series in the Carnegie
Museum first. Senectum is in general coloration intermediate between
latirostre and caniceps, but most fortunately our notes, made at the
time, state that Obidos birds are slightly grayer, less green above, wing
bars and edgings paler, less rusty buff, below grayer on throat and
chest, less yellow on flanks and belly than 18 specimens from the Rio
Purus and Rio Solimoes. Compared with Matto Grosso specimens of
ochropterus Allen, = typical latirostre fide Hellmayr, Obidos birds are
greener above and slatier, less brownish on the pileum. Compared
with east Ecuador specimens of caniceps, Obidos birds are not so dark
above, the pileum is not quite so clearly slaty, with less extension of the
gray on to the hindneck and mantle, while the underparts are much
nearer latirostre.

We were somewhat surprised when in November, 1937, Mr. Todd
described difficile from the Rio Purus and Rio Solimoes (south bank),
his series amply confirming the differences noted by Hellmayr in two
specimens from Teffe as compared with typical latirostre. But Mr.
Todd (1) never mentioned senectum, (2) his description of difficile
practically duplicates the characters ascribed to senectum, and (3) he
states that Obidos birds are indistinguishable from caniceps, whereas
we consider the same birds slightly paler than his difficile, which is paler
than caniceps!

It is obviously high time for someone to compare these proposed sub-
species with a real series from the Rio Madeira, and let us hope that the
American Museum possesses one. This series may prove that Pelzeln
had a slaty crowned bird after all, in which case the Matto Grosso
birds will bear the name ochropterus (Allen), and it would remain to be
determined how many races would prove recognizable in the Amazon
Valley proper.1

1 Zimmer (Amer. Mus. Novit., no. 1066, 1940) shows that difficile = latirostre, and recognizes
senectum.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 289

768. Todirostrum sylvia schulzi Berlepsch

Type locality: Ourem, Rio Guama
Santa Isabel, San Antonio do Prata (Snethlage)
1 9 , Benevides (Carnegie Mus.)

This species is as yet unrecorded between the upper Rio Branco and
the localities listed above.

769. EUSCARTHMORNIS STRIATICOLLIS GRISEICEPS (Todd)

Type locality: Santarem, Rio Tapajoz

Santarem (Chapman and Riker, Snethlage and Hellmayr) ; Cussary (Snethlage)
6 cf 6 9 1 ?, Rio Tapajoz, various localities, east bank
12 cf 5 9 , Santarem (Carnegie Mus.)

1 9 , Rio Tapajoz, Miritituba (do.)

An exceedingly local race known only from the Rio Tapajoz, right
bank.

770. Euscarthmornis striaticollis iohannis (Snethlage)

Type locality: Monte Verde, Rio Purus

1 9 , Rio Tapajoz, Villa Braga (Carnegie Mus.)

This specimen appears to agree very well with a series from the Rio
Purus, and not with the type series of griseiceps.

771. Euscarthmornis griseipectus Snethlage

Type locality : Alcobaga, Rio Tocantins, 5 c?
1 d\ Santarem (Carnegie Mus.)

This rare species is otherwise recorded only from southeastern Peru !
There are specimens from the Rio Purus in Pittsburgh.

772. Snethlagea minor (Snethlage)

Type locality: Arumatheua, Rio Tocantins

Also Rio Tapajoz, Villa Braga, Pinhel, Boim (Snethlage)

10 cf 4 9 , Rio Tapajoz, Apacy, Villa Braga and Itaituba (Carnegie

Mus.)

290 bulletin: museum of comparative zoology

773. Snethlagea minima minima Todd

Type locality: Itaituba, Rio Tapajoz

Also Rio Tocantins to Rio Madeira (Todd and Hellmayr)

1 o" 1 V, Rio Tapajoz, Tauary

4 c? 4 9 , (do.), both banks (Carnegie Mus.)

The male from Tauary is a most unsatisfactory specimen. In color
characters clearly minima and not minor. The wing length is 47 mm.,
which is too large for minima, but too small for males of minor. The
tail is exactly the same length as in a male minor before us. In other
words, this specimen is half way between these two supposed species,
which appear very dubious to us, so we do not recognize S. minor
snethlagea' H. Snethlage, 1937.

774. Taeniotriccus klagesi Todd

Type locality: Itaituba, left bank of Rio Tapajoz

Type examined by us. As Hellmayr points out, the unique type will
probably prove to be the female of T. andrci Berlepsch and Hartert
from the Caura River valley, Venezuela, only known from one imma-
ture male.

775. Lophotriccus vitiosus subsp.

Rio Jamunda, Faro; Obidos; Rio Jary (Snethlage)

2 d" 1 9 , Obidos (Carnegie Mus.)

These birds are certainly not eulophotes Todd from the Rio Purus,
but they can hardly be true vitiosus from "Peru".

776. Colopteryx galeatus (Boddaert)

Type locality: Cayenne

Common throughout our area (all collectors), except Marajo and Mexiana
Islands

1 9 , Rio Acara, Acara

13 cf 3 9 , Rio Tapajoz, various localities, east bank
2 d" 1 9 , Obidos (Carnegie Mus.)

2 cf , Benevides (do.)
5 cf, Santarem (do.)

2 cf 1 9 , Rio Tapajoz, both banks

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 291

777. Perissotriccus ecaudatus ecaudatus (Lafresnaye and

D'Orbigny)

Type locality: Yuracares, Bolivia

Peixe-Boi (Hellmayr); Santarem (Hellmayr); Para region, Rio Tocantins, Rio
Jamauchim, Rio Tapajoz, Rio Jary, Rio Jamunda (Snethlage)
3 d" 1 9 , Rio Tapajoz, various localities, east bank
1 9 , Obidos (Carnegie Mus.)

9 d" 7 9 , Rio Tapajoz, left bank (do.)
1 c? 2 9 , Santarem (do.)

Specimens from the north bank of the Amazon might prove referable
to miserabilis Chubb from British Guiana, of which we have no com-
parative material.

77S. Capsiempis flaveola flaveola (Lichtenstein)

Type locality: Bahia

Mexiana Island (Wallace, Hellmayr); Rio Tocantins, Rio Iriri, Rio Tapajoz,
Rio Maecuru, Obidos, Rio Jamunda' (Snethlage)

8 cf 3 9 , Rio Tapajoz, various localities, east bank

10 d* 5 9 , Obidos (Carnegie Mus.)

779. Stigmatura budytoides napensis Chapman

Type locality: Curary and Napo Rivers, Ecuador, Rio Tapajoz, Pinhel (Sneth-
lage) ; Santarem (Hellmayr)

1 o1 1 9 , Rio Tapajoz, east bank, Pinhy
3 o71 1 9 , Rio Tapajoz, west bank, Pinhel

780. Serpophaga hypoleuca pallida Snethlage

Type locality: Alcbbaga, Rio Tocantins

1 9 , Lago Grande, west of Santarem

2 cf 1 9 , Santarem (Carnegie Mus.)

We have no material for comparison. These specimens are distinctly
browner above than 1 hypoleuca from east Ecuador, and are also a
little larger (wing 51 mm.) ; in these two respects agreeing with the diag-
nosis of pallida from the Tocantins. They are not more purely white
below, however. A distinct buffy tinge on the flanks may or may not
be due to immaturity.

292 bulletin: museum of comparative zoology

781. Inezia subflava subflava (Sclater and Salvin)

Type locality: "Para"

Rio Tocantins to the Rio Tapajoz, both banks (Snethlage)

1 d" 3 9 , Rio Tapajoz, various localities, east bank
5 d1 3 9 , (do.) both banks (Carnegie Mus.)

782. Elainea flavogaster flavogaster (Thunberg)

Type locality: Rio de Janeiro

Apparently abundant throughout the region (all collectors)
1 cf , Para, Bosque
25 d" 4 9 1?, Rio Tapajoz, various localities, east bank
1 9 , Obidos
3 d1 4 9 , Benevides (Carnegie Mus.)
2 d" 1 9 , Santarem (do.)

783. Elainea albiceps albiceps (Lafresnaye and D'Orbigny)

Type locality: Yungas, Bolivia
Arumatheua, Rio Tocantins (Snethlage)

784. Elainea parvirostris Pelzeln

Type locality: Borba, Rio Madeira

1 9 , Rio Tapajoz, Santarem
1 cf 1 9 , Santarem (Carnegie Mus.)

785. Elainea pelzelni Berlepsch

Type locality: Lamalonga, Rio Negro, Brazil
Monte Alegre, Rio Maecuru, Obidos (Snethlage)
1 9 , Rio Tapajoz, Santarem
10 d1 6 9, Obidos (Carnegie Mus.)
6 cf 2 9 , Santarem (do.)

A rare and little known species, easily recognized by its large size
and brown coloration above. Previously unrecorded from the south
bank of the Amazon.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 293

786. Elainea cristata Pelzeln

Type locality: Goyaz, Brazil

Rio Tapajoz, Boim, and Monte Alegre (Snethlage)

1 cf, Rio Tocantins, Cametd
2 cf , 4 9 , Rio Tapajoz, Santarem

2 9 , Obidos (Carnegie Mus.)
1 9 , Benevides (do.)

6 cf 11 9, Santarem (do.)
1(^1 9, Rio Tapajoz (do.)

787. Elainea chiriquensis albivertex Pelzeln

Type locality: Ypanema, Sao Paulo

Marajo Island, Rio Tocantins, Rio Tapajoz (Snethlage)

2 cf 6 9 , Rio Tapajoz, various localities, east bank

788. Elainea gaimardii guianensis Berlepsch

Type locality: Camacusa, British Guiana

Peixe-Boi, San Antonio do Prata (Hellmayr) ; Rio Guamd (Stone) ; Para region
to right bank of Rio Tocantins, Obidos, and Rio Jamunda, Faro (Sneth-
lage)

5 o71 2 9 , Obidos (Carnegie Mus.)

789. Elainea gaimardii gaimardii (D'Orbigny)

Type locality: Yuracares, Bolivia

Left bank of Rio Tocantins to Rio Tapajoz (Snethlage)

11 c? 15 9 , Rio Tapajoz, various localities, east bank

7 c? 8 9, Santarem (Carnegie Mus.)

5 <? 19, Rio Tapajoz, both banks

790. Elainea flavivertex Sclater

Type locality: Ucayali, Peru

Mexiana Island, (Hellmayr); Monte Alegre, Rio Jamundd, Faro (Snethlage)

1 d\ near Obidos

3 d\ Rio Tapajoz, Santarem

7 o" 3 9 , Santarem (Carnegie Mus.)

294 bulletin: museum of comparative zoology

791. Elainea viridicata viridicata (Vieillot)

Type locality: Paraguay
Boim, Rio Tapajoz (Snethlage)

1 cf juv., Rio Tapajoz, Santarem

7 d1 4 9 , Santarem (Carnegie Mus.)

792. Suiriri affinis affinis (Burmeister)

Type locality: Lagoa Santa, Minas Geraes.

Serra de Erere, near Monte Alegre, north bank of Amazon, 1 cT (Snethlage)

One of the characteristic campo species, little known in our area. It
should be sought in the localities on the south bank, where this fauna
occurs.

793. SUBLEGATUS MODESTUS OBSCURIOR Todd

Type locality: Cayenne

Mexiana Island (Wallace, Hagmann); Monte Alegre and Erere (Snethlage);
Para (Stone)

4 o* 7 9 , Obidos (Carnegie Mus.)

794. SUBLEGATUS MODESTUS MODESTUS (Wied)

Type locality: Camamu, Brazil

1 c\ Rio Tapajoz, Tauary

2 a71, Santarem (Carnegie Mus.)

795. Ph.eomyias murina incompta (Cananis and Heine)

Type locality: Cartagena, Colombia
Monte Alegre (Snethlage)

5 cf, Obidos (Carnegie Mus.)

796. PlLEOMYIAS MURINA MURINA (Spix)

Type locality: Brazil

Para (Layard, Hellmayr, Stone); Mexiana Island Wallace, Hellmayr); San-
tarem (Hellmayr); Marajo Island (Hellmayr) ; Para region, Rio Tocantins,
Rio Tapajoz (Snethlage)

4 cf 8 9 1 ?, Rio Tapajoz, various localities, east bank.
1 cf, 3 9 , (do.), both banks (Carnegie Mus.)
5 9 , Santarem (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 295

797. Camptostoma obsoletum nap.eum Ridgway

Type locality: Diamantina, Santarem
Common throughout our area (all collectors)

3 c? 3 9, Rio Tapajoz, various localities, east bank
11 c? 13 9 , Obidos, Benevides, Santarem, Rio Tapajoz (Carnegie

Mus.)

798. Phyllomyias griseiceps (Sclater and Salvin)

Type locality: Babahoyo, Ecuador

3 cf 1 9 , Obidos (Carnegie Mus.)

These birds and another from French Guiana have smaller bills than
has a large series from Venezuela.

799. Tyranniscus acer Salvin and Godman

Type locality: Bartica Grove, British Guiana

Marajo Island (Snethlage); Caviana Island (Brodkorb); Para (Layard);
Peixe-Boi (Hellmayr); Para region, Rio Tocantins, Rio Jamunda (Sneth-
lage)

3 d71, Para, Val-de-Caes, and Bosque

5 d1, Rio Tapajoz, various localities, east bank

1 d* 2 9 , Obidos (Carnegie Mus.)

3 cf , Benevides (do.)

11 cf 7 9, Santarem (do.)

800. Tyrannulus elatus elatus (Latham)

Type locality: Cayenne

Para region (Spix, Hellmayr, Snethlage, Stone); Rio Tocantins and Rio

Tapajoz (Snethlage); Santarem (Ihering, Chapman and Riker); Monte

Alegre, Rio Jamunda, Faro (Snethlage)

2 cf 1 9 , near Obidos

4 cf 4 9 1 ?, Rio Tapajoz, various localities, east bank

17 cf 5 9 , Obidos, Benevides, Santarem, Rio Tapajoz (Carnegie
Mus.)

801. Ornithion inerme Hartlaub

Type locality: Bahia

Rio Guama, Para, Rio Tocantins (Snethlage) ; Rio Tapajoz (Hellmayr)

1 d71 2 9 , Rio Tapajoz, Pinhy, Tauary, Caxiricatuba

1 cf 2 9 , Benevides (Carnegie Mus.)

1 cf , Santarem (do.)

3 cf , Rio Tapajoz, both banks (do.)

296 bulletin: museum of comparative zoology

802. Pipromorpha oleaginea oleaginea (Lichtenstein)

Type locality: Bahia

Pard, San Antonio do Prata, Mexiana Island (Hellmayr)
1 d\ Rio Tapajoz, Caxiricatuba

1 cT 4 9 , Obidos (Carnegie Mus.)

2 cf 2 9 , Benevides (do.)
6 cf , Santarem (do.)

3 <? 2 9 , Rio Tapajoz, left bank

803. Pipromorpha macconnelli macconnelli Chubb

Type locality: Camacabra Creek, British Guiana
2 d\ Obidos (Carnegie Mus.)

804. Pipromorpha macconnelli amazona Todd

Type locality: Buenavista, Bolivia
Peixe-Boi, Benevides, Ourem (Hellmayr)

1 cf 1 9 , Para, Bosque and Val-de-Caes

2 c? 1 9 , Rio Tapajoz, various localities, east bank
5 d71, Benevides (Carnegie Mus.)

1 9 , Santarem (do.)
11 cf 2 9 , Rio Tapajoz, both banks (do.)

All birds in the lower Amazon valley were called oleaginea, until
Todd showed there were two species in 1921. Numerous old records,
including many by Snethlage from throughout our area, cannot be
allocated without reexamination.

Family OXYRUNCIDAE

805. Oxyruncus cristatus hypoglaucus (Salvin and Godman)

Type locality; Mt. Roraima, British Guiana
lo*l ?, Belem, Val-de-Caes

A notable range extension for this little known family of birds.
Typical cristatus is the form of southeastern Brazil, and hypoglaucus
was supposed to be confined to Mt. Roraima, of which we have a single
topotypical male. Our male from Belem has the spotting below ap-
preciably paler, but this degree of difference is covered by individual
variation in other subspecies.

GRISCOM AND GREEN WAY: BIRDS OF LOWER AMAZONIA 297

Family HIRUNDINIDAE
806. Progne subis subis (Linnaeus)

Type locality: Hudson's Bay

Cussary and Rio Jamunda, Faro (Snethlage)

1 cf , Rio Tapajoz, Goyana Island, Dec. 23, 1919, in (Carnegie Mus.)

Apparently the winter home of the Purple Martin is chiefly in
northern and eastern Brazil, but there are still very few records.

807. Progne chalybea chalybea (Gmelin)

Type locality: Cayenne

The species is common throughout our area (nearly all collectors)
1 o71 1 9, north bank of Amazon, near Obidos
5 cf 3 9,1?, Rio Tapajoz, east bank (approaching domestica)
1 9 , Obidos (Carnegie Mus.)
1 9 , Santarem (do.)
1 cT, Rio Tapajoz, Goyana Island (do.)

The Tapajoz series is so perfectly intermediate that we list them
separately as such. The birds are half way between the two races in
size, and some of the males approach domestica in having whiter un-
derpays, while others do not.

808. Progne chalybea domestica (Vieillot)

Type locality: Paraguay

1 d\ Rio Acara, Villa Acara

The Acara specimen is typical domestica in coloration, but is a little
smaller than southern birds. It is much larger, however, than Surinam
specimens of chalybea. The facts are that the lower Amazon valley is
occupied by variously intermediate birds, the majority of which are
nearer to chalybea than domestica, and there is every possibility that
domestica may be partly migratory alsQ.

S09. Phaeoprogne tapera tapera (Linnseus)

Type locality: Cayenne

Common throughout our area (Snethlage)

10 cf 3 9 , Rio Tapajoz, various localities
1 9 , Santarem, (Carnegie Mus.)
7 d" 6 9, Rio Tapajoz, Miritituba (do.)

298 bulletin: museum of comparative zoology

The southern race fusca (Vieillot) is migratory, and is recorded from
parts of Amazonia, Venezuela and British Guiana. It should, conse-
quently, be expected in our area.

[The North American Petrochelidon pyrrhonota pyrrhonota (Vieillot)
should be expected in our area on migration.]

810. Stelgidopteryx ruficollis ruficollis (Vieillot)

Type locality: Rio de Janeiro, Brazil

Apparently common throughout our area (numerous collectors)
2 d\ Para\ Val-de-Caes

4 9 , Rio Tapajoz, east bank

2 c? 5 9 , Benevides (Carnegie Mus.)

3 cf 1 9 , Santarem (do.)

2 cf 3 9 , Rio Tapajoz, both banks (do.)

811. Pygochelidon cyanoleuca cyanoleuca (Vieillot)

Type locality: Paraguay

1 9, Benevides, Sept. 13, 1918 (Carnegie Mus.)

Purely a migrant or winter visitant to our area.

812. Atticora fasciata (Gmelin)

Type locality: Cayenne

Rio Capim (Gmelin) ; Rio Capim and Cunany (Snethlage)

The wide range of this Swallow makes the lack of records in our
area quite inexplicable, unless it is migratory.

813. Atticora melanoleuca (Wied)

Type locality: Rio Belmonte, Bahia, Brazil

Rio Tocantins, Rio Xingu, Rio Jamauchim (Snethlage)

814. Riparia RiPARiA riparia (Linnaeus)

Type locality: Sweden

7 d> 9 9 , Lago Jauary, near Obidos, March 16-18, 1933
1 cf 1 9 , Obidos, March 23 and 25, 1921 (Carnegie Mus.)

Apparently previously overlooked in our area, where it is presumably
a transient visitant on migration.

GRISCOM AND GREENWAY : BIRDS OF LOWER AMAZONIA 299

815. Hirundo rustica erythrogaster Boddaert

Type locality: Cayenne

One seen in February at Para docks (Bond, in Stone) ; Para, San Antonio do
Prata, Marajo and Mexiana Islands, Rio Jamunda (Snethlage); Mexiana
Island (Wallace, Hagmann); Santarem, March 1, 1889 (Chapman and
Riker)

5 cf 7 9 1 ?, north bank of Amazon, near Obidos, March 16, 1933

1 cf4 9, Obidos, March 24-26, 1921 (Carnegie Mus.)

1 d\ Santarem, March 25, 1919 (do.)

1 9 , Rio Tapajoz, Itaituba, March 13, 1920 (do.)

There was obviously a heavy migration of North American swallows
in March, 1933, the Bank Swallows having been taken at the same
time. One of these Barn Swallows is inseparable from numerous
European specimens of rustica.

816. Iridoprocne albiventer albiventer (Boddaert)

Type locality: Cayenne

Common throughout our area (all collectors)

8 d71 4 9 , Rio Tapajoz, various localities, east bank
1 9 , Obidos (Carnegie Mus.)

2 d\ Santarem (do.)

1 c? 1 9 , Rio Tapajoz, Villa Braga (do.)

This series agrees with others from the Rio Tocantins and topotypes
from Surinam in color. The wings of males from Surinam are 99-104
mm., from the lower Amazon 99-107 mm., and from southeastern
Brazil 107-108 mm. Southern birds are progressively larger, but
there is no color change. A fine adult male from Amazonian Ecua-
dor has the wing 108 mm. It differs radically in color from eastern
birds in having the head bluish green, the back green with no bluish
tinge, and the tail with no steel blue reflections, the feathers dusky to
hoary stone gray. Should further specimens confirm these differences
and show a definite geographic range, there would be a perfectly valid
subspecies, which would probably have to be called aequatorialis
Lawrence, though we quite agree with Hellmayr (1935, p. 73, footnote)
that the characters originally claimed have no validity whatever.

300 bulletin: museum of comparative zoology

Family TROGLODYTIDAE
817. Heleodytes turdinus hypostictus (Gould)

Type locality: Rio Ucayali, Peru
Left bank, Rio Tapajoz (Snethlage)

1 9 , Rio Tapajoz, Boim, wing 84 mm.

818. Heleodytes turdinus subsp.

Santarem (Chapman and Riker) ; Rio Xingii and Rio Tocantins (Snethlage)

I 9 , Obidos (Carnegie Mus.)
13 d* 7 9 , Santarem (do.)

Lack of topotypes from eastern Peru makes any final disposition of
lower Amazonian birds impossible. While hypostictus in the aggregate
differs from typical turdinus in more heavily spotted underparts,
specimens are well known to be exceedingly variable. There is, how-
ever, a marked increase in size, the break occurring on the Rio
Tapajoz. The table of wing measurements below brings this out
clearly.

d" 9

II Rio Solimoes 87-92 79-87
8 Rio Purus 85-88 82-83
1 Rio Tapajoz, Boim 84
1 Obidos 91

20 Santarem 91-97 88-91

In color characters, the large birds from Santarem agree with "hy-
postictus", but the female from Boim is in this respect indistinguishable
from turdinus. The meager evidence before us suggests that adequate
series covering the enormous range assigned to hypostictus will show
the desirability of recognizing several subspecies differing in average
characters.

819. Odontorchilus ctnereus (Pelzeln)

Type locality: Salto de Girao, Rio Madeira

Rio Iriri, a tributary of the Rio Xingii (Snethlage); Rio Tapajoz, Miritituba
and Colonia do Mojuy (2 9 in Carnegie Museum examined)

This very rare species is still known from only four specimens.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 301

820. Thryothorus leucotts tceniopterus Ridgway

Type locality: Diamantina, Santarem, Brazil

Common throughout our area (all collectors) ; recorded from Marajo, but not
Mexiana Island

2 <j\ near Obidos

8 cf 3 9 , Rio Tapajoz, Tauary and Pinhy

5 d\ Obidos (Carnegie Mus.)
9c?6 9„ Santarem (do.)

4 cf 1 9 , Rio Tapajoz, left bank (do.)

With 37 specimens before us of typical albipectus, and 38 specimens
representing tceniopterus, we agree with Todd in recognizing the latter
as a valid subspecies. While freely admitting the great amount of
individual color variation in this species, these great series show con-
clusively that typical albipectus is more richly colored below, more cin-
namomeous or tawny. There is also a marked size difference, which
alone would entitle tceniopterus to recognition. Cayenne birds average
5-8 mm. longer wing, with a notably longer bill.

821. Thryothorus genibarbis genibarbis Swainson

Type locality: Bahia, Brazil

Throughout our area on the south bank of the Amazon from the Para region to
the Rio Tapajoz

6 a71, Para, Bosque, and Val-de-Caes
11 cf, 11 9, Benevides (Carnegie Mus.)

8 cf 5 9 , Rio Tapajoz (do.)

This variable species is widely spread south of the Amazon. The
typical form is replaced by juruanus on the left bank of the Rio Ma-
deira westward.

822. Thryothorus coraya coraya (Gmelin)

Type locality: Cayenne

Rio Jamunda, Obidos and Rio Jary (Snethlage)
6 cf 6 9 , Obidos (Carnegie Mus.)

823. Thryothorus coraya herberti Ridgway

Type locality: Diamantina, Santarem

Abundant from the Rio Tocantins to the Rio Tapajoz

16 c? 13 9 1 ?, Rio Tapajoz, various localities

4 cf 3 9, (do.), both banks (Carnegie Mus.)

17 d31 10 9, Santarem (do.)

302 bulletin: museum of comparative zoology

The relationships between this species and genibarbis on the south
bank of the i\mazon are as yet unrecorded. The latter is abundant
in the Para region, and the present species unknown there, in spite of a
wider geographic range.

824. Troglodytes musculus clarus Berlepsch and Hartert

Type locality: Bartica Grove, British Guiana

Common in clearings throughout our area (all collectors)

1 cf , Para, Bosque

4 cf 1 ?, Rio Tapajoz, various localities

2 cf 2 9 , Lago Grande

3 cf 1 9 , Obidos (Carnegie Mus.)

1 cf 3 9 , Benevides (do.)

3 cf 3 9 , Rio Tapajoz, Apacy

The males from Para are more richly colored than typical clarus
and distinctly approach musculus, which ranges as far north as Ceara.
So far as we can find, House Wrens from Maranhao have yet to be
critically studied.

825. Microcerculus bambla bambla (Boddaert)

Type locality: Cayenne

2 cf , Obidos (Carnegie Mus.)

This species is new to Brazil. The two adults are not separable from
Cayenne topotypes.

826. Microcerculus marginatus marginatus (Sclater)

Type locality: "Bogota"

San Antonio do Prata and Peixe-Boi (Hellmayr); Para region, various localities
(Snethlage)

I 9 , Rio Tapajoz, Tauary

7 cf 4 9 , Benevides (Carnegie Mus.)
6 cf , Santarem (do.)

5 cf , Rio Tapajoz, Villa Braga, and Miritituba (do.)

Inseparable from 29 specimens from the Rio Purus and Rio Solimoes.

827. Leucolepis arada arada (Hermann)

Type locality: Cayenne

Obidos and Rio Jary (Snethlage)

II cf 6 9, Obidos (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 303

828.. Leucolepis arada griseolateralis Ridgway

Type locality: Diamantina, Santarem, Brazil

Santarem (Chapman and Riker) ; Rio Jamauchim (Snethlage)

5 cf 3 9, Rio Tapajoz, Tauary, Caxiricatuba, Pataua
16 cf 15 9 , Santarem (Carnegie Mus.)

9 cf 10 9 , Rio Tapajoz, right bank (do.)

829. Leucolepis arada interposita Todd

Type locality : Villa Braga, left bank of Rio Tapajoz
Also Apagy (Todd), west to the right bank of the Rio Madeira
12 a1 12 9 , the type series in Carnegie Mus.

Family MIMIDAE

830. Mimus gilvus antelius Oberholser

Type locality: Bahia
Cajutuba (Natterer)

831. Mimus saturninus saturninus (Lichtenstein)

Type locality: Rio Tapajoz, Pard,

Santarem (Allen, Riker and Chapman, Snethlage) ; Monte Alegre (Snethlage)

The rarity of this genus in lower Amazonia is just what would be
expected of open country birds in a prevailingly forested area. M.
saturninus is a common bird in the campo country south of our area
and is currently called f rater Hellmayr, but the validity of this race
should be checked with an adequate series of typical saturninus.

832. Donacobius atricapillus" atricapillus (Linnaeus)

Type locality: eastern Brazil

Common on the north side of the Amazon, the Para region, and Mexiana Island
(all collectors); Santarem (Chapman and Riker); unrecorded from Marajo
Island and anywhere else on the south bank between Para and the Tapajoz
3 d> 2 9 , near Obidos
1 a", Obidos (Carnegie Mus.)
2^29, Santarem (do.)

304 bulletin: museum of comparative zoology

Family TURDIDAE

833. Turdus phaeopygus poiteaui Bonaparte

Type locality: Cayenne

Obidos (Todd, as cayennensis Todd).

1 cf 4 9 , Obidos (Carnegie Mus.)

834. Turdus phaeopygus coloratus Todd

Type locality: Colonia de Mojuy, Santarem.

Para region, common (all collectors); Rio Acara Hellmayr); Rio Tocantins
(Snethlage)

6 c? 1 9, Rio Tapajoz, east bank

1 d71 1 9 , Benevides (Carnegie Mus.)

3 d1 3 9 , Santarem (do.)

Whether albicollis and phacopyus are conspecifie or not is a matter of
opinion. Mr. Todd (Proc. Biol. Soc. Wash., 44, 1931, p. 51) considers
specimens from the Para region to be a third subspecies which he does
not describe for lack of material, but these birds seem to us to be inter-
mediate between the two races listed here, which appear very
distinct.

835. Turdus nudigenis gymnophthalmus Cabanis

Type locality: Cayenne, designated by Berlepsch
Amapa, Rio Jamunda, Faro (Snethlage)

A fine series from Cayenne is easily separable from an equally fine
series from Venezuela (typical nudigenis) in being darker, more umber
brown on throat and chest below.

836. Turdus nudigenis extimus Todd

Type locality: Santarem, Brazil, Cussary (Snethlage)
3 d" 3 9 , Santarem, (Carnegie Mus.)

There is disagreement between Messrs. Todd and Hellmayr on this
species also. The former assigns all records from the Lower Amazon
to extimus. The latter restricts extimus to the south bank without,
however, even having seen any Amazonian specimens. As usual, the
material in the Carnegie Museum fully endorses Mr. Todd's conclu-
sions. The present subspecies is strikingly distinct.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 305

837. Turdus fumigatus fumigatus (Lichtenstein)

Type locality: southeastern Brazil

Para region common (all collectors); Caviana Island (Brodkorb); Mexiana
Island (all collectors) ; RioTocantins (Snethlage) ; Santarem (Chapman and
Riker)

3 cf 2 9 , Rio Acara, Acara
2 cf, Rio Tapajoz, Pinhy
1 d\ Obidos (Carnegie Mus.)

1 cf 3 9 , Santarem (do.)

2 cf 4 9 , Rio Tapajoz, left bank (do.)

Replaced by hauxwelli Lawrence from the Rio Madeira westward.

838. Turdus amaurochalinus Cabanis

Type locality: Brazil

1 d\ Para (Snethlage)

This is a common thrush south of our area, ranging north along the
coast of Maranhao. One wonders if the record from Para is really au-
thentic.

839. Turdus leucomelas ephippialis Sclater

Type locality: Bogata

Amapa, Monte Alegre, Rio Jamunda (Snethlage)

840. Turdus leucomelas albiventer Spix

Type locality: Para

Common, Mexiana, Marajo, Para region (all collector); Caviana Island
(Brodkorb); Rio Tapajoz (Snethlage)

1 cf 1 9 , Para, Bosque

11 c? 8 9 , Rio Tapajoz, various localities
1 9 , Benevides (Carnegie Mus.)

2 cf 3 9 , Santarem (do.)

Family SYLVIID.E

841. Polioptila paraensis Todd, 1937

Type locality: Benevides, Para, Brazil

1 cT nearly adult (possibly a 9 ), type locality

This interesting little bird will probably prove to be related to the
guiancnsis-schistaceigula complex when further specimens turn up.
Above all, positively adult males are badly needed.

306 bulletin: museum of comparative zoology

842. Polioptila plumbea plumbea (Gmelin)

Type locality: Surinam

Pard region (Hellmayr, Snethlage, Stone); Marajo Island (Hellmayr, Sneth-
lage); Santarem (Hellmayr); Monte Alegre (Snethlage)

1 c71, Para, Val-de-Caes

3 cT 1 9 , near Obidos

1 d1, Rio Tapajoz, Pinhy

7J2 9, Obidos (Carnegie Mus.)

1 9, Benevides (do.)
17 d 9 9 , Santarem (do.)

843. Ramphocaenus melanurus austerus Zimmer, 1937

Type locality: Pedral, Baiao, Rio Tocantins, Brazil

Para region, numerous records and localities (Sclater, Hellmayr, Snethlage,
Stone); Rio Tocantins, east bank (Snethlage)

2 d\ Benevides (Carnegie Mus.)

844. Ramphocaenus melanurus albiventris Sclater

Type locality: Surinam

Rio Jary, S. Antonio de Cachoeira (Snethlage) ; Faro (Zimmer)

845. Ramphocaenus melanurus amazonum Hellmayr

Type locality: Teffe, Rio Solimoes, Brazil

Common from the west bank of the Rio Tocantins westward (Snethlage,
Zimmer)

25 cf 6 9 , Rio Tapajoz, various localities, east bank.
10 d* 1 9 , Santarem (Carnegie Mus.)
5 cf 3 9 , Rio Tapajoz, both banks (do.)

Family MOTACILLIDAE

846. Anthus lutescens lutescens Pucheran

Type locality: Rio de Janeiro, Brazil

Para (Snethlage, Stone); Benevides, Quati-puru (Snethlage); Mexiana Island
(Wallace, Hellmayr); Marajo Island (Hellmayr, Snethlage); Rio Maecuru
(Snethlage); Santarem (Chapman and Riker)

7 o71 7 9 , Lago Grande, west of Rio Tapajoz
20 c? 8 9 , Santarem (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 307

Family CYCLARHIDAE

847. Cyclarhis gujanensis gujanensis (Gmelin)

Type locality: French Guiana

Pard, region and entire area south of the Amazon to the Rio Tapajoz (numerous
records, all collectors); Monte Alegre (Snethlage)
1 9 , Para, Bosque
7 cf 5 9 , Rio Tapajoz, various localities, east bank.
2 d* 1 9 , Obidos (Carnegie Mus.)
2 cf 1 9 , Benevides (do.)
2 cf 3 9 , Santarem (do.)
2 cf 1 9 , Rio Tapajoz, Aveiros, Miritituba (do.)

Family VIREOLANIIDAE
848. Smaragdolanius leucotis leucotis (Swainson)

Type locality: "Africa"

1 cf, S. Antonio de Cachoeira, Rio Jary (Snethlage)

849. Smaragdolanius pulchellus simplex (Berlepsch)

Type locality: Santa Elena, Rio Jamauchim, Brazil.

Recorded only from the type locality, Boim on the Rio Tapajoz, and Aruma-

theua on the Rio Tocantins (Snethlage) in our area, and Barao Melgaco,

northern Matto Gross (Naumburg)

2 cf 1 9 , Santarem (Carnegie Mus.)

1 cf 1 9 , Rio Tapajoz, Bella Vista, Miritituba (do.)

There is as yet no record of this species between the Tapajoz and
the Rio Purus. Five specimens from this river in the Carnegie Museum
differ from bolivianus (1 spec.) in being brighter greener olive, the cap
more bluish, less slaty gray with more of a greenish tinge forward, and
the throat and breast appear a brighter more golden yellow.

Family VIREONIDAE

850. Vireo chivi chivi (Vieillot)

Type locality: Paraguay

Santarem and Miritituba, Rio Tapajoz (Todd)

2 9 , Rio Tapajoz, Pinhy and Caxiricatuba, June and July

2 cf , Santarem, April 30, and May 12, 1919. (Carnegie Mus.)
1 cf , Rio Tapajoz, Miritituba, February 23, 1920 (do.)

A migrant to our area from the south.

308 bulletin: museum of comparative zoology

851. Vireo chi vi vividior Hellmayr and Seilern

Type locality: Caparo, Trinidad
Localities on the Rio Tapajoz (Todd)

1 d\ Rio Tapajoz, Pinhy, March 5, 1933

1 9 , Rio Tapajoz, Tauary, November 21, 1932

A migrant to our area from the north.

852. Vireo chivi solimoensis Todd

Type locality: Sao Paulo de Olivenca, Rio Solimoes

2 cf, Para, Val-de-Caes

27 cf 13 9 , Rio Tapajoz, various localities, east bank
5 c? 7 9, Santarem (Carnegie Mus.)
5 cf, Rio Tapajoz, both banks (do.)

There is hopeless disagreement at the moment on the proper treat-
ment of this Vireo and its racial variation, (cf. Hellmayr, 1935, pp.
136-143, and Todd, Auk, 1931, pp. 407-412). Hellmayr and Zimmer
claim that chivi is conspecific with the virescens group. Todd not only
denies this, but divides chivi into two species, all the race's of chivi from
the Amazon Valley northward being a second species, the earliest
name for which is caucae. Mr. Todd's position is based on very fine
series of specimens of every proposed form, and the discovery of chivi
and other northern subspecies at the same place and season in Ama-
zonia. Hellmayr has fortunately examined some of these critical
specimens, and while admitting the justification of Mr. Todd's identifi-
cations, suggests (1) that there is a higher degree of individual varia-
tion or (2) that the examples of chivi might be migrants. Mr. Todd's
review leaves the resident bird (if any) of lower Amazonia unsettled,
but Dr. Hellmayr maintains that all specimens seen by him east to
Para cannot be separated from solimoensis.

The very large series before us throw some new light on these ques-
tions. The very great majority of the birds are solimoensis both in
color and size, and as these are the breeding birds, it is solimoensis
which is the resident race just as Dr. Hellmayr claims. On the other
hand, seven specimens are readily picked out of this series by virtue of
their much larger size. Two in their brighter and lighter green color
are clearly vividior. Five others are strikingly grayer and duller, and
are inseparable from typical specimens of chivi. These birds conse-
quently endorse Todd's conclusions and also Dr. Hellmayr's admission
that certain specimens are indeed strikingly like chivi. We disagree

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 309

most emphatically, however, in the identification of Todd's specimens
of vividior from the Lower Amazon, and refer all of them to solimoensis.

We see no occasion for adopting the extreme conclusions of either
author. We cannot agree in reducing chivi to the virescens "formen-
kreis," but see no necessity for splitting chivi into two species for the
reasons alleged above. Field experience has now overwhelmingly
demonstrated that flawviridis and calidris, the other two tropical
species of this group, are highly migratory. It is, we think, a much
more common sense position to regard solimoensis as the resident race,
to infer that vividior is migratory from the north, and chivi migratory
from the south. The new evidence before us, on Mr. Todd's reasoning,
would now require chivi to be split into three instead of two species,
which seems to us a patent absurdity.

The type series of griseola Todd seems separable to us on the charac-
ter claimed, though Hellmayr is doubtful if griseola and solimoensis
are really separable. He has also examined Todd's 2 Obidos specimens
of griseola and cannot distinguish them from solimoensis; neither can
we.

Needless to say, it is quite impossible to allocate any of the older
records and references subspecifically without reexamination. The
species is recorded abundantly by all collectors from every part of our
area except Marajo Island.

853. Vireo calidris barbatula (Cabanis)

Type locality: Cuba

1 9 , Rio Tapajoz, Caxiricatuba, July 31, 1932

1 9 , Obidos, January 20, 1921 (Carnegie Mus.)

1 d", Rio Tapajoz, Villa Braga, December 10, 1919 (do.)

We wish to call attention to the surprising date of capture (July) of
one specimen. It is in exceedingly worn and frowsy plumage, but is
readily identified as barbatula rather than typical calidris by the
much smaller bill. The color characters so obvious in fresh birds are
completely obscured in this specimen, but it cannot be barbadensis, as
the superciliary is buffy, not grayish.

Winter specimens of this West Indian vireo are exceedingly few.
Recently collected specimens in American Museums from eastern
Panama and the north coasts of Colombia and Venezuela have proved
to be barbatula. There are a few older winter records from South
America in European museums, but these specimens have never been
critically determined subspecifically, with the single exception of

310 bulletin: museum of comparative zoology

Natterer's specimen from Borba on the Rio Madeira, which Hellmayr
shows is straight calidris. (Novit. Zool., 1910, p. 268). This, therefore,
is the first definite record of barbatula from Brazil.

854. Hylophilus thoracicus griseiventris
Berlepsch & Hartert

Type locality: Suapure, Caura River, Venezuela
Obidos (Hellmayr, 1935)

1 d\ Obidos (Carnegie Mus.)

855. Hylophilus semicinereus semicinereus Sclater & Salvin

Type locality: Para, Brazil

Common on the south bank of the Amazon from the Para region westward (all
collectors)

9 d1 11 9 , Rio Tapajoz, various localities, east bank

5 d" 3 9, Benevides (Carnegie Mus.)

6 <d\ Rio Tapajoz, both banks, (do.)

856. Hylophilus semicinereus viridiceps (Todd)

Type locality: Pied Saut, French Guiana

Rio Jary, Obidos, Rio Jamunda (Snethlage and Todd)

4 d\ Obidos (Carnegie Mus.)

857. Hylophilus pectoralis Sclater

Type locality : Villa Bella de Matto Grosso

Mexiana Island (Hagmann, Hellmayr); Marajo Island, Quati-puru, Rio To-
cantins, Arumanduba, Monte Alegre (Snethlage)
1 d1 1 9 , Rio Tapajoz, Santarem

5 d1 3 9 , (do.) (do.) (Carnegie Mus.)

858. Hylophilus muscicapinus muscicapinus Sclater & Salvin

Type locality: Oyapock, French Guiana
Rio Jary and Obidos (Snethlage)

14 d1 3 9 , Obidos (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 311

859. Hylophilus muscicapinus griseifrons (Snethlage)

Type locality : Villa Braga, Rio Tapajoz

Otherwise known only from Boim (Snethlage), the Rio Madeira and northern
Matto Grosso

6 d1 3 9 , Rio Tapajoz, Villa Braga, and Apacy (Carnegie Mus.)

860. Hylophilus brunneiceps inornatus (Snethlage)

Type locality: Cameta, Rio Tocantins

Also Rio Jamauchim (Snethlage) and Rio Tapajoz (Todd)

1 d\ Rio Tapajoz, Caxiricatuba

4 o" 1 9, Santarem (Carnegie Mus.)

5c?2 9, Rio Tapajoz, Miritituba (do.)

861. Hylophilus hypoxanthus albigula (Chapman)

Type locality: Santa-Julia, Rio Iriri, Rio Xingu
Rio Xingu (Snethlage)

862. Hylophilus luteifrons Sclater

Type locality: Bartica Grove, British Guiana
San Antonio de Cachoeira, Rio Jary (Snethlage)
1 d\ Obidos (Carnegie Mus.)

863. Hylophilus rubrifrons rubrifrons Sclater and Salvin

Type locality: Para

Para, (Natterer, Snethlage, Stone) ; other localities near Para and Peixe- Boi

(Snethlage)

3 d" 2 9 , Benevides (Carnegie Mus.)

This race is probably restricted to the area east of the Rio Tocantins.

864. Hylophilus rubrifrons lutescens (Snethlage)

Type locality: Boim, Rio Tapajoz

Rio Xingu, Victoria and Rio Tapajoz, Villa Braga (Snethlage)
1 d1 1 9 , Rio Tapajoz, Pataua and Tauary
6 d1 1 9 , Rio Tapajoz, left bank (Carnegie Mus.)
9 d1 1 9 , (do.) , right bank and Santarem (do.)

312 bulletin: museum of comparative zoology

The series from the right bank of the Tapajoz is intermediate.
Above, these birds are nearer lutescens, but they have the distinctly
brownish throat and breast of rubrifrons.

Family COEREBIDAE

865. Chlorophanes spiza spiza (Linna?us)

Type locality: Surinam

Peixe-Boi (Hellmayr); Utinga (Beebe); Para (Stone); Para region, Rio Tocan-
tins, Rio Jamauchim (Snethlage)

3 d\ Para, Bosque

4 d\ Obidos (Carnegie Mus.)
6c?l 9, Benevides (do.)

1 d\ Rio Tapajoz, Villa Braga (do.)

866. Cyanerpes cyaneus cyaneus (Linnpeus)

Type locality: Surinam

Para region, abundant (all collectors); Rio Tocantins, Monte Alegre, Obidos
(Snethlage)

21 c? 14 9 , Rio Tapajoz, various localities, east bank
1 c? 2 9, Obidos (Carnegie Mus.)
3d" 19, Benevides (do.)

5 d? 4 9, Santarem (do.)
Id1 19, Rio Tapajoz (do.)

867. Cyanerpes caeruleus caeruleus (Linnaeus)

Type locality: Surinam

Common throughout our area (all collectors), but unrecorded from Marajo
and Mexiana Islands

11 d" 5 9 , Para, Bosque, and Val-de-Caes

6 d\ Benevides (Carnegie Mus.)

1 d1 2 9 , Santarem (do.)

2 d" 2 9 , Rio Tapajoz, Villa Braga (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 313

868. Dacnis cayana cayana (Linnaeus)

Type locality: Cayenne

Common almost throughout our area (all collectors), but as yet unrecorded
from Mexiana Island, although collected on Mara jo
1 cf 3 9 , Para, Bosque
21 cf 8 9 , Rio Tapajoz, various localities, both banks
1 cf, Obidos (Carnegie Mus.)
4 cf 1 9 , Benevides (do.)

3 cf 4 9 , Santarem (do.)

4 d" 2 9 , Rio Tapajoz (do.)

869. Dacnis lineata lineata (Gmelin)

Type locality: Cayenne

Para (Stone); Igarape-Assu, Benevides (Hellmayr); Para, Rio Acara (Sneth-
lage)

1 cf , Obidos (Carnegie Mus.)

This species is better known as angelica Bonaparte. It is certainly
curious that it is unrecorded between Para and the Rio Madeira.

870. Dacnis flaviventer Lafresnaye and D'Orbigny

Type locality: Yuracares, Bolivia

Rio Iriri, Rio Tapajoz, Rio Jamauchim (Snethlage)

1 cf , label lost, but presumably Rio Tapajoz

1 cf 1 9 , Santarem (Carnegie Mus.)

3 cf 1 9 , Rio Tapajoz, Goyana Island, Apacy (do.)

871. Coereba luteola chloropyga (Cabanis)

Type locality: Bahia

Maraca, Rio Jary, Monte Alegre, Rio Jamauchim (Snethlage) and Caviana
Island (Brodkorb) all as minima Bonaparte from Cayenne; abundant
throughout our area south of the Amazon (all collectors.)
7 cf 4 9 , Para, Bosque and Val-de-Caes
1 1 cf 7 9 2 ?, Rio Tapajoz, various localities, east bank

2 cf 1 9 , Obidos (Carnegie Mus.)
1 cf 2 9 , Benevides (do.)

1 cf 2 9 , Santarem (do.)

3 cf , Rio Tapajoz (do.)

314 bulletin: museum of comparative zoology

This fine series averages slightly smaller and darker slate, less
brownish above, than birds from eastern and southeastern Brazil.
They thus slightly approach the Guiana race. As Hellmayr has duly
noted, this bird is exceedingly variable individually, and we agree that
there is really no need for more than two races in the area outlined
above. It is merely unfortunate, as so often happens, that Bahia topo-
types of chloropyga are not typical of the southern extreme. Obidos
birds are inseparable, and show no approach to minima, to which Hell-
mayr assigns them.

872. Ateleodacnis speciosa spectosa (Temminck)

Type locality: Rio de Janeiro

Marajo Island (Snethlage), the northeastern limit

873. Ateleodacnis speciosa amazonum Hellmayr

Type locality: Tarapato, northern Peru

Erere, Serra de Paituna, Rio Tocantins, Rio Tapajoz (Snethlage)

2 d" 1 9 , Obidos (Carnegie Mus.)

Snethlage has recorded her birds as Dacnis speciosa and D. analis
Lafr. and d'Orb. It remains to be determined whether some or all of
these records really belong to amazonum or not.

874. Ateleodacnis bicolor bicolor (Vieillot)

Type locality; Cayenne

Cajutuba (Natterer); Mexiana Island (Hagmann); Para, Marajo Island,
Aqriqui, Arumanduba (Snethlage)

3 d" imm., near Obidos

5 d" 2 9 , Obidos (Carnegie Mus.)
8 a1 4 9 , Santarem (do.)

875. Ateleodacnis bicolor minor Hellmayr

Type locality: Rio Madeira, right bank below Rio Mahisi

Santarem (Chapman and Riker) ; Villa Bella Imperatriz (Zimmer)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 315

Family COMPSOTHLYPIDAE

876. Dendroica aestiva aestiva (Gmelin)

Type locality: Canada

Chaves, Marajo Island, 1 d" (Snethlage)

877. Geothlypis aequinoctialis aequinoctialis (Gmelin)

Type locality: Cayenne

Para, (Stone); Caviana Island (Brodkorb); Mexiana Island (Wallace, Hag-
mann, Hellmayr, Snethlage); S. Antonio do Prata and Rio Xingu (Sneth-
lage)

3 d\ Santarem (Carnegie Mus.)
2 cf 1 ?, Obidos (do.)

878. Granatellus pelzelni pelzelni Selater

Type locality: Destacamento de Ribeirao, Rio Madeira

Santarem (Hellmayr); Rio Tocantins, Rio Tapajoz and Rio Jary (Snethlage).

4 d* 2 9 , Rio Tapajoz, both banks

2 c? 2 9, (do.) , (do.) (Carnegie Mus.)

1 d\ Santarem (do.)

879. Granatellus pelzelni paraensis Rothchild

Type locality: S. Antonio do Prata, Para
Also Santa Maria de San Miguel, Rio Guama (Snethlage)
1 9 , Rio Acara, Acara

880. Basileuterus rivularIs mesoleucus Sclater

Type locality : Demerara, British Guiana

Rio Muraiteua (Stone); Para, Peixe-Boi, San Antonio do Prata (Snethlage);
Rio Tapajoz, Villa Braga, and Colonia de Mojuy (Todd)
1 d" 1 9 , Rio Tapajoz, east bank

1 9 , Benevides (Carnegie Mus.)
1 cf 1 9 , Santarem (do.)
1 d\ Villa Braga (do.)

•)

16 bulletin: museum of comparative zoology

Family ICTERIDAE

881. Gymnostinops bifasciatus (Spix)

Type locality: Maranhao and Para

Para (Natterer, Spix, Snethlage); Peixe-Boi (Hellmayr); Arumatheua, Rio
Tocantins (Snethlage)

1 d1 ad., "Lower Amazons"

882. Gymnostinops neiv^e Snethlage

Type locality: islands in Rio Iriri, a tributary on the west bank of the Rio

Xingu
Santarem (Allen)

3 9 , Santarem (Carnegie Mus.)

This remarkable bird is almost half way between the chestnut and
blackish bifasciatus and the chestnut and yellow yuracares. Further
specimens of this genus from the south bank of the Amazon are badly
needed. Hellmayr has already commented on the greener, less yellow
coloration of Rio Madeira specimens of yuracares which suggests the
first step in an approach to neivae.

883. OSTINOPS DECUMANUS DECUMANUS (Pallas)

Type locality: Surinam

Rio Muria (Natterer); Rio Capim (Goeldi); Marajo Island, Amapa, Cunany,
Rio Tocantins (Snethlage); Santarem (Chapman and Riker)
6 cf 2 9 , Rio Tapajoz, east bank
1 9 , Obidos (Carnegie Mus.)
1 cf 1 9 , Santarem (do.)

There would seem to be a distinct decrease in size southward in
eastern South America. Measuring only adult males and females, we
get the following wing lengths —

Surinam 1 cf 141 4 9 74-80

Rio Tapajoz 5 & 124-133 2 9 60-65

Bahia 1 9 66

This difference could perhaps receive nomenclatural recognition
were size not equally variable in other parts of the range of the species.
Birds from upper Amazonian Brazil and east Ecuador are fully as large

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 317

as Surinam specimens. Birds from western Colombia and Panama,
however, are small again like Brazilian birds, and there are no color
differences if birds of similar age are compared. The youngest birds
are more chestnut, less blackish above; the next stage, particularly
in males, is dull blackish with a short crest, and only fully adult males
are glossy black, with an elongated crest, and radically larger than
females.

884. Ostinops viridis viridis (P. L. S. Miiller)

Type locality: Cayenne

Para (Natterer, Wallace, Layard); Caviana Island (Brodkorb); Peixe-Boi,
San Antonio do Prata (Hellmayr); Rio Capim (Goeldi); Capanema, Rio
Guaraa, Rio Tocantins (Snethlage)
3 d, Rio Tapajoz, Boim
1 9 , Obidos (Carnegie Mus.)
1 d, Santarem (do.)
1 d" 2 9 , Rio Tapajoz, Villa Braga (do.)

These birds are much nearer typical viridis than the very distinct
flavescens Bangs and Penard of the Peruvian Amazons. They differ
from a series from French and Dutch Guiana, however, in being dis-
tinctly yellower on throat and chest, thus showing a slight approach to
flavescens.

885. Cacicus cela (Linnaeus)

Type locality: Guiana

Common throughout our area (all collectors)

5 d 5 d, Para, Bosque and Val-de-Caes

2d 19, Rio Tocantins, Cameta

Id" 19, Rio Xingii, Tapara and Porto do Moz
14 d1 11 9, Rio Tapajoz, east bank

1 d\ Obidos (Carnegie Mus.)

1 d1, Santarem (do.)

1 cf 19, Rio Tapajoz, Apacy

886. Cacicus haemorrhous haemorrhous (Linnteus)

Type locality: Brazil in error, = Cayenne

Para region, common (all collectors); no records from Marajo or Mexiana
Islands; Villa Braga, Rio Tapajoz (Snethlage)
3 d 2 9 , Rio Tapajoz, Tauary and Boim
1 9 , Benevides (Carnegie Mus.)

2 d, Santarem (do.)

7 d 2 9 , Rio Tapajoz, Villa Braga (do.)

318 bulletin: museum of comparative zoology

887. Archiplanus solitarius (Vieillot)

Type locality: Paraguay

Marajo Island (Snethlage); Caviana Island (Brodkorb); Arumanduba, Monte
Alegre, Rio Jamunda (Snethlage); Santarem (Chapman and Riker)

2 cf 1 9 , near Obidos

1 d1, Rio Tapajoz, Santarem

7 d> 4 9 , Santarem (Carnegie Mus.)

This Cacique occurs in our area, only where there is some open coun-
try. It will unquestionably be found locally elsewhere.

888. Psomocolax oryzivorus oryzivorus (Gmelin)

Type locality: Cayenne

Para (Natterer, Wallace, Stone); S. Antonio do Prata, Ipitinga (Hellmayr);
Rio Capim (Goeldi); Rio Guama, and Monte Alegre (Snethlage) ; Santarem
(Chapman and Riker)

4^19, Rio Tapajoz, east bank
2 9 , Santarem (Carnegie Mus.)

889. Molothrus bonariensis riparius Griscom and Greenway,

1937

Type locality: Pinhy, Rio Tapajoz, Brazil

Cajutuba (Natterer); Mexiana Island (Wallace); Marajo Island (Hellmayr);
Quati-puru, Amapa, Monte Alegre, Cussary, Rio Tapajoz (Snethlage);
Santarem (Pelzeln, Chapman and Riker, Hellmayr)

13 a71 3 9 , Rio Tapajoz, various localities, east bank
1 9 , Lago Grande, west of Rio Tapajoz

1 d\ Obidos (Carnegie Mus.)
5 cf 3 9 , Santarem (do.)

2 d> 1 9 , Rio Tapajoz, Goyana Island (do.)

It seems to us clearly proved that lower Amazon Valley Cowbirds
are neither typical bonariensis nor the dwarf minimus Delmas from
Tobago, Trinidad, and the Guianas. Both these races have been
ascribed to our area. The females are not dimorphic, apparently, as in
bonariensis, and there are perfectly good size characters in addition,
though riparius is intermediate in this respect.

890. Agelaius cyanopus Vieillot

Type locality: Paraguay

Arumanduba (Snethlage); the extreme northern limit

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 319

891. Agelaius icterocephalus icterocephalus (Linnaeus)

Type locality: Cayenne

Amapd, Arumanduba, Monte Alegre, Marajo Island (Snethlage); Para
(Stone); Santarem (Chapman and Riker)
12 cf 8 9 , Santarem (Carnegie Mus.)

892. Agelaius ruficapillus frontalis Vieillot

Type locality: Cayenne

Mexiana Island (Hagmann) ; Rio Guama (Snethlage)

893. Leistes militaris militaris (Linnaeus)

Type locality: Surinam

Common in marshes and open country throughout our area (all collectors).

8 cf 3 9 , Rio Tapajoz, localities on both banks
2 cf, Santarem (Carnegie Mus.)

1 cf , Itaituba, Rio Tapajoz (do.)

894. Gymnomystax mexicanus (Linnaeus)

Type locality: Cayenne

Recorded from practically all parts of our area, where marshes and open

country occur

1 cf 2 9 , Lago Grande

9 cf 4 9 , Rio Tapajoz, both banks

1 9 , Obidos (Carnegie Mus.)
1 cf 2 9 , Santarem (do.)

895. Sturnella magna subspecies

Savannahs of Rio Tocantins (Snethlage, 1926); Marajo Island (Brodkorb)

American ornithologists have generally overlooked Snethlage's
record of the genus Sturnella ranging south to Lower Amazonia. She
reported her birds as meridionalis Sclater, which is, of course, out of the
question. On the other hand, Brodkorb identified his specimens as
praticola Chubb of the lowlands of British Guiana, certainly the proper
allocation until a series proves them separable. The completely isolated
birds on the savannahs of the Tocantins might be something else again.

320 bulletin: museum of comparative zoology

896. Icterus cayanensis cayanensis (Linnaeus)

Type locality: Cayenne

Bemfica, Ipitinga, (Hellmayr); Castanhal (Stone); Para and Marajo Island
(Wallace); Para, Providencia, San Antonio do Prata, Rio Tocantins
(Snethlage)

1 d\ Rio Tapajoz, Boim
1 cf 1 9 , Rio Tapajoz, Santarem
1 9 , Obidos (Carnegie Mus.)
1 cf 2 9 , Santarem (do.)

1 o", Rio Tapajoz, Miritituba (do.)

897. Icterus croconotus croconotus (Wagler)

Type locality: Guiana

Monte Alegre, Rio Maecuru, Rio Jamunda (Snethlage); Santarem (Chapman
and Riker)

2 o\ Lago Grande

1 d\ Rio Tapajoz, Santarem

2 9 , near Obidos

3 cf 1 9 , Obidos (Carnegie Mus.)

4 <? 4 9 , Santarem (do.)

Family TERSINIDAE

898. Tersina viridis viridis (Illiger)

Type locality: Brazil

1 cf juv., Para (Snethlage)

This record requires confirmation. The Swallow Tanager is other-
wise unknown between the Rio Madeira (occidcntalis) and the region
between Bahia and Rio de Janeiro.

Family THRAUPIDAE
899. Tanagra musica intermedia (Chubb)

Type locality: Roraima, British Guiana

1 c? juv., Monte Alegre (Snethlage)

This is the species more familiarly known as nigricoUis Sclater (nee
Vieillot) or cyanocephala Vieillot (nee P.L.S. Miiller). This blue-capped

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 321

species seems to have an interrupted distribution, the Lower Amazon
Valley being one of the areas where it is almost wholly absent. South
of our area the race aureata (Vieillot) is common from southern Brazil
southward. If the Antillean musica is not regarded as conspecific, the
specific name of the South American bird is, of course, aureata.

900. Tanagra chlorotica chlorotica (Linnaeus)

Type locality: Cayenne

Marajo Island (Hellmayr, Snethlage); Rio Iriri, Rio Tapajoz (Snethlage);

Santarem (Chapman and Riker); Rio Tapajoz, Itaituba (Hellmayr);

Monte Alegre (Snethlage)

3 c? 4 9 , Rio Tapajoz, east bank
8 cT1 6 9 , Obidos (Carnegie Mus.)
2 cf 1 9 , Santarem (do.)

4 cf 1 9 , Rio Tapajoz (do.)

This species has a wide South American range, and is now divided
into at least four races. All authors agree that the present one ranges
south to the north bank of the Amazon valley. Southward we have the
very poorly characterized violaceicollis (Cabanis), a group of variants
too near serrirostris (Lafresnaye and d'Orbigny). It is still possible
that the Amazon River may prove to be the best dividing line; in
which case all specimens from the south bank would be called either
violaceicollis or serrirostris, according to the author's taste in "split-
ting."

901. Tanagra xanthogaster subsp.

Rio Jamauchim (Snethlage).

2 cf, Santarem (Carnegie Mus.)

Compare Hellmayr's remarks on Amazonian records of this species
(Birds of Americas, pt. 1, Oct., 1936, p. 23, footnote).

902. Tanagra minuta mellea Bangs and Penard

Type locality: Iquitos, northeastern Peru

Souza (Hellmayr) ; Providencia, Rio Tocantins, Rio Tapajoz (Snethlage)

1 d", Rio Tapajoz, Caxiricatuba
4 d\ Obidos (Carnegie Mus.)

2 d" 2 9 , Benevides (do.)
1 cf, Santarem (do.)

1 9 , Rio Tapajoz, Aveiros (do.)

322 bulletin: museum of comparative zoology

This species is widely distributed in northern South America and
southern Central America, but is little known in the Amazon valley.
Typical minuta has been recorded from Manaos; and Obidos birds
might belong here too.

903. Tanagra violacea violacea (Linnaeus)

Type locality: "Surinam"

Common throughout our area (all collectors), but not a.s yet recorded from
Mexiana Island

2 o\ Para, Val-de-Caes
13 cf , 4 9 , Rio Tapajoz, both banks
2 cf 1 9 , Benevides (Carnegie Mus.)
4 cf 1 9 , Santarem (do.)

2 cf 2 9 , Rio Tapajoz (do.)

904. Tanagra melanura (Sclater)

Type locality: Barra do Rio Negro, Brazil

1 d\ near Obidos, and 1 cf , do. (Carnegie Mus.)
4 cf , Rio Tapajoz, east bank

Apparently not previously recorded east of the Rio Madeira.

905. Tanagra rufiventris rufiventris (Vieillot)

Type locality: Brazil

1 cf , Boa Vista, Rio Xingu (Snethlage)

1 cf , Rio Tapajoz, Villa Braga (Carnegie Mus.)

A species of upper Amazonia, otherwise unreported east of the Rio
Madeira.

906. Tanagra cayennensis (Gmelin)

Type locality: Cayenne

Para region, common (all collectors) ; Rio Jary (Snethlage)

1 cf 1 9 , Para, Bosque and Val-de-Caes
1 9 , Rio Acara, Acara

1 cf , Obidos (Carnegie Mus.)

3 cf 1 9 , Benevides (do.)

This well known species reaches its extreme southern limit in north-
ern Maranhao, just south of our area.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 323

907. Tanagra chrysopasta subsp.

1 9 , St. Antonio de Cachoeira, Rio Jary (Snethlage)
1 d" 1 9 , Obidos (Carnegie Mus.)
1 d\ Rio Tapajoz, Villa Braga (do.)

The two specimens from Obidos are minutely smaller than upper
Amazonian birds, but show no suggestion of the very marked color
characters of nitida Penard. It will be recalled that the type is the only
definite record of this species in the Guianas. In the Carnegie Museum
there is a male imm. from Pied Saut, French Guiana which also shows
no real approach to nitida. It is apparent that nitida is one of the highly
local coastal races in Surinam ; this furnishes additional evidence that
the "Guianas" are by no means an integral geographic unit. Typical
chrysopasta is definitely recorded as far east as the Rio Madeira.

90S. Tanagrella velia velia (Linnaeus)

Type locality: Surinam

1 9 , Obidos (Carnegie Mus.)

Not previously recorded so far south.

909. Tanagrella velia signata Hellmayr

Type locality: Para

Para (Layard, Hellmayr, Snethlage, Stone); Souza (Hellmayr); Providencia,
Peixe-Boi, Rio Maeryubim (Snethlage)

1 cT 1 9 , Benevides (Carnegie Mus.)

This genus is lacking in Lower Amazonia between Para and the Rio
Purus, where T. callophrys has been collected.

910. Tangara chilensis chilensis (Vigors)

Type locality: Bolivia

2 d\ Rio Tapajoz, Villa Braga (Carnegie Mus.)

The northeastern limit of this species, previously reported from
Calama, Rio Madeira.

324 bulletin: museum of comparative zoology

911. Tangara punctata punctata (Linnaeus)

Type locality: Surinam

Para region, numerous localities (Hellmayr, Snethlage); Faro, Rio Jamunda
(Snethlage)

3 9 , Obidos (Carnegie Mus.)

2 cf 4 9 , Benevides (do.)

1 <?, Rio Tapajoz, Villa Braga (do.)

912. Tangara varia (P.L.S. Miiller)

Type locality: Cayenne

1 9 , Villa Braga, and 1 6", Miritituba, Rio Tapajoz (Snethlage and Hellmayr)
1 9 , Villa Braga (Carnegie Mus.)

A rare species, otherwise not reported authentically from our area.

913. Tangara cayana cayana (Linnaeus)

Type locality: "Cajania"

Monte Alegre (Snethlage); Santarem (Chapman and Riker)

7 d1 3 9,2 imra., Rio Tapajoz, Santarem

4 d1 12 9 , Santarem (Carnegie Mus.)

We have elsewhere (1937) commented on the size variations of this
species and the factor of post mortem color change.

914. Tangara cayana huberi (Hellmayr)

Type locality: Marajo Island

Only known from Marajo Island (Hellmayr, Snethlage).

915. Tangara gyrola albertinae (Pelzeln)

Type locality: Salto de Girao, Rio Madeira

Igarape- Assu, S. Antonio do Prata (Hellmayr); Peixe-Boi, Rio Tocantins, Rio
Jamauchim (Snethlage)

1 d" 1 9 , Para, Benevides (ex Carnegie Mus.)

2 o71 2 9 , Benevides (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 325

916. Tangara mexicana mexicana (Linnaeus)

Type locality: Cayenne

Maraca and Monte Alegre (Snethlage)

2 cT 1 9 , near Obidos

2 cf 2 9 , (do.) (Carnegie Mus.)

917. Tangara mexicana lateralis Todd

Type locality: Rio Tapajoz, Apagy

Common throughout the south side of the Amazon in our area, from the Para
region westward (all collectors)

12 d" 4 9 , Rio Tapajoz, east bank
3 cf , Benevides (Carnegie Mus.)
2 d1 1 9 , Santarem (do.)

5 a" 2 9 , Rio Tapajoz (do.)

We are at a loss to follow Hellmayr's strictures on this subspecies,
which he synonymizes with boliviano,. The 29 specimens in the type
series of lateralis are readily told at a glance from 7 Bolivian skins.

918. Thraupis episcopus episcopus (Linnseus)

Type locality: Cayenne

Abundant throughout the area (all collectors)

1 cT 1 9 , north bank of Amazon near Obidos
1 9 , Para Bosque

13 cf 7 9 , Rio Tapajoz, east bank

1 cT, Obidos (Carnegie Mus.)

2 cf 1 9 , Benevides (do.)

3 cf 1 9 , Santarem (do.)

1 c? , Rio Tapajoz, Apagy (do.)

This fine series differs from typical episcopus in having a larger and
snowy white shoulder patch (rarely pale china blue), and a tendency
to more extensive paler edgings to the wingcoverts, this slightly
approaching coelestis (Spix) of upper Amazonia, which ranges east to
the Rio Madeira.

326 bulletin: museum of comparative zoology

919. Thraupis palmarum melanoptera (Sclater)

Type locality: eastern Peru

Common throughout the area (all collectors), except Marajo and Mexiana
Islands; Caviana Island (Brodkorb)

1 cf 3 9 , Para, Bosque and Val-de-Caes

1 cf , Rio Acara, Acara

9 cf 5 9 , Rio Tapajoz, east bank

1 9 , Obidos (Carnegie Mus.)

2 9, Benevides (do.)
7 cf 2 9, Santarem (do.)

1 cf 1 9 , Rio Tapajoz, Apacy (do.)

This entire series is obviously nearer melanoptera than typical
palmarum. We note, however, that Hellmayr refers "Para" birds to
palmarum.

920. Ramphocelus carbo carbo (Pallas)

Type locality: Surinam

Abundant throughout our area (all collectors)

4 cf ad., 3 cf imm., 1 9 , Para, Bosque and Val-de-Caes
18 cf ad., 6 cf imm., 9 9 ad., 8 9 imm., Rio Tapajoz, east bank.

1 cf 1 9 , Obidos (Carnegie Mus.)

1 cf , Benevides (do.)

1 cf , Santarem (do.)

4 cf 2 9 , Rio Tapajoz (do.)

There is a rapid post mortem change in adult males of this species.
Our fresh specimens are a brighter crimson on the throat and blacker
on the abdomen than Guiana specimens, but old specimens from the
Amazon are just like Guiana birds.

921. Ramphocelus nigrogularis subspecies

Monte Alegre, north bank, and Cussary, directly opposite on the south bank
(Snethlage)

2 cf , Obidos (Carnegie Mus.)

5 cf 2 9 , Santarem (do.)

These nine specimens differ from nine topotypes from the Rio Soli-
moes in having the black abdominal patch more restricted, the sides and
flanks consequently more extensively scarlet and often clearly separat-
ing the black thighs from the other black areas; under tailco verts far

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 327

more extensively scarlet, the black tips much narrower; female with
the abdominal patch more restricted also. In our opinion we have
here an eastern subspecies worthy of recognition. Our agreement with
the Carnegie Museum, however, does not permit us to describe their
material, and Mr. Todd writes that he does not care "to take a chance"
on this Tanager.

922. PlRANGA FLAVA SAIRA (Spix)

Type locality: Brazil = Caxias, Piauhy

Serra de Erere, Monte Alegre (Snethlage) ; Santarem (Hellmayr)

According to Hellmayr (1929, p. 283), specimens from the north
bank are nearer saira, and not macconnelli Chubb of British Guiana,
as might have been expected.

923. Habia rubica peruviana Taczanowski

Type locality: Yurimaguas, Peru

Rio Tapajoz, Boim, Villa Braga (Snethlage)

5 <?, Rio Tapajoz, Villa Braga, etc. (Carnegie Mus.)

It would appear that peruviana ranges east to the left bank of the
Tapajoz. We cannot distinguish these birds from series from the Rio
Purus and Rio Solimoes, and Dr. Hellmayr states that he cannot
separate topotypes and birds from the Rio Madeira.

924. Habia rubica hesterna Griscom and Greenway, 1937

Type locality: Pataua, right bank, Rio Tapajoz
Santarem (Chapman and Riker) ; Rio Jamauchim (Snethlage)
1 cT 5 9 , Rio Tapajoz, right bank

A paler bird below, with a pinker, less scarlet throat.

925. Lanio versicolor parvus Berlepsch

Type locality: Santa Elena, Rio Jamauchim

Rio Tocantins (Snethlage) ; Santarem (Chapman and Riker)

6 c? 2 9 , Rio Tapajoz, various places east bank
1 cf , (do.) Miritituba (Carnegie Mus.)

3 c? 1 9 , Santarem (do.)

The males measure 76-80 mm. in wing length, the majority being
80 mm. Hellmayr records versicolor from the Rio Madeira on the basis
of a male measuring 82 mm.

328 bulletin: museum of comparative zoology

926. Lanio fulvus (Boddaert)

Type locality: Cayenne

St. Antonio de Cachoeira, Rio Jary

5c?l 9, Obidos (Carnegie Mus.)

927. Tachyphonus rufus (Boddaert)

Type locality: Cayenne

Para region, common (all collectors); Rio Tocantins (Snethlage); Santarem
(Chapman and Riker)

1 cf 3 9 , Para, Val-de-Caes

1 cf 2 9 , Benevides (Carnegie Mus.)

1 cT 1 9 , Rio Tapajoz, Miritituba (do.)

928. Tachyphonus luctuosus luctuosus

Lafresnaye & D'Orbigny

Type locality: Guarayos, Bolivia

Throughout the south bank from the Rio Tapajoz to the Rio Tocantins (Chap-
man and Riker, Snethlage, Hellmayr) ; Rio Guama (Snethlage) ; through-
out the north bank in our area (Snethlage, Hellmayr)
13 o71 5 9 , Santarem (Carnegie Mus.)
1 cf 2 9 , Obidos (do.)
4 c? 5 9 , Rio Tapajoz (do.)

929. Tachyphonus phoenicius Swainson

Type locality: East Peru

1 cf1, Boim, Rio Tapajoz (Snethlage)

930. Tachyphonus cristatus cristatus (Linnaeus)

Type locality: "Cayania"

Rio Jamunda, Faro (Snethlage) and Obidos (Snethlage, Hellmayr)
4 cT 5 9 , Obidos (Carnegie Mus.)

2 d* 1

1 c? 1

11 cT 5

1 <? 2

8 <? 9

5(^2

1 c?

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 329

931. Tachyphonus cristatus brunneus (Spix)

Type locality: Rio de Janeiro

Common throughout our area on the south bank of the Amazon east of the
Tapajoz (all collectors), but unrecorded from Mexiana Island
1 9 , Pard, Val-de-Caes

Rio Acara, Acara

Santarem

Rio Tapajoz, east bank

Benevides (Carnegie Mus.)

Santarem (do.)

Rio Tapajoz, Villa Braga (do.)

(do.) Miritituba (do.)

Hellmayr in the Cat. Birds America, Pt. IX, thinks that madeirae
Hellmayr of the Rio Madeira might extend east to the left bank of the
Rio Tapajoz and west to the Rio Solimoes. While we have seen no
material from the Rio Madeira, an unexpected situation develops.
Birds from the left bank of the Rio Tapajoz are indistinguishable from
Santarem and Para series. Twelve specimens from the Rio Purus and
the Rio Solimoes are also absolutely inseparable from Lower Amazon
birds and not a single one displays any of the characters ascribed to
madeirae.

932. Tachyphonus surinamus surinamus (Linnaeus)

Type locality: Surinam

1 d% Obidos (Snethlage)

This form ranges west to Manaos, and is replaced further west by
brcvipes Lafresnaye (= napensis Lawrence).

933. Tachyphonus surinamus insignis Hellmayr

Type locality: Bemfica, Pard

Pard region, common (all collectors); Rio Tocantins (Snethlage)

2 cT 2 9 , Para, Bosque
1 cT 1 9 , Santarem

9 d1 4 9 , Rio Tapajoz, east bank

4 cf 3 9 , Benevides (Carnegie Mus.)

3 d" 2 9 , Santarem (do.)

4 cf 1 9 , Rio Tapajoz, both banks (do.)

This race ranges west to the Rio Madeira.

330 bulletin: museum of comparative zoology

934. eucometis penicillata penicillata (spix)

Type locality: Brazil

Para region, common (numerous collectors); Mexiana Island (Wallace,
Hagmann); Rio Tocantins, Cussary, Rio Jamunda (Snethlage)
1 9 , near Obidos

1 cf, Obidos (Carnegie Mus.)

2 d\ Santarem (do.)
1 9 , Benevides (do.)

Series from Cayenne and two birds from Obidos do not differ in
either color or size from series from the Rio Solimoes and Rio Purus,
the wing of males being 85-90; Santarem specimens, 89-91, 95; Bene-
vides, 1 9 95. It will be apparent, therefore, that birds near the city
of Para may prove to be a large local race.

935. Nemosia pileata pileata (Boddaert)

Type locality: Cayenne; Not Para (Sclater and Salvin), fide Hellmayr
Mexiana Island (Wallace, Hagmann, Hellmayr); Marajo Island (Hellmayr,

Snethlage); Cajutuba (Natterer); Rio Tocantins, Arumanduba, Monte

Alegre, Erere (Snethlage)

1 d* 1 9 , Rio Tapajoz, Pinhy

3 cf 3 9 , Obidos (Carnegie Mus.)

2 <? 1 9 , Santarem (do.)

936. Hemithraupis guira guira (Linnaeus)

Type locality : northeastern Brazil = Pernambuco

Caviana Island (Brodkorb); Pard and Rio Capim (Stone); Par&, Rio Moju,
Rio Tocantins (Snethlage)

3 c? 2 9 , Para, Val-de-Caes

1 d\ Rio Tapajoz, Caxiricatuba

5 cf 2 9 , Santarem (Carnegie Mus.)

2 0"! 9 , Rio Tapajoz, Miritituba (do.)

Not previously recorded west of the Tocantins. Series from the Rio
Purus and Rio Solimoes in the Carnegie Museum are also guira.

937. Hemithraupis guira nigrigula (Boddaert)

Type locality : Cayenne

Arumanduba, Rio Maecuru, Rio Jamunda, Faro (Snethlage)

3 d71, Obidos (Carnegie Mus.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 331

938. Hemithraupis flavicollis flavicollis (Vieillot)

Type locality: Cayenne

3 cf 6 9 , Obidos (Carnegie Mus.)

For so widely distributed a bird, its absence from most of the Lower
Amazon seems curious. The race centralis (Hellmayr) is recorded from
the Rio Madeira.

939. Lamprospiza melanoleuca (Vieillot)

Type locality: Guiana

Para (Natterer); Igarape-Assu, Ipitinga, Benevides (Hellmayr); Para region,
Rio Tapajoz, Rio Jamunda, Faro (Snethlage); Santarem (Chapman and
Riker)

2 cf 1 9 , Para, Bosque

2 c? 1 9 , Obidos (Carnegie Mus.)

2 d" 2 9, Benevides (do.)
6 cf 2 9 , Santarem (do.)

3 cf 2 9 , Rio Tapajoz, left bank (do.)

This genus is as yet unreported in Amazonia west of our area.

940. Cissopis leveriana leveriana (Gruelin)

Type locality: Cayenne

2 cf , Rio Tapajoz, Itaituba (Carnegie Mus.)

941. SCHISTOCHLAMYS MELANOPIS MELANOPIS (Latham)

Type locality: Cayenne

Santa Isabel (Snethlage; Santarem (Berlepsch)
1 cf , Para, Val-de-Caes

5 cf 2 9,4 imm., Rio Tapajoz, Santarem

6 cf 3 9 , Benevides (Carnegie Mus.)
13 cf , Santarem (do.)

These birds show no approach whatever to the alleged characters of
olivina Sclater. It is quite remarkable that this widely diffused species
is practically unrecorded in our area.

332 bulletin: museum of comparative zoology

Family FRINGILLIDAE

942. Cyanocompsa cyanoides rothschildii (Bartlett)

Type locality: Carimang River, British Guiana

Para region, common (all collectors); Rio Tocantins (Snethlage); Santarem
(Chapman and Riker) ; Rio Jary, Monte Alegre (Snethlage)
3 9 , Rio Acara, Acara

2 cf 1 9 , Rio Tapajoz, east bank

1 d\ Obidos (Carnegie Mus.)

3 cf 2 9 , Benevides (do.)

4 cf , Santarem (do.)

2 cf 2 9 , Rio Tapajoz (do.)

943. Oryzoborus angolensis torridus (Scopoli)

Type locality: unknown; north coast of Venezuela by Hellmayr

Para (Wallace, Snethlage, Stone); Mexiana Island (Hellmayr, Snethlage); Rio

Tocantins, Cussary, Rio Tapajoz (Snethlage); Rio Jamunda, Faro

(Snethlage)

9 cf 11 9 , Rio Tapajoz, various localities, east bank.

1 cf 19, Obidos (Carnegie Mus.)

2 cf 2 9 , Benevides (do.)
1 cf 19, Santarem (do.)

1 cf 19, Rio Tapajoz, Miritituba (do.)

944. Oryzoborus crassirostris crassirostris (Gmelin)

Type locality: Guiana

Mexiana Island (Hagmann, Snethlage); ? Cussary (Snethlage)

945. Sporophila schistacea longipennis Chubb

Type locality: Mt. Roraima, British Guiana
Peixe-Boi, Snethlage (Snethlage, as S. grisea)

This form is as yet unrecorded between the type locality and Para.

946. Sporophila leucoptera mexianae Hellmayr

Type locality: Mexiana Island

Mexiana Island (Hagmann, Hellmayr, Snethlage)

A local race of a species widely distributed south of our area.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 333

[Sporophila leucoptera cinereola (Temminck)

Type locality: Bahia

Para, by Graham (Sharpe, in Cat. Birds Brit. Mus.)

There is a probable error of labelling here, this race, the old S.
hypolcuca (Licht.), not being definitely known north of Maranhao.]

947. Sporophila plumbea whiteleyana (Sharpe)

Type locality: Mt. Roraima, British Guiana

Mexiana Island (Hagmann, Hellmayr, Snethlage); Marajo Island (Hellmayr)

948. Sporophila castaneiventris castaneiventris Cabanis

Type locality: Guiana

Obidos (Hellmayr); Arumanduba, Monte Alegre (Snethlage)

1 o71 1 9 , Obidos (Hellmayr)

949. Sporophila castaneiventris rostrata Todd

Type locality: Santarem, Brazil

Rio Tapajoz (Snethlage) ; Santarem (Chapman and Riker, Hellmayr)

2 c?, Rio Tapajoz, Pinhy and Caxiricatuba.
12 cf 3 9 , Santarem (Carnegie Mus.)

1 cf, Rio Tapajoz, Goyana Island (do.)

950. Sporophila minuta minuta (Linnaeus)

Type locality : Surinam

Nazare, (Layard); Mexiana Island (Hagmann); Marajo Island (Snethlage);

Santarem (Chapman and Riker); Quati-puru, Maraca, (Snethlage)

Rio Tocantins (.Snethlage);

1 cf , Rio Tapajoz, Tauary

2 cT 1 9 , Santarem (Carnegie Mus.)

951. Sporophila bouvreuil bouvreuil (Muller)

Type locality : Bahia, Brazil

Mexiana Island (Hagmann); Marajo Island (Snethlage)

The extreme northern limit for this well known east Brazilian species.

334 bulletin: museum of comparative zoology

952. Sporophila Americana Americana (Gmelin)

Type locality: Cayenne

Pard (Wallace, Spix); Mexiana Island (Wallace, Hellmayr); Para region, nu-
merous localities, Marajo Island, Rio Tocantins, and whole of north shore
(Snethlage); Santarem (Chapman and Riker)
5 d" 2 9 , near Obidos
4 d" 2 9 , Obidos (Carnegie Mus.)

1 9 , Benevides (do.)
3 c? 3 9, Santarem (do.)

The alleged racial variation of this species in our area seems most in-
consistent and unsatisfactory. A fine series from Cayenne proves a
great deal of individual variation in color. Entirely apart from wear,
the amount of white in the wing of males is quite variable. Females are
astonishingly variable. Older birds are apparently darker and more
richly colored, while younger ones are much paler and grayer, es-
pecially below. Size seems quite constant, the wing of males 56-58,
9 54-56. The series from Santarem and Obidos were described as
dispar Todd. All five females are pale and grayish brown below, but
not distinguishable from Cayenne birds in similar plumage. The
males have more white in the wing than the majority of Cayenne
males, but no more than those with the maximum amount of white
from Cayenne. These birds are minutely larger, wing of cf 59-61,
9 56-59. The single female from Benevides is as richly colored as any
9 from Cayenne, the wing 55. Finally a series from the Solimoes is
inseparable from Cayenne topotypes. Hellmayr has pointed out that
leuco pterygia Spix is available for Amazon birds, but the type locality
is Para. Our own view is that the absence of dark females from Obidos
and Santarem is a pure accident of small series, that the remaining
differences are inconsistent and trifling, and that all birds from Lower
Amazonia can be called americana.

953. Sporophila caerulescens caerulescens (Vieillot)

Type locality: Brazil
1 <?, Sta. Julia, Rio Iriri (Snethlage)
1 cf , Rio Tapajoz, Pinhy
Our single male is notably paler gray above, with less blackish on
the pileum, thus agreeing with Hellmayr's comments on the Rio Iriri
specimen. The Lower Amazon birds are isolated, and will probably
prove separable, when a proper revision of the species can be at-
tempted.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 335

954. Sporophila nigricollis nigricollis (Vieillot)

Type locality: Brazil

Para region, numerous records, (most collectors); Ilha das Oncas, Mexiana
Island, Monte Alegre, Rio Tocantins (Snethlage)

955. Sporophila lineola (Linnaeus)

Type locality: Surinam

Para region to Rio Tapajoz (Snethlage); Urucurituba, Santarem (Hellmayr).
10 cf 2 9 , Rio Tapajoz, various localities, east bank

1 9 , Para, Bosque

1 cf, Santarem, (Carnegie Mus.)

1 cf , Rio Tapajoz, Villa Braga (do.)

956. Sporophila bouvronides (Lesson)

Type locality: Trinidad

North side of River Amazon and Mexiana Island (Wallace, fide Sharpe in Cat-
Birds Brit. Mus., as S. amazonica Sharpe); Obidos (Snethlage)
3 d" 2 9 , Obidos (Carnegie Mus.)

Hellmayr regards S. ocellata, S. trinitatis Sharpe and S. amazonica
Sharpe as synonyms of bouvronides. We agree that there is a good
chance that this species will prove to be a mutation of lineola.

957. Volatinia jacarina jacarina (Linnaeus)

Type locality: northeastern Brazil
Common throughout our area (all collectors)

5 o" 1 9, Para, Bosque

1 9 , Rio Acara, Acara

4 c? 1 9 , Rio Tapajoz, east bank

2 d* 1 9 , Benevides (Carnegie Mus.)

1 d\ Santarem (do.)

1 cT imm., Rio Tapajoz, Itaituba (do.)

336 bulletin: museum of comparative zoology

958. Pitylus grossus grossus (Linnaeus)

Type locality: Cayenne

Para region, common; Rio Tocantins, Rio Xingii, Rio Tapajoz. Rio Jamauchim,
Rio Jary (Snethlage)

1 9 , Para, Val-de-Caes

1 c?, Rio Acara, Acara

7 c? 2 9 , Rio Tapajoz, various localities, east bank.

1 c?, Obidos (Carnegis Mus.)
1 9 , Benevides (do.)

4 c? 4 9 , Santarem (do.)

8 c? 2 9 , Rio Tapajoz, both banks

959. Periporphyrus erythromelas (Gmelin)

Type locality: Cayenne

Para region, common (all collectors)

1 c?, Rio Tapajoz, Caxiricatuba

1 c?, Benevides (Carnegie Mus.)

This Grosbeak has been overlooked on the north bank of the Amazon
in our area. Its occurrence on the Rio Tapajoz is a slight extension in
its extreme southern range.

960. Caryothraustes canadensis canadensis (Linnaeus)

Type locality: Cayenne

Para region, common (all collectors); Rio Tocantins (Snethlage)

5 c? 1 9 , Para, Val-de-Caes and Bosque
4 cf, Benevides (Carnegie Mus.)

The distributional features of these two Grosbeaks in our area are
practically identical.

961. Saltator maximus maximus (P.L.S. Muller)

Type locality: Cayenne

Common throughout our area (all collectors), but unrecorded from Mexiana
and Marajo Islands

6J3 9 2?, Para, Val-de-Caes

1 ?, Rio Tocantins, Cameta

16 c? 11 9 , Rio Tapajoz, various localities

2 9, Obidos (Carnegie Mus.)

2 c? 3 9, Benevides (do.)
4 c? 2 9, Santarem (do.)

2 c? 2 9, Rio Tapajoz, both banks (do.)

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 337

962. Saltator caerulescens mutus Sclater

Type locality: Mexiana Island

Para region (Snethlage and Stone); Mexiana Island (Wallace, Hagmann, Hell-
mayr); Marajo Island (Snethlage); Rio Tocantins, Rio Jamauchim and
north shore localities (Snethlage)

4 cf 1 9 , south bank of Amazon, Lago Grande
3 c? 1 9, Obidos (Carnegie Mus.)
3 cf 1 9 , Santarem (do.)

1 d" 1 9 , Rio Tapajoz, Itaituba (do.)

The specimens from Santarem and the Rio Tapajoz may or may not
properly represent mutus of Mexiana Island. This race is replaced by
azarce from the Rio Madeira westward, by typical coerulescens from
Matto Grosso southward, and supcrciliaris (Spix) in eastern Brazil.

963. Sicalis columbiana goeldii Berlepsch

Type locality: Paricatuba, Santarem

Santarem (Chapman and Riker, Hellmayr); Rio Tapajoz, Maraca, Monte
Alegre, Erere, Rio Jamunda (Snethlage); Obidos (Hellmayr)

2 cf, near Obidos

3 cf , 7 9 , Rio Tapajoz, various localities on both banks.
2 cf , Obidos (Carnegie Mus.)

9 cf 4 9 , Santarem (do.)

This subspecies is quite isolated from true columbiana, and has a
very scattered distribution of its own. It is characteristic of grassy
areas along river banks.

964. Sicalis luteiventris chapmani Ridgway

Type locality: Diamantina, Santarem

Santarem (Chapman and Riker); Rio Tapajoz, Boim and Pinhel (Snethlage)

2 cf , Lago Grande, west of Rio Tapajoz

8 cf 8 9 , Santarem (Carnegie Mus.)

As now restricted, chapmani is exceedingly local.

338 bulletin: museum of comparative zoology

965. Sicalis luteiventris flavissima Todd

Type locality: Rocana, Para, Brazil

Mexiana Island (Wallace, Hagmann, Snethlage) ; Marajo Island (Hellmayr,
Snethlage); Monte Alegre (Snethlage, as arvensis chapmani.)

Type series in Carnegie Museum examined.

[Sicalis flaveola Linnaeus
"Para" by R. Graham in Brit. Mus. (Sharpe)

If the specific identification is correct, the locality is probably
erroneous.]

966. Brachyspiza capensis subsp.

Rio Acara, Monte Alegre (Snethlage)

These birds might be typical capensis, or an unnamed form.

967. Myospiza humeralis humeralis (Bosc)

Type locality: Cayenne

Mexiana Island (Wallace, Snethlage); Marajo Island (Hellmayr, Snethlage);
Caviana Island (Brodkorb); Monte Alegre, Rio Jamunda (Snethlage)

3 d\ Rio Tapajoz, Santarem

4 d1 1 9 , Santarem (Carnegie Mus.)

968. Myospiza aurifrons aurifrons (Spix)

Type locality: Fonteboa, Rio Solimoes

Common throughout the area (all collectors), but unrecorded from Marajo
and Mexiana Islands

3 d" 2 9 , Rio Tapajoz, Santarem

5 cf 5 9 , Benevides (Carnegie Mus.)
2 d\ Obidos (do.)

1 c? , 1 9, Santarem (do.)

8 d1 3 9, Rio Tapajoz, various localities (do.)

969. Emberizoides herbicola subsp.

Mexiana Island (Wallace, Hellmayr)

The only adult is stated to be intermediate between herbicola and
sphenurvs (Hellmayr).

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 339

970. Coryphospingus cucuLLATus cucullatus (P.L.S. Miiller)

Type locality: Cayenne

Para region, common (all collectors)

8 d" 1 9 , Benevides (Carnegie Mus.)

971. Paroaria gularis gularis (Linnaeus)

Type locality: Guiana

Common throughout our area, but unreported in the Par;! region on the south
side of the Amazon

27 cf 18 9,5?, Rio Tapajoz, various localities.
3d" 19,1?, (do.), (do.) (Carnegie Mus.)
Id" 19, Obidos (do.)
6 d" 19, Santarem (do.)

972. Arremon taciturnus taciturnus (Hermann)

Type locality: Cayenne

Common throughout our area on the south bank of the Amazon (all collectors) ;
Obidos (Snethlage); unreported from Mexiana and Marajo Islands

2 d", Para, Val-de-Caes

1 d" 3 9 , Rio Acara, Acara, and Buenos Aires

19 d" 7 9 , Rio Tapajoz, various localities

1 d", Obidos (Carnegie Mus.)

3 d", Benevides (do.)

2 9 , Santarem (do.)

1 d", Rio Tapajoz, Aveiros (do.)

2 d", (do.), Villa Braga

The two males from Villa Braga differ in having a narrower pectoral
collar which does not meet across the chest.

340 bulletin: museum of comparative zoology

BIBLIOGRAPHY

It is assumed in this bibliography that Snethlage's Aves Amazonicas
and Hellmayr's volumes on the Birds of the Americas are the founda-
tion stones of our knowledge on the systematics and local distribution
and occurrence of birds in lower Amazonia. In the case of the former
work, a bibliography of all earlier faunal papers is given. In the latter
not only are there ample critical notes, but complete references to all
papers with locality records. With few exceptions all papers are
omitted from this bibliography, which are cited in those works. Thus
Todd's descriptions of new Formicariidae are omitted, as they are
cited in extenso in Hellmayr's volume on this family, but Zimmer's
papers are given, since they appeared after it.

No attempt has been made to cite every paper mentioning or dis-
cussing Amazonian birds in every sort of connection. To do so would
entail a bibliography of gigantic length. Such were formerly thought to
add to the learning of the authors, but in these hard times are a waste
of severely restricted publication funds.

Beebe, C. W.

1916. Zoologica, no. 2, pp. 55-106.

General notes on ecology, habits, moult, etc. of certain birds in the
vicinity of Para.

Bangs, O. and Penard, T. E.

1921. Bull. Mus. Comp. Zool., 64, no. 4, pp. 362-398.

A paper of miscellaneous systematic notes, many of which affect
species of our area.

Brodkorb, P.

1937. Occas. Papers Mus. Zool. Univ. Michigan, no. 349, March 18,
1937, pp. 1-7.

First record of a collection made by Prof. J. B. Steere in 1871
and 1879 on Caviana and Marajo Islands. The list for the
former island numbers 49 species, and is the only one as yet
published. Quite a number of earlier records are overlooked.

Butler, A. L.

1926. Bull. Brit. Orn. Club, 46, p. 56.

Description of Tapaza pella microshyncha from Utinga, Para.

Chapman, F. M.

1921. Amer. Mus. Novit., no. 2, p. 1.

Description of Capito brunneipectus.

GRISCOM AND GREENWAY: BIRDS OF LOWER AMAZONIA 341

Cherrie, George K.

1916. Bull. Amer. Mus. Nat. Hist., 35, pp. 395, 396.

Descriptions of two new subspecies from our area.

Chubb, Charles

1917. Bull. Brit. Orn. Club, 38, p. 32.

Description of Columba plumbea wallacei (Rio Capim).
1919. loc. cit., 39, p. 42.

The genus Poliolaema proposed for certain species of Myrmo-
therula in our area.

Conover, H. B.

1934. Proc. Biol. Soc. Wash., 47, pp. 119-120.

Description of Psophia viridis dextralis.
1937. loc. cit., 50, pp. 191-192.

Description of Tinamus major olivascens (Rio Acara).

Cory, C. B.

1918-19. Field Mus. Nat. Hist., Zool. Series, 13, pts. 1 and 2.

Catalogue of the birds of the Americas from the Parrots through
the Woodpeckers. A useful reference work.

Ctjnha Vieira, C. O. da

1935. Rev. do Mus. Paulista, 19, pp. 327-398.

More or less popular resume of the Cotingidae of Brazil. Lists
specimens from our area in the museum, including several
records of interest.

Griscom, L. and Greenway, J. C, Jr.

1937. Bull. Mus. Comp. Zool., 81, no. 2, pp. 417-437.

Descriptions of new subspecies and critical notes on lower
Amazon birds.

Hartert. Ernst and Goodson, A.

1917. Novit. Zool., 24, pp. 410-419

Contains systematic notes on several species in our area.

Hellmayr, C. E.

1924-1938. Field Mus. Nat. Hist., Zool. Series, 13, pts. 3-11, the con-
tinuation of the Catalogue of Birds of the Americas, begun
by Cory, including the entire order Passeres.
Invaluable for its numerous critical notes, and locality
references, including practically all faunal papers dealing
with our area.

Hellmayr, C. E. and Gyldenstolpe, H.
1937. Arkiv for Zool., 29, pp. 1-3.

Description of several new forms from Amazonia.

342 bulletin: museum of comparative zoology

MlRANDA-RlBEIRO, AlIPIO DE.

1920. Rev. Mus. Paulista, 12, no. 2, pp. 1-82.

A review of the Parrots of Brazil, with descriptions of new

genera.
1927. Bol. do Mus. Nac. do Rio Janeiro, 3, no. 2, June, pp. 1-11.

Notes on a few birds from Santarem, collected by Hagmann.

Sakesphorus hagmanni described as new.

Naumburg, Elsie M. B.

1933. Amer. Mus. Novit. no. 648, July 21.

A review of Zenaida auriculata, including the characters and
range of jessiae.

1937. Bull. Amer. Mus. Nat. Hist., 73, art. 3, Dec. 31, pp. 139-205.

Notes on Conopophagidae, Rhinocryptidae and Forimcariidae,
with occasional reference to lower Amazonian forms.

Neumann, Oscar

1927. Ornith. Monatsberichte, 35, p. 89.

Description of Pyrrhura perlata ancrythra from the Rio Tocan-
tins.
1927. Verhand. Ornith. Gesell, Bayern, 17, pp. 428-431.
Review of Pyrrhura perlata Spix.

1931. Mitt. Zool. Mus. Berlin, 17, p. 442.

Description of Brotogerys st. thomae takatsukasae from lower
Amazonia.
1933. Bull. Brit. Ornith. Club, 63, pp. 93-95.
Review of Penelope super ciliaris.

Olalla, A. M.

1935. Rev. Mus. Paulista, 19, pp. 419-423.

Interesting account of Berlepschia rikeri.

Oliveira Pinto, O. M. de.

1938. Rev. Mus. Paulista, 22, pp. 1-566.

Part I of a catalogue of the birds of Brazil, all the orders and
families through the superfamily Furnariides of the Passeres.
A very useful compilation, giving the original description, type
locality, general distribution and Brazilian range of every form.

Pinto-Peixoto, Pedro.

1923. Archiv. Museu Nac. do Rio de Janeiro, 24, 1923, pp. 267-273.

Brief list of species collected on Mara jo Island during a ten day
visit.

Sassi, M.

1932. Ornith. Monats., 40, pp. 120-121.

Description of Notharchus macrorhynchus paraensis.

GRISCOM AND GREEN WAY: BIRDS OF LOWER AMAZONIA 343

Snethlage, Emilia

1910. Bol. Mus. Goeldi, 6, pp. 226-235.

A most valuable discussion and summary of the isolated
savannahs in Lower Amazonia, which contain birds peculiar to
the "campo" fauna.

1914. Bol. Mus. Goeldi, 8, for 1911-12, pp. 1-534.

This is the well known Aves Amazonicas, the foundation work
for our area. It summarizes all preceding papers, which are
consequently not listed again in this bibliography.

1913. Jour. f. Ornith., 61, no. 3, July, pp. 469-539.

A most important paper on the distribution of birds in Lower
Amazonia.

1914. Ornith. Monatsber. 22, pp. 39-44.

Descriptions of new birds from Lower Amazonia.
1924. Journ. f. Ornith., 72, pp. 446-450.

Descriptions of new birds from Lower Amazonia.
1925a. Bol. Mus. Nac. Rio de Janeiro, 1, no. 6, pp. 407-412.

A reprint in Portuguese of the preceding article.
1925b. Journ. f. Ornith., 73, pp. 264-274.

Descriptions of new birds from Lower Amazonia.
1926. Bol. do Mus. Nac. Rio de Janeiro, 2, no. 6, Nov. 15, pp. 35-70.

A report on study in Europe in 1924-25, of collections made in

Brazil at various times 1914-23. The introduction contains

general remarks of interest on ecological and avian boundaries

in Ceara, Maranhao, and Para. Lists of the collections follow.

List B deals with 143 species from Para, collected 1914-17,

including some records of interest from our area. Unfortunately

specific localities are usually omitted.
1930a. ibid, 6, no. 1, p. 10.

Records a specimen of Liosceles thoracicus from Villa Braga,

Rio Tapajoz, June 19, 1917.
1930b. Journ. f. Ornith., 78, pp. 58-65.

Further discussion of the distribution of birds in Brazil.
1936. Bol Mus. Nac. Rio de Janeiro, 12, no. 2., pp. 83-92.

List of Brazilian Woodpeckers in the museum. Includes some

records from Lower Amazonia.

Stone, Witmer

1928. Proc. Acad. Nat. Sci. Phila., 80, pp. 149-176.

A list of birds collected near the city of Para by Bond and de
Schauensee.

Todd, W. E. Clyde

1937a. Proc. Biol. Soc. Wash., 50, Oct. 28, pp. 175-178.
Review of Crypturellus variegatus and its allies.

344 bulletin: museum of comparative zoology

1937b. ibid, pp. 183-184.

Description of Chaetura spinicauda acthalea (Para).
1937c. ibid, pp. 185-190.

The Pigeons of the Columba plumbea group.
1937d. Annals Carnegie Mus., 25, art. XIX, Nov. 16, pp. 243-255.

Descriptions of numerous new birds from our area.

Zimmer, John T.

1925. Proc, Biol. Soc. Wash., 38, p. 87.

Pipra iris Schinz the correct name for P. opalizans Pelzeln.
1929. Field Mus. Nat. Hist., Zool. Ser., 17, pp. 3-18.

A review of Deconychura.
1931-1938. Amer Mus. Novitates, nos. 500, 509, 523, 524, 538, 545,
558, 584, 646, 647, 668, 703, 728, 753, 756, 757, 785, 819, 860,
861, 862, 889, 893, 894, 917, 930, 962, 963, 994.

These papers constitute Studies of Peruvian Birds, Nos. I-XXIX.
They are really critical reviews of the genera of the families
Formicariidae, Furnariidae, Dendrocolaptidae, Pteroptochidae,
Conopophagidae, Pipridae and Cotingidae, which occur in Peru.
Whenever pertinent the great Olalla collections of these birds
from Lower Amazonia are listed. Numerous taxonomic notes and
descriptions of new forms from our area. The last four numbers
begin the Tyrannidae. The first twenty-five numbers constitute a
"volume", and a title page, contents, and index were issued
July 20, 1937.
1938. Proc. Biol. Soc. Wash., 51, March 18, pp. 47-52.

Critical notes on the members of the Crypturellus noctivagus
group, occurring in our area.

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE

Vol. LXXXVIII, No. 4

THE RECENT MOLLUSKS OF THE FAMILY NERITIDAE
OF THE WESTERN ATLANTIC

By Henry D. Russell

With Seven Plates

CAMBRIDGE, MASS., U. S. A.

PRINTED FOR THE MUSEUM

August, 1941

No. 4. — The Recent Mollusks of the Family Neritidae
of the Western Atla7iticl

By Henry D. Russell

TABLE OF CONTENTS

Page

Introduction 347

Acknowledgments 348

The Family Neritidae 349

Geographical Distribution 350

The Radula 350

Fossil Neritidae 355

Importance of Color Pattern 355

General Ecology 355

Ranges 356

Explanation of Maps 357

Systematics and Taxonomy 357

Measurements 357

Abbreviations 357

Classification of the western Atlantic Neritidae 357

Key to the genera 358

Genus Nerita 359

Genus Fluvinerita 367

Genus Puperita 369

Genus Neritina 373

Genus Smaragdia 396

Genus Neritilia 398

Spurious western Atlantic Neritidae 401

Bibliography 404

INTRODUCTION

This report considers the several species in the family Neritidae that
are known to occur in the Western Atlantic, a region which extends
from Bermuda and North Carolina throughout the West Indian
islands to southern Brazil. The majority of the species of the Neritidae
found in this area live in salt or brackish water; a few, however, in-

1 Published with the aid of a special gift from Mr. George R. Agassiz.

348 bulletin: museum of comparative zoology

habit the freshwater rivers of Central America and certain of the West
Indian islands, especially those islands composing the Greater Antilles.

From a taxonomic standpoint only two of the species and sub-
species are distinct enough in their characters to have remained more
or less clearly understood from the time of their discovery to the
present. These are Nerita ■pcloronta and N. versicolor. The remaining
forms have been very imperfectly known, and considerable confusion
has resulted as to just what names should be applied to the various
species and what limits assigned to the several varieties and sub-
species.

Perhaps one of the most important factors in connection with this
family is that, as individuals, they are abundant and form a very
conspicuous part of the fauna throughout the region that they occupy.
There are but few habitat stations in brackish water or on rocky coasts
within the American tropics where some member or members of this
family do not exist, usually in great profusion. As a consequence,
presence or absence of the commoner species of this family, from any
area, gives us an index of the amount of collecting that has been ac-
complished. Certainly this is true as regards the collections in the four
largest museums in the United States. Distributional maps for this
family will present a fairly accurate picture of the amount of collecting
that has been done on the islands and mainland for all the marine
mollusks of the West Indian region.

The purpose of this report is to distinguish the several species and
subspecies from one another, and also to give as much of the natural
history of these animals as has been possible to gather from the litera-
ture and personal field experience.

The economic importance of the Neritidae is slight. However,
Martin and Uhler, p. 97 have found that Neritina reclivata Say is of
great significance for game ducks. Probably other species are used as
food by wading birds such as flamingos.

A cknoivledg ments

The author is exceedingly grateful for helpful suggestions and
criticism to many individuals. Chief among these is Mr. William J.
Clench, Curator of Mollusks at the Museum of Comparative Zoology
at Harvard, for the privileges granted in the use of the molluscan
collections of that institution and for his kindly criticism of the ma-
terial and its arrangement for the present paper. Also I am grateful to
Dr. Arthur W\ Weysse of the Department of Biology, Boston Uni-

RUSSELL : MOLLUSKS OF THE FAMILY NERITIDAE 349

versity, for many helpful suggestions. I am also indebted to Doctor
Paul Bartsch of the National Museum, Washington, for the use of the
Neritid collection stored there, and to his assistant, Doctor Harold
Rehder. Gratitude is also due Doctor Henry A. Pilsbry of the Academy
of Natural Sciences, Philadelphia, for the use and privileges of the
Neritid collection of that institution, and to his assistant, Mr. Richard
A. McLean. Thanks are expressed to Mr. Calvin Goodrich of the Mu-
seum of Zoology, Ann Arbor, Michigan, for sending the West Indian
Neritid collection of that Museum to the author for examination. I am
very grateful to Doctor Ethan A. Andrews of Johns Hopkins Uni-
versity, Maryland, who has greatly aided the present work by his
valuable shipments of material from Jamaica.

During the years 1934-1938 five expeditions were made to the West
Indian Region under the auspices of the Museum of Comparative
Zoology at Harvard for the purpose of carrying on biological surveys
and making collections of the fauna of certain islands in this area.
The author took part in four of these trips and the present paper is
based largely upon material collected at these times. In 1934 investiga-
tions were carried on at Soledad and Gavilan, Cienfuegos, and Viniales
Cuba, under the direction of Mr. W. J. Clench. I was associated with
Mr. Clench, in 1935, on an expedition to Cat Island (Clench 1938, A),
Bahamas. In the spring of 1936 Mr. Clench visited the northern
Bahama Islands, collecting on Grand Bahama, the Abacos, and
Eleuthera. During the summer of the same year, the author headed an
expedition to Long Island in this archipelago. The island of Mari-
guana was visited by Emanuel Williams, our local field collector, in
December of the same year. I was again associated with Mr. Clench
on an expedition to Hispaniola during the summer of 1937. Monte
Cristi, Puerto Plata, and Santa Barbara de Samana were intensively
studied at this time.

A detailed list of expeditions and collectors in the Bahamas has been
published (Clench, 1938, B).

Other collections of Neritidae were made by the author in Bermuda
(1930) and southern Florida (1934 and 1936).

The Family NERITIDAE

In form, members of this family range from the rather thin, smooth-
shelled, crepiduloid type of Navicella to the globose, thick-shelled,
rough types of Nerita; also from the smooth, subpatelliform types of
Neritilia and Smaragdia to some of the globose, spiny forms of Theo-

350 bulletin: museum of comparative zoology

doxus. In size they range from a few millimeters to over 5 centimeters
in length and width. Some members possess a thick or thin peri-
ostracum; others have none. Some possess parietal teeth; others have
these much reduced or lacking. In all known forms the operculum is
calcareous. The sculpture varies throughout the group from smooth
to very rough, coarse ridges. The members of this family are all
dextral with a lunate aperture and a more or less prominent spire
according to the species examined. The food is believed to be of a
vegetable nature according to Tryon, (p. 3), probably consisting of
algae and even detritus adhering to the substratum.

Geographical D istrib ution

The Ncritidae as a family is nearly world-wide in its distribution.
It exists, however, mainly in a tropical to temperate belt girdling the
earth. For example, we find members of this group from about
northern Florida south into Argentina in the western Atlantic and from
Great Britain, Northern Europe, and the Mediterranean Sea to Cape
Town, South Africa, in the eastern Atlantic. The belt extends from
Cape Town to the Red Sea, to India and throughout the Indian
Ocean. In the western Pacific we find species existing from Southern
Australia and New Zealand through the Polynesian Islands north to
China and Japan. In the eastern Pacific, the belt extends from
approximately Lower California south to Peru. It is highly probable
that more intensive work on the Atlantic and Pacific coasts of South
America would materially extend the range of the family southward
in these areas. However, we believe that temperature is probably the
limiting factor in the distribution of the majority of species.

Like many marine and brackish families of mollusks, the family
Neritidae is far richer in species in the tropical Indo-Pacific than in
the West Indies. However, most of the genera and subgenera occur
in the West Indies.

The island of Jamaica, B. W. I., possesses a number of the most
interesting and intermediate forms among the Nerites found in the
Wrest Indian Region. It is the only locality in this area, as far as is
known, that possesses a species peculiar to it, Fluvinerita tenebncosa
C. B. Adams.

The Radula

H. B. Baker (p. 128), states: "Although more variable than that of
the Helicinidae, the radula of the Neritidae is still surprisingly stable.
The most variable character appears to be the number of cusps on the

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 351

E-lateral, although even these differences appear to indicate a certain
amount of racial differentiation or can be correlated with age (com-
pare Neritina virginea)."

The author has found this to be true to a large extent among the
Neritas and Neritinas of the West Indian region. They vary from
Fluvincrita tenebricosa C. B. Adams, with slightly roughened E-laterals
and undentated marginals to Neritina cJenchi Russell from fresh water,
with about thirty fine but distinct denticles on the E-lateral and
highly denticulate marginals.

In general, the terminology of H. B. Baker's 1923 work is used here
for the radula teeth. It is as follows:

"R-central rhachidian central

unpaired central (Fischer 1885; middle plate Troschel, 1866-
1882).

central A-plate A-central; 1st paired central, 2nd cen-

tral (Fischer); 1st "between"-plate (Troschel); "wing"-plate (von
Martens, 1879); 1st lateral.

A-B-C plates A-, B-, C-centrals; paired centrals

(Fischer); "between"-plates (Troschel), laterals.

D-plate D-lateral; comb-lateral or T-lateral

(Helicinidae).

E-plate E-lateral; accessory plate (Helicinidae).

D-, E-, plates Capituliform tooth (Fischer), or com-

plex; lateral (Fischer), or laterals; inner lateral "edge", or "mush-
room"-plate (Troschel); "umbrella"-plate (von Martens).

Uncini Marginals; lateral lamellar or lamellae

of the outer lateral (Troschel); border teeth (von Martens)."

Apparently, there is some confusion in Baker's system of naming
the various teeth of the radula, since in his key of definitions on p. 120
the D-lateral is described as the "comb lateral or T-lateral." This can
only be the large lateral tooth with an anterior portion that bears
denticles, if there are any, and is usually the largest, or principal tooth
of the radula. It is usually quite distinct and unmistakable. The
E-lateral is defined on the same page as an "accessory plate." This is
the Y-shaped tooth pressed against the base of the D-lateral so closely
that it is like a part of that tooth. Later, p. 122, the D-lateral is
referred to (pi. 9, fig. 2D), which is a Y-shaped tooth, and the text
describes this tooth and refers to the lettering as given in that figure.

352 bulletin: museum of comparative zoology

P. 123 refers to pi. 9, fig. 2E, which is the large eusp-bearing tooth or
T-lateral of his key. The text describes this tooth and refers to the
lettering on the plate so there can be no mistake as to what is meant
by this description. Obviously, there has been a confusion of designa-
tion either in the key or the text. I have chosen to call the Y-shaped
tooth "D-lateral" and the large T- or cusp-bearing lateral, the "E-
lateral" in accordance with the figures on plate 9 of Baker's work.

The Neritoid radula is a primitive one composed of hundreds of
teeth, and these teeth vary extremely among the genera and species.
A general description of it will suffice here. The radula of each species
will be discussed under that species and the degree to which it differs
from the general one. The lingual ribbon is composed of a central
(R-central) tooth which varies from subtriangular to square in out-
line. The A-central which is the next tooth laterally from the center is
claviform, bearing a more or less prominent outer posterior shoulder.
The B-central is small, suboval, and bears an irregular ridge on its
exposed surface. Next to this is the C-central which is somewhat
larger, with a larger irregular ridge that articulates with that of the
B-central. It is about the same shape as the B-central. These two
teeth are followed laterally by the D and E-laterals which, as Baker
says, are often in such close contact with one another that they cannot
be separated without breaking one or the other in that process. The
E-lateral resembles a mushroom which has been sectioned vertically
into two sagittal halves and a hollow made on the sectioned surface
of the "pileus." The "stipe," or pillar, of the second E-lateral articu-
lates with the first in this groove. The outer edge of the "pileus" may
be cusped by a few large denticles, smooth, slightly roughened, or
bearing many small cusps. The D-lateral is a Y-shaped tooth with a
curved lower end. It fits anteriorly across the stipe of the E-lateral.
The marginals vary very much among themselves in the same radula.
They may be broad or narrow at the base and bear at the free end a
hook with longer or shorter denticles or may be smooth at this end but
merely like a claw. The outermost marginals are usually long and
wedge-shaped with a ridge at the broad, free end which may or may
not bear denticles. The accompanying diagram will aid in following
this description.

The diagram illustrates the radula of Neritina reclivata taken from
Baker 1923, and has been selected for containing the necessary parts
for discussion. This is relative to Nerita and Neritina, and not for
Smaragdia or Neritilia. Wherever the word "typical" is used in the
descriptions of the radulae in the species sections, it is in relation to

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

353

fig. 1 and not in connection with what may be typical for that species
or other species. The diagram is merely used as a "yarcl stick" or
standard as a basis for comparison.

Fig. 1. Typical Neritinoid Radula
R — R central, A — A central, B — B lateral, C — C lateral,
D — D lateral, E — E lateral.

1 . posterior lobe and point

2. outer projection of inner arm of Y-thickening

3. base

4. major inner cusp

5. base of C-lateral

6. anterior wing directed downward for support

7. base or pillar of E-lateral underlying D-lateral

8. ridge of B-lateral

9. ridge of C-lateral

10. body of E-lateral

11. lateral end of A-lateral

12. central end of A-lateral

13. elliptical reflection

14. thickened upper rim or cusp edge of R-central

15. cusp of A-lateral

Baker, p. 140, speaking of the differences between the shells and
opercula of his Neritina virginea Linne and Theodoxus meleagris
Lamarck states, "The above differences will separate practically all

354 bulletin: museum of comparative zoology

specimens of the two species but the radula appears to be the only
infallible character yet described ..."

"Typical AT. virginea from Matanzas, Cuba, has 13-15 cusps on the
E-lateral; those from Cienfuegos, Cuba, 15-19." Therefore the num-
ber of cusps on the same tooth varies by six in the same species.
"Typical N. reclivata Say from Florida shows 22-27, but a young
specimen from Mississippi has only 14."

Age apparently makes a difference in the number of cusps according
to Baker, and this factor causes the number of cusps to fall well within
the range for N. virginea Linne. Where is the line to be drawn be-
tween the two specimens, since there is greater variance between the
mature specimens of N. reclivata and a young specimen than there is
between the young of N. reclivata and N. virginea? The number of cusps
on the E-lateral does not seem, therefore, to be an infallible character
for determination. Based upon this point alone, it would be im-
possible to differentiate between the two.

Baker further states, "The R-centrals of the specimens of AT. vir-
ginea and those of N. reclivata from Florida show an outline that is
longer than broad, while those from Venezuela with as high as 29 cusps
on the E-lateral of the older specimens shows an outline that is broader
than long. In all specimens the shape of the rim is quite constant
when viewed so that it is horizontal. The first lot of zigzag shows
14-15 cusps on the E-marginals, the second 23-25. Nevertheless, the
specimens of the same age from a single lot are quite constant. It
seems very probable that a careful study of the entire area will show
N. virginea actually does consist of a number of species, each with its
definite range, but, at present, no sharp distinctions seem possible."

Roger, pp. 78-79, finds that the denticles of Theodoxus vary in
specimens in a small area.

The author finds also that the number of denticles is variable among
specimens of the same species. The radula, therefore, is not an
"infallible" character for the determination of species and their
varieties, but merely an added factor pro or con in the establishment
of the relationship among specimens. It must be considered in its
proper proportion to the other characters of anatomy, shell, and
operculum that constitute the individuals. No one of these characters
can be set aside as a final and unerring court of appeal whose dictum
may be accepted as infallible. The reason for this is that all and each
of them is changeable and it is only when the specimen at hand
possesses more characters of one kind than another that we can say it
is this species or that, this variety or that, or that it is new to science.

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 355

Fossil Neritidae

There are over two hundred living species in the Neritidae with an
additional sixty or more fossil species. The fossil forms occur from the
early Jurassic to the present, with well developed color patterns, and
become numerous in the Miocene and Pliocene.

The following paragraphs from "Miocene Mollusks from Bowden,
Jamaica, (W. P. Woodring, pp. 423-434) gives a very concise account
of the Neritidae in tropical America.

"Nerita has not yet been recorded from the later Tertiary deposits of
tropical America, though it is represented in collections of Pliocene
Age from Costa Rica. The absence of this genus in the Miocene
deposits of this region probably is due to the scarcity among fossils of
shells that clung to rocks. Five species of Neritina and one of Smarag-
dia have been described from the Oligocene, Miocene, and Pliocene
beds of tropical America.

"For some reason fossil Neritinas retain their color markings in a
remarkable manner. Among the Miocene mollusks from tropical
America they have more distinct color markings than any other
genus."

There has, however, been very little change from these earlier forms
to the present. Of the two main groups found in the West Indian
region, Nerita is found from the Jurassic to the present and Neritina
is found from the Cretaceous to the present.

As far as we know there are no fossil West Indian forms living today.

The Importance of Color Pattern

As in many other groups of animals color per se of the individuals of
the family Neritidae is of little or no value as a specific determinant.
It is exceedingly variable, especially with Neritina virginea Linne and
has probably been one of the chief causes responsible for its extensive
synonymy. The pattern of the colors, however, is of far greater im-
portance, since certain species possess a quite constant type of color
application to the shells.

General Ecology

The general ecology of the Neritidae inhabiting the West Indian
region is extremely varied. In habitat members of this group are found
in fresh and brackish to salt and hyper-saline waters, and they can
withstand temperatures of from nearly freezing to over ninety degrees

356 bulletin: museum of comparative zoology

Fahrenheit. No known single species is capable of standing all of
these various vicissitudes, of course, but collectively the species do.
Certain species are found living on exposed rocks under completely
marine conditions. These rocks are beaten constantly by heavy swells,
and the Neritidae occurring here are able to withstand the same
amount of pounding as the chitons with which they are associated.
These Neritas usually cling to the rocks from a few feet above high
tide mark well into the intertidal zone. One inhabits the undersides
of broken boulders in salt water. Others live in the splash pools above
high tide mark. These pools range from almost entirely fresh to
hypersaline, and the temperature rises to over 90° F. during the day.
Several species prefer the quieter waters of sheltered bays and will
even tolerate slightly brackish waters. Some of these brackish water
forms are rock lovers, while others dwell in the mud of mangrove
swamps. The mangrove swamp Neritina virginea L. is often found
clinging to the aerial roots of these trees at the lower margin of the
intertidal zone. For many hours at a time individuals will remain
above the water, subject to desiccation and the heat of the sun. We
find certain species that will tolerate both brackish water and com-
pletely freshwater conditions, moving back and forth between the two.
Finally there are several that inhabit freshwater entirely. These occur
on stones or other hard objects usually in riffles or swiftly running
water. As far as is known there are no freshwater forms in North
America, nor are there any truly terrestrial or semi-arboreal forms in
the West Indian region. Outside of the Western Atlantic region
Neritina cornea Linne is found inhabiting the trees and gardens of the
natives in the Solomon Islands, Western Pacific, according to Eyer-
dam, pp. 44-46. Theodoxus fluviatilis Linne can survive in thermal
springs and also temperatures near freezing in the rivers of Europe.

Ranges

The ranges of the various species have been divided into the follow-
ing geographic areas for the sake of clarity and conciseness: United
States, Bermuda, East coast of Central America (we are not con-
cerned here with species from the west coast or those found in the
west coast drainage systems), East coast of South America (the same
as above applies here also), Bahama Ids., Greater Antilles, and Lesser
Antilles. The Greater and Lesser Antilles are divided at the Anegada
Channel between the Virgin Ids. and Anguilla, St. Martin, St. Barth-
olomew, Saba, and St. Eustacius Islands. The reason for this is that
faunistically the islands west and north of the Anegada Channel seem

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 357

to be more closely related among themselves than they do to those
islands south and east of them. Likewise, the islands south and east
of this Channel seem more closely knit faunistically with themselves
than with the islands to the northwest.

Explanation of Maps

The maps showing the distribution of certain of the Neritidae from
the West Indian region are based entirely upon specimens examined
by the author. No records from the literature have been used. Each
lot was examined, determined as to species, and the locality dotted
upon a map. Therefore, these maps represent the analysis of a single
person. Many of the localities were so close together that a single
dot was sufficient for more than one record.

SYSTEMATICA AND TAXONOMY

Measurements

Wherever possible, fifty specimens of each species, ten individuals
from each of five different lots, were selected at intervals throughout
the range of the species and measured to strike an average for the
species. Four measurements were made of each specimen, length,
maximum width, minimum width and aperture length.

Abbreviations

Throughout this report the names United States National Museum,
Academy of Natural Sciences of Philadelphia and Museum of Com-
parative Zoology, at Harvard, have been abbreviated to U.S.N.M.,
A.N.S.P. and M.C.Z. respectively.

The synonymy followed in this report is the same as that of
Baker 1923, except as particularly noted among the various species.

Classification of the Western Atlantic Neritidae
Subfa mily Neritinae

Genus — Nerita s.s. Linne 1758

Nerita peloronta Linne 1758

Nerita versicolor Gmelin 1791

Nerita tessellata Gmelin 1791

Nerita fulgurans Gmelin 1791
Genus — Fluvinerita Pilsbry 1932

Fluvinerita tenebricosa C. B. Adams 1851

358 bulletin: museum of comparative zoology

Genus — Puperita Gray 1857
Puperita pupa Linne 1767
Puperita tristis Orbigny 1842
Genus — Neritina Lamarck 1816

Subgenus — Nereina Cristofori and Jan 1832
Neritina virginea Linne 1758
Neritina clenchi Russell 1940
Neritina mcleagris Lamarck 1822
Neritina reclivata Say 1822
Neritina piratica Russell 1940
Neritina zebra Bruguiere 1792
Subgenus — Neritina s.s. Lamarck 1816
Neritina punctulata Lamarck 1816

Subfamily Smaragdinae
Genus — Smaragdia Issel 1899

Smaragdia viridis weyssei Russell 1940
Subfamily Neritilinae
Genus — Neritilia von Martens 1879
Neritilia succinea Recluz 1841

Key to the genera of West Indian Neritidae

1. Shell solid, thick; columellar area ridged or papillose, palatal teeth

present Nerita

Shell thin; columellar area smooth, palatal teeth absent 2

2. Shell globose 4

Shell subpatalliform 3

3. Shell greater than 10 x 10 mm 6

Shell less than 10 x 10 mm 7

4. Operculum lacking peg and periostracal layer along the palatal

margin Fluvinerita

Operculum with peg and periostracal layer along the palatal
margin 5

5. Operculum opaque white; aperture interior lemon to orange yellow;

color pattern dull chalky white with black zigzag lines to black

with chalky white spots Puperita

Operculum opaque, black; aperture cream to bluish white; color
pattern shiny, composed of lines and spots on yellow, olive,

purple, pink or black background Neritina

Subgenus Nereina

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 359

6. Operculum opaque pinkish white to salmon pink; color pattern

rather dull, composed of suboval light yellowish brown spots on a
light brown or reddish background Subgenus Neritina s. s.

7. Operculum opaque, greenish; color pattern shiny peagreen, entire

or with scattered suboval white spots often with brownish red

posterior edges Smaragdia

Operculum slightly transparent horn-colored; color pattern the
same, dull Neritilia

Genus Nerita Linne
Key to the species of Nerita

1 . Color pattern, alternating black and white spots 2

Color pattern entirely black or with irregularly scattered patches

of white 3

Color pattern dirty yellow to dirty white with spots or zigzags of
black and red 4

2. Operculum black, convex, papillose Nerita tessellata

3. Operculum bluish grey to yellow, convex, papillose . Nerita fulgurans

4. Operculum consisting of a raised smooth, reddish-horn color part

and a lower dark brown slightly papillose part 5

Operculum brownish grey, concave papillose 6

5. Parietal teeth strong one or two in number 7

6. Parietal teeth strong four to five in number 8

7. Columellar area concave and bearing a reddish orange spot

Nerita peloronta

8. Columellar area convex, white Nerita versicolor

Nerita peloronta Linne

pi. 1, fig. 1, 2; pi. 5, fig. 1.

Nerita peloronta Linne 1758, Syst. Nat. ed. 10, p. 778.

Type locality. "O. Asiae Ad Bandam" (Linne).

Description. Shell globose and solid. Ground color dirty yellow.
Pattern consisting of zigzags of black and red. Shell deeply impreg-
nated by pigment. Whorls 3, rounded in cross section. Spire low, cast
at an angle of 120° and often somewhat corroded. Aperture lunate and
cast at an angle of 66°. Palatal lip thin, faintly scalloped, sharp,
marginate with the callosity bearing 15-23 small teeth. Superior and
inferior palatal teeth larger than others; with the interior hinge tooth
strong. Parietal area thick, undulated, concave, white to dirty yellow

360

bulletin: museum of comparative zoology

with irregular red to orange central spot, and bearing one or two
strong teeth, one usually larger than the other. Suture faintly im-
pressed. Sculpture consisting of spiral ridges crossed by many faint
axial lines of growth which are always present. The spiral ridges may
vary greatly in number and from coarse to smooth.

Operculum opaque, calcareous and with interior coral pink. The
surface is undulated and shiny with the rib absent. Peg much reduced
and pitted. Periostracal layer present along the palatal margin of
exterior. Operculum consisting of two sections, the first raised, smooth
and orange in color, the second and lower part smooth to papillose and
colored a brownish-green.

Radula. R — central oblong in shape.

A — central very much heavier than in fig. 4 and with a
rounded prominence between the two ends of the tooth instead of at 1 .

B — central with "s" shaped ridge.

C — central typical.

D — lateral about typical but with outer projection of
inner arm placed lower on inner arm.

E — lateral with very heavy and large elliptical reflection
and bearing no cusps.

Inner marginals with large claw-like anterior portion bearing no
cusps. The middle marginals develop about 20 fine cusps and these
become approximately 23 coarser and longer denticles. The outermost
marginals bear about 20 very fine denticles.
Measurements.

length

Cuba: Punta de la Sabanilla,

Cienfuegos Bay 30.5

Florida: 24 mi. s.w. of Ta ver-
nier Key 29.0
Key West 41.7

Caribbean Sea: Swan Id. 32.0

Bahamas: Arthur's Town, Cat

Id. 28.7

/ Hispaniola: Puerto Sosua 29.3

Remarks. A series from Piscadera Bay, Curacao, is of interest in that
the ground color is lemon yellow in some specimens. The usual red
color is lacking and the yellow may be almost entirely obscured by a
dark brownish-blue pigment. Also, the marginate teeth on the palatal

maximum
width

34.1

minimum aperture
width length

20.7 26.9 mm.

32.3

20.2

26.0

45.1

26.1

36.7

34.5

20.6

28.0

30.9

17.5

25.4

31.0

19.0

26.0

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 361

callosity are lacking except for the superior two. Some of the specimens
possess only one parietal tooth.

Nerita peloronta Linne is strictly a marine species. It is found
throughout the West Indies on the rocky coasts, even in places exposed
to the full force of the sea. It is seldom found where there is any ad-
mixture of fresh water. While collecting this species one is struck with
the fact of their ability to hold on to the rocks with a great deal of
tenacity, a character which enables this species to maintain its hold on
the substratum even during severe storms. This is even more im-
portant when one realizes that it is not only the direct pounding of the
sea that must be sustained but also a tremendous amount of lateral
wave pressure caused by the uneven sea worn rocks that characterize
most of the coast line of these West Indian islands.

This species will tolerate but little brackish water. It soon disap-
pears within the mouths of harbors where the sea water has become
somewhat freshened from shore drainage.

Range, (see pi. 5, fig. 1).

The range extends from Saint Augustine, Florida, and the Bermudas
on the north through Central America and the West Indian islands as
far south as Trinidad.

Nerita versicolor Gmelin

pi. 1, fig. 3, 4; pi. 5, fig. 2.
Nerita versicolor Gmelin 1791, Syst. Nat. I, pt. 6, p. 3684.

Type locality. "Ad Insulas Antillas" (Gmelin).

Description. Shell solid and globose; color consisting of a variable
number of irregular black and red rectangular spots on a dirty white
background. The red may be lacking as in specimens from the island of
Trinidad. Shell is deeply impregnated by pigment. Whorls 3-3^,
rounded in cross section. Spire low and pointed; sometimes con-
siderably corroded, though rarely as much as in Nerita tessellata or
Nerita fulgurans. Spire cast at an angle of 109°. Aperture lunate and
cast at an angle of 583/2°. Palatal lip thin and sharp, irregularly
scallopped and spotted with red, white and black at the margin. It is
marginate, marginate callosity bearing 10-15 distinct teeth; superior
and inferior teeth stronger than others; interior palatal hinge tooth
faint. Parietal area thick, slightly convex and irregularly ridged. It
is white to yellowish in color and bears four to five strong teeth —
usually four. Suture faintly impressed. No periostracum present.
Sculpture consisting of a variable number of rounded, spiral ridges

362

bulletin: museum of comparative zoology

often dividing once as they approach the aperture. They bear many
faint axial lines of growth.

Operculum opaque, calcareous and brownish grey. Rib strong; peg
faint to lacking; if present, it is slightly pitted. There is a slight
periostracal layer along the palatal margin. Exterior finely papillose
and slightly concave.

Raclula. R-central more rectangular in outline than is typical.

A-central broad, as wide at central as at lateral end.
Posterior lobe prominent and about midway between the two ends.

B-central larger than typical, more elongate.

C-central with narrower base than typical.

D-lateral with larger anterior wing than typical and with
outer projection of inner arm of Y-thickening much reduced.

E-lateral with broad elliptical reflection which bears no
cusps. There is a large wing at the lateral end of the tooth which is
parallel to the base or pillar of it.

Marginals with single claw-like projection. These develop 9-12 long,
slender denticles arranged semicircularly as we progress laterally.
The outermost marginals bear about 25 denticles.
Measurements.

Cuba: Punta de los Colorados,

Cienfuegos Bay
Trinidad : Toco
Bermudas
Honduras : Oak Ridge,

Ruatan Id.
Florida: Plantation Key

Remarks. There has been a
Neriia versicolor Gmelin in the
defined by its shape and arrangement of the columellar teeth which
separate it from any other known member of the family Neritidae in
the West Indies. The color pattern varies somewhat but, as a rule, is
very stable.

N. versicolor, like Neriia pcloronta Linne lives on the most exposed
coastal rocks in salt water. The author has gathered specimens of
both these species from the most exposed rocks of the outermost
islands of the Bahamas where the full force of the Atlantic swells beat
unchecked on the rocky headlands. Both these species exist side by
side, yet never seem to interbreed. They are not found on sand or in
fresh water, but will occasionally exist in slightly brackish water.

maximum

minimum aperture

length

width

width length

21.6

21.1

15.3 18.3 mm.

25.6

28.3

17.6 22.7

26.9

28.1

17.2 23.4

21.7

23.8

14.5 19.4

20.

21.1

13.4 17.9

certain amount

of confusion with

past,

but today

the species is well

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 363

Range, (see pi. 5, fig. 2).

The range extends from Saint Augustine, Florida, and the Bermudas
through the West Indian islands and Central and South America as
far as Para, Brazil.

Nerita fulgurans Gmelin

pi. 1, fig. 5, 6; pi. 6, fig. 1.

Nerita fulgurans Gmelin 1791, Syst. Nat. I, pt. 6, p. 3685.
Nerita antillarum Gmelin 1791, Syst. Nat. 1, pt. 6, p. 3685.

Type locality. "Frequens ad Insulas Antillas" (Gmelin).

Description. Shell globose and solid. Color consisting of an irregular
marbling of black and white areas; often the white is indented below
the black, the white appearing to be worn away leaving the black
portion raised above it. Shell deeply impregnated with pigment.
Whorls 1x/2 to 3%, rounded in cross section. Spire low, often badly
corroded and cast at an angle of 132°. Aperture lunate and cast at an
angle of 59°. Palatal lip thin, sharp, irregularly scalloped with a thin
band, black or irregularly spotted with white at the margin. Marginate
with a callosity bearing 12 to 15 distinct teeth, superior two teeth and
inferior one tooth stronger than others; interior palatal hinge tooth
weak. Parietal area thick, concave and irregularly papillose, white to
yellowish, bearing two distinct central teeth. Suture faintly impressed.
No periostracum present. Sculpture consists of a variable number
of crenulated spiral ridges showing many faint spirewardly curved,
axial lines of growth. Irregular black portions stand above the worn
white ones as noted above. Spiral ridges tend to divide once or twice
as they approach the aperture.

Operculum opaque, calcareous, slightly convex and bluish grey to
yellowish in color. Exterior irregularly and finely papillose. Rib
usually strong with peg pitted and much reduced; periostracal layer
developed along the palatal margin.

RaduJa. R — central typical in outline and bearing a "V" shaped
base (fig. 1, R, 3).

A — central heavier than typical with less prominent
posterior lobe, but with a rounded prominence between the ends of the
tooth on that side.

B — central with more "s" shaped ridge than typical.
C — central typical.

D — lateral with reduced to lacking outer projection of
inner arm of Y-thickening.

364

bulletin: museum of comparative zoology

E — lateral with large heavy elliptical reflection bearing no
cusps.

Inner marginals with smooth claw-like anterior portion. Proceeding
laterally, however, these teeth become cusped with 8 or more long
ones. The outermost marginals also appear to possess fine dentations.

Measurements.

length

maximum
width

minimum
width

aperture
length

Florida: Tahiti Beach

20.2

22.8

14.2

19.2 mm.

Key West

25.9

28.6

17.3

24.4

Cuba: Punta La Milpa, Cien-

fuegos Bay

19.4

21.6

13.5

19.9

Santo Domingo: Sta. Barbara

de Samana

17.1

18.9

12.0

16.3

Trinidad: Toco

22.4

24.5

15.0

19.9

Brazil : Puerto Seguro

15.3

17.3

10.8

14.7

Remarks. Nerita fulgurans Gmel. can always be distinguished from
Nerita tessellata Gmel. by the bluish-grey to yellow cast of the oper-
culum which is decidedly black in the latter form. The color patterns
are quite different; that of true N. tessellata is a series of distinct black
and white alternating dots while that of Ar. fulgurans Gmel. is far
more blurred and irregular. The spiral ridges of Ar. fulgurans are
usually more numerous. The aperture is wider in relation to its
length in N. fulgurans and the columellar teeth are more prominent.
The most striking difference is the operculum which separates these
two species at a glance since that of N. fulgurans is bluish-white, while
that of N. tessellata is black. N. fulgurans, as a species, is usually some-
what larger than N. tessellata.

Nerita fulgurans Gmelin is only found where there is a moderate
amount of brackish water. It generally replaces N. tessellata within
the mouths of harbors where the water is only slightly freshened but
not, however, strictly brackish. Conditions of this sort only exist
where there are fairly sizable rivers entering rather large harbor en-
closures, where the admixture of salt and fresh water is more or less
constant at all times. These would explain the absence of this form in
the Bahamas where conditions of this sort do not exist, as well as many
other of the smaller islands found in the West Indies.

Where found, this species is exceedingly abundant and its absence
from so much of the territory throughout the West Indies is due to
its peculiar ecologic requirements.

Range, (see pi. 6, fig. 1)

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 365

The range extends from Fort Lauderdale, Florida, and the Ber-
mudas through the West Indian islands and Central and South
America to about Porto Seguro, Brazil.

Nerita tessellata Gmelin
pi. 1, fig. 7, 8. pi. 6, fig. 2.
Nerita tessellata Gmelin 1791, Syst. Nat. I, Pt. 6, p. 3685.

Type locality, "ad insulas inter Africam et Medium Americanam
intersitas" (Gmelin). These are probably the West Indian islands
since specimens similar to Gmelin's description are found there and
not in the Madeira, Cape Verde or Canary Islands.

Description. Shell solid and globose. Color consisting of alternating
irregularly shaped, raised black spots and indented, dirty-white spots.
Shell deeply impregnated with pigment. Whorls 2^, rounded in cross
section. Spire low cast at an angle of 118° and often considerably
corroded. Aperture lunate, cast at an angle of 63°. Palatal lip thin,
sharp, irregularly scallopped and marginate. Marginate callosity
bearing 13 to 18 distinct teeth. Superior and inferior palatal teeth
larger than the others. Interior palatal hinge tooth faint. Parietal
area thick, concave, papillose, bluish-white and bearing two weak
central teeth. Suture faintly impressed. No periostracum present.
Sculpture consisting of a variable number of spiral ridges broken
irregularly by alternating high and low areas ; often faint axil lines are
present and often the ridges break up forming two or three smaller
ones as they approach the aperture.

Operculum opaque, calcareous and black. Exterior finely papillose.
Rib strong and peg much reduced and pitted. Periostracal layer
present along the palatal margin. Exterior slightly convex.

Radula. R-central typical with two small projections at the rounded
end.

A-central narrow towards the center, but very heavy
laterally with a prominent posterior lobe, tooth shaped like a leg of
mutton with the small end towards the center.

B-central about typical with "s" shaped ridge.
C-central typical.

D-lateral with outer projection of inner arm of Y-thicken-
ing lacking.

E-lateral with large elliptical reflection which is smooth
on the edge towards the center and develops a variable number of
small denticles laterally. Inner marginals with single large claw-like

366

bulletin: museum of comparative zoology

maximum

minimum

aperture

length

width

length

14.9

17.0

10.6

14.6 mm

18.5

21.1

13.3

16.2

14.8

16.7

10.5

13.8

15.0

17.0

10.7

13.7

18.1

20.8

12.7

16.7

anterior portion. Laterally the teeth become cusped first with about 8
short, broad, blunt ones and later with about 12 longer finer denticles
semi-cireularly arranged. The outermost laterals bear 12 to about 15
small cusps.

Measurements

Santo Domingo: Sta. Barbara

de Samana
Cuba: Cayo Frances, Caibar-

ien, St. Clara Prov.
Florida: Biscayne Bay

Tahiti Beach
Bermuda
Bahamas : Clarence Town,

Long Id. 16.3 17.4 10.5 14.7

Remarks. Nerita tessellata Gmelin is more closely allied to N.
fulgurans Gmelin than to any other species from the West Indian
region. There has been a certain amount of confusion with these two
species among earlier authors, but as discussed under N. fulgurans
they may be readily distinguished.

Nerita tessellata Gmelin, similar to N. peloronta and versicolor, is
found only where purely marine conditions exist. It will not, however,
be found where wave action is strong unless there is suitable protec-
tion in rather deep cracks and crevices. It appears to be most abun-
dant where there is plenty of broken rock which affords the necessary
protection. It will, however, stand a little more brackish water than
both of the above two species and will extend into the territory occu-
pied by N. fulgurans.

This species has a tendency to congregate in rather large numbers,
clustering together sometimes as many as 200 under a single small
rock. It would appear to have a distinct association complex, as speci-
mens will be found grouped usually in a single mass under a rock with
no stragglers in other spots where, apparently, there is just as much
protection. Species of other mollusks, which also seem to need pro-
tection, will be found scattered as individuals well over the underside
of the same rock.

Range, (see pi. 6, fig. 2).

The range extends from Jacksonville, Florida, and the Bermudas
through the West Indian islands and Central and South America to
Para, Brazil.

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 367

Genus Fluvinerita Pilsbry 1932
Genotype Nerita (Fluvinerita) alticolor Pilsbry

Fluvinerita tenebricosa (C. B. Adams)
pl. 2, fig. 1, 2

Neritina tenebricosa C. B. Adams 1851, Contrib. to Conch. 1, p. 175; von Mar-
tens 1879, Conchy .-Cab. 2, pt. 10, p. 260.

Nerita (Fluvinerita) alticolor Pilsbry 1932, Proc. Acad. Nat. Sci. Phila. 84, pp.
12-13, fig. 1-2.

Type locality. Black River, Jamaica (C. B. Adams).
Description. Shell globose and thin. Ground color dark brown to
horn-color with very angular, dark brown, zigzag lines traversing the
whorls with the angles directed from aperture towards the spire.
Many of the specimens are of such a dark purplish-brown that these
angular lines are not always visible. It is only when they cross a light,
horn-colored area that they become apparent. Shell is thinly impreg-
nated by pigment which scrapes off easily leaving a bluish-white sur-
face exposed. Whorls S-3}4. Spire low, cast at an angle of 126°.
Spire tends to be occasionally corroded, though not extensively so.
Aperture lunate, cast at an angle of 58°. Palatal lip thin, sharp,
smooth and bluish. Palatal hinge tooth absent. Parietal area slightly
convex, smooth and bearing no teeth. Suture faintly impressed.
Periostracum present and thin. Sculpture consists of many faint axial
growth lines.

Operculum opaque, calcareous and brownish-white. Periostracal
layer absent along the palatal margin. Rib strong and peg lacking.
Exterior face of operculum bearing lines of growth radiating from the
nucleus.

Radula. R-central approximately square with anterior edge pro-
longed into a long rounded point.

A-central as in fig. 1 with prominent anterior shoulder.

B & C-centrals typical.

D-lateral typical but with outer projection of inner arm of

Y-thickening much reduced.

E-lateral with heavy large elliptical reflection bearing on
cusps.

Marginals — those nearest the center with a few blunt denticles. As
they proceed away from the center the denticles increase in size and
number to about 8. Outermost laterals seem to have no denticles.

laximum

minimum

aperture

width

length

8.0

5.2

7.0 mm

10.0

6.4

8.6

12.0

7.0

9.8

368 bulletin: museum of comparative zoology

Measurements. All localities are in Jamaica.

length

Catadupa St., near Spanish

Dam 7.5

Gt. River, Hanover Side 9.3

Gt. River 10.9

Remarks. Nerita (Fluvinerita) alticolor Pilsbry is identical with
specimens described by C. B. Adams, I.e. 1851, as Neritina tenebricosa
from the Black River, Jamaica. Pilsbry's specimens (I.e. 1932) are from
the Great River, Jamaica, 18 miles from the mouth and at an elevation
of 1000 ft. Specimens named by Adams as examples of his species
have been examined (U.S.N.M.). No differences are to be found
between Pilsbry's and Adams' species. Adams' name takes priority
in being of an earlier date. However, Pilsbry created a new subgenus,
Fluvinerita, with his species alticolor as the type. Fluvinerita is here
considered as of generic rather than subgeneric rank, because of radula
and operculum characters. The holotype for Fluvinerita alticolor is
A.N.S.P. no. 153559 and one paratype in the M.C.Z. no. 83952.

In shape Fluvinerita tenebricosa resembles Neritina virginca Linne
but is more globose than is typical for that species. The operculum,
however, resembles that of Theodoxus fluviatiUs Linne in shape and in
lacking a peg. The color of the two opercula is quite different ; that of
Fluvinerita tenebricosa being brownish blue, while that of Theodoxus
fluviatiUs is chalky white. Except for the above points of similarity,
these two species are very different. Fluvinerita tenebricosa is a globose
form with a low but distinct spire, a relatively shorter palatal lip and
narrower columella area than Theodoxus fluviatiUs, which is subpatelli-
form in shape with an extremely low spire. The color patterns also
vary markedly; that of F. tenebricosa is a dark brown to horn color
while that of T. fluviatiUs may be red, purple or brown and with white
spots. The radula of both is distinctly Neritoid, but that of F. tene-
bricosa possesses a more rectangular A-central and very fine irregular
denticles on the E-lateral, while T. fluviatiUs has a more triangular
shaped A-central with a very prominent posterior lobe, and the
E-lateral bears 17-18 distinct denticles. Furthermore, F. tenebricosa
is limited to freshwater rivers of Jamaica while T. fluviatiUs occurs in
the freshwater rivers, thermal springs and even in sea water in Europe.

Fluvinerita tenebricosa C. B. Adams is distinctly a freshwater
species, in the deep, strong current of the Great River, Jamaica. Dr.
Pilsbry, as above, records Nerita {Fluvinerita) alticolor — Fluvinerita

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 369

tencbricosa Adams, from eighteen miles above the mouth of the Great
River, Jamaica, and at an elevation of 1000 feet above sea level. It
was originally reported from the Black River, Jamaica, by C. B.
Adams 1851.

Range. Known only from the rivers of Jamaica.

Genus Puperita Gray

Key to the species of Puperita

1. Color pattern chalky white with zigzag black lines 2

Color pattern black with white spots 3

2. Aperture light to dark orange yellow 4

3. Aperture bluish yellow to bluish white 5

4. In general the larger of the two species Puperita pupa

5. In general the smaller of the two species Puperita tristis

Puperita pupa (Linne)

pl. 2, fig. 3, 4
Nerita pupa Linne 1767, Syst. Nat., ed. 12, 1253-1254.

Type locality. Mauritius (Linne).

Mauritius, the type locality, as given by Linne, is unquestionably
an error. It has not been reported from there since so far as I know.

Description. Shell thin and globose. Ground color chalky white
crossed by a few or many zigzag, bifurcating black lines. These lines
vary greatly in number and width; often they outline the white areas
making irregular spots of them. The spots vary in size, number and
shape. Shell thinly impregnated by pigment which scrapes off easily
leaving a chalky surface exposed. Whorls 2}/$, rounded in cross sec-
tion. Spire very low, cast at an angle of 113°; often considerably
corroded. Aperture lunate cast at an angle of 57°. Palatal hinge tooth
faint to lacking. Palatal lip thin and sharp, smooth and greyish
yellow to deep yellow. Parietal area thick, smooth, flat to slightly
concave, and bearing 3-5 small irregular teeth. Suture faintly im-
pressed. Periostracum present. Sculpture consisting of very faint
axial growth lines. Usually a hand lens is necessary to see them.

Operculum opaque, calcareous and yellowish. Peg smooth and has
a tendency to be reduced. Rib strong. Periostracal layer present along
the palatal margin. Exterior bearing radiating lines of growth from
the nucleus.

Radula. R-central, shaped as a rather long heraldic shield.

370

bulletin: museum of comparative zoology

A-central with heavy posterior lobe nearer the R-central
tooth than in the typical form.

B-eentral typical.

C-central typical.

D-lateral typical with much reduced outer projection of
inner arm of Y-thickening.

E-lateral with heavy, large elliptical reflection and often
bearing a variable number of many fine cusps. Inner marginals with
single large claw-like anterior portion. As we proceed outwards, 8 or
10 long denticles are developed. The outermost marginals appear to
bear about fifteen very small cusps.
Measurements.

length

maximum
width

minimum
width

aperture
length

Bahamas: Gt. Inagua

Id.

Matthew Town

9.6

10.4

6.6

8.8

Gd. Bahama Id.

Eight Mile Rock

10.1

10.6

6.8

9.1

Santo Domingo:

Puerto Sosua

9.5

10.0

6.2

8.5

Cuba: Havana

8.9

9.7

5.8

8.2

Florida : Key Vacca,

East Sister Key

10.5

9.8

7.4

9.5

Remarks. According to Thiele, p. 73, there are a few species of his
subgenus, Puperita, that occur in the Pacific Ocean. Whether or not
these are definitely members of this genus, I cannot say. They do not
concern us here. However, since Gray 1857 has established Nerita pupa
L. 1767 as the type of his subgenus, Puperita, this subgenus must either
be placed under the genus Neritina, as has been done by previous
authors, or it must be given generic rank. Because, as given above, it
does not entirely exhibit the characters of either the genus Nerita or
those of Neritina, I prefer to consider it a separate genus standing
between the two and closer in relationship to the genus Neritina.

Puperita pupa Linne is one of the most curious species of the family
Neritidae from the point of view of its habitat. It dwells in pools in the
coastal rock within the spray zone and splash from the waves. At
times during heavy rains these pools are almost entirely fresh and at
other times, as during storms, they are entirely hypersaline or at
least as salt as the ocean water. This species has been found inhabiting
splash pools in Santo Domingo that were over 90° F. and exposed to
the full brilliance of the tropical sun. A few moments later a travelling
rain cloud, or "chawasco", would pass. For perhaps twenty minutes
the heavy rains deluged the coast. Then the weather would clear.

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 371

This might happen several times in an afternoon. To what degree the
storms would lower the temperature of the splash pools, I do not know,
but it must be considerable since tropical rain feels very cold to those
drenched by it. Not only the salinity, but also the temperature of the
environment of P. pupa must, therefore, be radically changed in a
short period of time. With a calm sea, such as during the calms of
August when there is little splash from the waves and the sun beats
down on these splash pools, evaporating the water from them, the salt
content must consequently rise considerably and these pools must
become hypersaline. This is another environmental change, though
not as rapid as those above. P. pupa must be able to withstand it.
Unlike Nerita peloronta and N. versicolor, P. pupa is not found above
the water, but always immersed in it. Presumably, it feeds upon
algae in the splash pools. Where this species is found, it usually occurs
in considerable numbers.
Range.

United States: Florida — Big Pine Key.

Bahama Islands: Grand Bahama Island — Settlement Point; Eight
Mile Rock.

New Providence — Nassau

Andros — Mangrove Cay

Little San Salvador — 18 mi. N.W. of Cat Island

Cat Island — Arthur's Town

Long Island — - Cape Santa Maria; Clarence Town

Mayaguana — Northeast Point

Great Inagua — Mathew Town
Greater Antilles: Cuba — Cape San x\ntonio; Havana; Cabarien;
Cochinos Bay; Cienfuegos Bay. Santiago

Grand Cayman — George Town

Little Cayman Island

Gonaive Island

Santo Domingo — Puerto Plata ; Santa Barbara de Samana

Jamaica — Negril Point; Annotta Bay

Virgin Islands — St. Thomas ; St. Croix
Lesser Antilles : Barbados — Bridgetown
South America : Colombia — Cartegena

Puperita tristis (D'Orbigny)
pi. 2, fig. 5, 6
Neritina tristis D'Orbigny 1842, in Ramon de la Sagra Histoire Physique, Poli-
tique et Naturelle de L'lle de Cuba, 2, pp. 47^8, pi. 27, fig. 35.

372

bulletin: museum of comparative zoology

Type locality. Cuba (D'Orbigny).

Description. Shell, globose and thin. Ground color black with many
irregular, chalky white spots varying in size, number and shape. Shell
thinly impregnated with pigment which scrapes off easily leaving a
yellow to white surface exposed. Whorls 2 1/3, rounded in cross sec-
tion. Spire very low, cast at an angle of 124°, usually partially to much
corroded. Aperture lunate, cast at an angle of 73°. Palatal lip thin,
sharp, smooth, greyish or bluish yellow, never the deep yellow of
Nerita pupa L. Palatal hinge-tooth weak. Parietal area flat to slightly
concave, smooth and bearing 3-5 small irregular teeth along the
columellar margin. Suture faintly impressed. Periostracum present
but thin. Sculpture consists of very faint axial growth lines; a hand
lens is usually necessary to see them.

Operculum calcareous, opaque, bluish white to yellowish. Peg
smooth and strong; rib strong. Periostracal layer present along the
palatal margin. Exterior face of operculum bearing lines of growth
radiating from the nucleus.

Radula. R-central more rectangular than typical.

A-central heavier at central end than typical and with a
much greater rounded posterior lobe nearer the center of the tooth
than typical.

B-central with "s" shaped ridge.
C-central typical.

D-lateral with outer projection of inner arm of Y-thickening
lacking.

E-lateral with elliptical reflection broad and smooth on the
edge, slightly roughened as if by wear or bearing twenty or more small
denticles; major inner cusp often much larger than typical. Inner
laterals with single, large claw-like anterior portion. This claw de-
velopes about 10 heavy, short denticles and these become curved,
long, and slender and arranged somewhat semicircularly as we pass
laterally. Outermost laterals appear to be minutely denticulate.

Measurements.

Guadeloupe Id.
Virgin Id.: Guana Id.,

Tortola
Honduras
Cuba: Guantanamo

Havana

maximum

minimum

aperture

length

width

length

7.7

8.0

5.0

7.1 mm

7.6

8.1

5.1

6.9

9.8

9.3

5.9

8.0

8.4

8.8

5.5

7.6

8.8

9.4

5.8

8.7

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 373

Remarks. Puperita tristis D'Orbigny has previously been considered
a subspecies or variety of Puperita pupa Linne. Try on, (p. 42) considers
Puperita tristisT>'Ovbignya synonym of Neritina pupa and von Martens
(Conchy. -Cab. 2, pt. 10, 1877, p. 130) considers Neritina tristis
D'Orbigny a variety of Neritina pupa Linne.

In view of similarities of shell shape and type of color pattern, as
well as the presence of a periostracum, I believe that Puperita should
be considered a genus, standing between Nerita and Neritina. The
radula possesses characters of both genera. Puperita tristis D'Orbigny,
in most cases, possesses a color pattern distinct from Puperita pupa
Linne though there are occasional intergrades between the two forms.
P. tristis is usually smaller and has a pale yellow interior of the aper-
ture rather than the yellow orange of P. pupa. The ranges of the
two species overlap but are not coextensive and the radulae differ
somewhat in that the E-lateral of tristis tends to be far more dis-
tinctly cusped than that of pupa. For the above reason tristis is here
considered a full species.

Puperita tristis D'Orbigny has the same ecologic habits as P. pupa.
Range. Greater Antilles: Cuba — Corrientes Bay; Havana; Cien-
fuegos Bay; Santiago

Rosario Key 45 mi. W. of Isle of Pines
Haiti — Miragoane; Saltrou; Isle a la Tortue (Tortuga)
Jamaica — Montego Bay; Annotta Bay; South East Point
Virgin Islands — St. Thomas ; Tortola
Lesser Antilles: Marie Galante
Barbados — Bridgetown

Genus Neritina Lamarck

Key to the species of Neritina

1 . Shell globose 2

Shell subpatelliform 3

2. Operculum black to bluish black 4

3. Operculum pinkish white to salmon pink N eritina punctulata

4. Ground color of shell often containing red, lavender or purple . . .5
Ground color of shell olivaceous green or yellow 6

5. Color pattern consisting of dark zigzag lines and lighter spots with

the leading edge outlined in heavy black. . . .Neritina virginea

Color pattern consisting of dark zigzag lines and lighter spots with

the leading edge outlined with white, white and black or white

and red ; suggests imbricating scales Neritina meleagris

374 bulletin: museum of comparative zoology

6. Color pattern consisting of zigzag dark more or less parallel lines

and lighter spots 10

Color pattern consisting of black more or less parallel lines, very

closely set 7

Color pattern consisting of black more or less parallel lines, not

very closely set 8

Color pattern consisting of black lines forming a reticulated

triangular or diamond shaped network Neritina piratica

7. Palatal lip sharp not marginate 9

8. Palatal lip usually marginate Neritina zebra

9. Edge of parietal area opposite parietal teeth, bluish white or with

faint yellowish tinge Neritina reclivata

10. Edge of parietal area opposite parietal teeth outlined with dark
orange yellow Neritina clenehi

Subgenus Nereina Cristofori & Jan 1832
Neritina virginea (Linne)
pi. 2, fig. 7, 8; pi. 7, fig. 1
Nerita virginea Linne 1758, Syst. Nat. ed. 10, T.I. p. 778.

Type locality. Mediterranean Sea (Linne).

Description. Shell globose and thin. Color very variable consisting
of crooked axial lines varying in color from black to pink to purple and
red, and irregular olivaceous to white spots, varying in size, shape, and
number. The ground color of the shell is olivaceous, yellowish or white,
and the color is given to the pattern mainly by the crooked axial lines
which make the color deeper the closer together they occur, and fainter
as they move apart. On many specimens the color occurs solidly or in
spiral bands, sometimes solid, sometimes broken by spots. The
pattern varies from solid black with a few small, white spots to white
with a few irregular black lines. The "leading" or aperture edge of the
spots is almost invariably outlined with a heavy, solid black margin.
The "trailing" or spireward edge is not. The shell is thinly impreg-
nated by pigment which scrapes off easily, leaving a yellowish white
surface exposed. Whorls, 3, rounded in cross-section. Spire, low and
cast at an angle of 105°. Aperture, lunate and cast at an angle of 74°.
Palatal lip long, thin, and sharp, smooth, and bearing no teeth,
yellowish to bluish white. Palatal hinge tooth weak. Parietal area
smooth, convex, white to yellowish, bearing a variable number of

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

375

small, irregular teeth. Suture faintly impressed. Periostracum thin.
Sculpture consisting of many faint axial lines.

Operculum opaque, calcareous and usually black. Exterior bearing
faint lines of growth radiating from the nucleus. Rib smooth, rib and
peg strong. Periostracal layer present along the palatal margin.

Radula. R-central square to rectangular; cusp edge sometimes
develops two lateral processes that resemble blunt to rather long,
curved horns.

A-central typical but with a more curved cusp edge than
typical.

B-central typical.

C-central typical to slightly wider than typical.

D-lateral typical.

E-lateral with wide elliptical reflection bearing 8-9 large
pointed denticles, very much larger than typical. Major inner cusp
much more prominent than typical.

Inner marginals bearing 7-9 pointed denticles on the anterior edge.
Middle marginals with 9-11 longer, more curved denticles arranged
somewhat semicircular ly. Outermost laterals appear uncusped.

length

maximum
width

minimum
width

aperture
length

Santo Domingo: Rio Manjoi

Puerto Plata

18.9

19.1

12.1

16.0 mm

Florida: New River, Fort

Lauderdale

16.1

15.6

9.9

12.6

Brasil: Bahia

9.3

9.1

6.1

8.2

Guatemala: Puerto Barrios

10.0

10.6

6.9

9.0

Bahamas: New Providence Id.

Old Fort, Nassau

11.8

12.1

7.9

10.1

Remarks. Nerita listen Pfeiffer. Pfeiffer, p. 255 describes a shell that
might be one of several species from the West Indian region and refers
for a figure to Lister, Hist. Sive Synop. Method. Conchyl. et tabularum
anatomicarum pi. 604, fig. 26, 27, 1688. Lister's figures 26 and 27,
pi. 604 of this work are without question Neritina virginca Linne.
Neritina, listen of the literature has previously been ascribed to
Pfeiffer, but, as above, it is here considered to be a synonym of N.
virginea L. since Lister's description is somewhat doubtful as to the
species meant and the figure to which he refers is definitely that of
N. virginea L.

Neritina virginca (Linne) is one of the most abundant and variable
forms occurring in the West Indian region. It extends almost through-

376 bulletin: museum of comparative zoology

out the region and thus has the most extensive range of any of the
species found within these limits. Its synonymy is enormous as
Tryon, p. 40 says. This is due in part to the great variety and number
of color patterns exhibited by this species. Also, the extensiveness
and the abundance of specimens throughout the range has brought it
to the attention of many travellers and investigators in these regions.
Thus, specimens of this species have found their way into many
museum collections of natural history and many names have been
given to the various color-pattern forms.

N. virginea (Linne) finds its counterparts in regard to variability of
color pattern in Neritina ualanensis Lesson from the Indian Ocean and
Polynesia and in Theodoxus fluviatilis Linne from Europe. Neritina
ualanensis Lesson is a lower spired, more globose form, usually with a
more angulated type of color pattern and tends in the majority of
cases to be smaller than N. virginea Linne, and it also possesses an
operculum with a peg and rib of about equal length and firmly united
by a central section or buttress. These characters readily separate the
two species. They are both inhabitants of brackish water.

Theodoxus fluviatilis passes through many variable color pattern-
forms somewhat similar to N. virginea, but may easily be distinguished
from the latter by its subpatelliform shape with low spire, flat parietal
area and usually smaller size and white operculum which lacks a peg.
It is an inhabitant of fresh water, thermal and salt springs, brackish
and even sea water.

An examination of the specimens of N. virginea (Linne), A", clenchi
Russell, and N. reclivata Say brings out the following points of interest:
N. virginea is a brackish water species with richly or brightly colored
color pattern. The radulae among these species vary in the dimen-
sions of the teeth and proportionate measurements. This takes place
from individual to individual in the same lot but principally the
denticles of the E-Iateral vary and fall into certain rather elastic
groups. For example, we find that the E-lateral denticles of N. vir-
ginea from brackish water number from about five large coarse ones to
ten finer ones. The next group is composed of intermediate specimens
which, in regard to color pattern, fall between N. reclivata and N. vir-
ginea. The denticles of the E-lateral number approximately 10-14.
Following this, are specimens of N. reclivata with E-lateral denticles
numbering from 8 or 9 to from 17 to 22. The latter figures were in the
majority of those counted. The next group is N. clenchi which bears
about 21-29 E-lateral denticles. From the point of view of ecology, we
find that the specimens with the least number of E-lateral denticles,

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 377

that is, N. virginea occur in brackish water and that as the water
becomes more fresh the number of E-lateral denticles increases until
where the water is entirely fresh we find the greatest number in N.
clenchi. Correlated with this denticular change, are changes in shell
size, color pattern, and the operculum. Nothing can be speculated
concerning these facts at this time since adequate data concerning the
subject are lacking. It may be that at some future date further informa-
tion from other regions may make an explanation possible.

Neritina virginea (Linne) is a species of the mangrove swamps and is
strictly a brackish water form. It does, however, prefer the lower
margin of such swamps where there is only a short exposure to the air
at low tide. It appears to be most abundant around the mouths of
creeks and rivers though it will not advance beyond the area of
brackish water.

Range, (see pi. 7, fig. 1). The range extends from Saint Augustine,
Florida, and the Bermudas through the West Indian islands and
Central and South America as far south as Itabapinana, Brasil.

Neritina clenchi Russell
pi. 3, fig. 1

Neritina clenchi Russell 1940, Memorias de la Sociedad Cubana de Hist. Nat.,
14, no. 4, pp. 261-262, pi. 46, figs. 1, 2.

Type locality. Rio Manjon, 7 km. s. e. of Puerto Plata, Santo
Domingo.

Description. Shell globose and thin. Color black to olivaceous con-
sisting of crooked axial black lines and white to olivaceous irregular
spots that vary in shape, size and number. The "leading", or aperture,
edge is outlined in heavy black; the "trailing", spireward, edge is not.
The shell is thinly impregnated by the pigment which scrapes off easily
leaving a yellow-white area exposed. Whorls, 2j/2, rounded in cross
section. Spire low, usually much corroded, and cast at an angle of
115°. x\perture lunate, and cast at art angle of 75°. Palatal lip long
and thin, sharp and smooth and bearing no teeth, edge yellowish to
bluish-white interior. Palatal hinge tooth weak. Parietal area smooth,
convex and white with a yellow to orange edge except where the small
irregular teeth are borne on the columellar edge. Suture faintly
impressed. Periostracum thin. Sculpture consisting of many faint
axial lines.

Operculum opaque, calcareous and black to pink. Exterior bearing

378

bulletin: museum of comparative zoology

faint lines of growth radiating from the nucleus. Peg strong, rib has a
tendency to be reduced. Peg smooth. Periostracal layer present along
the palatal margin.

Radula. R-central slightly longer than wide.

A-central heavy; rather short in proportion to its width
with broadly rounded posterior lobe.
B-central typical.
C-central typical.

D-lateral typical but with reduced outer projection of inner
arm of Y-thickening.

E-lateral typical and bearing 22-29 small, distinct denticles on the
anterior edge of the elliptical reflection.

Inner marginals with long claw-like anterior portion bearing 9-10
rather short denticles. The middle marginals bear about the same
number of longer, more curved cusps arranged in an oval formation.
The outermost marginals appear typical and uncusped.
Measurements.

Santo Domingo: Rio Manjon,

Puerto Plata
Cuba: Rio Arimao, Cienfuegos 18.9
Jamaica : Great River

Montego River
Cuba: Rio San Juan,

Mantanzas Prov. 19.9 19.1 13.7 16.4

Guatemala: Cavech River,

Livingston 20.1 19.1 12.7 16.7

The specimens from the Great River, Jamaica, were very abnormal
in shape in that they were badly corroded at the spire and the animal
had tried to cover growths of calcareous algae on the whorls with shell.
The algae grew mainly on the body or last whorl at the base of the
parietal area.

The holotype for this species is M.C.Z. no. 115,701.

Remarks. Certain specimens are from Salt Pond, Montego Bay,
Jamaica, where there is a town dump and which is sometimes open to
the ocean and sometimes not. These specimens resemble Neritina
smithi Gray from Calcutta, India, in shape and color pattern, but are
smaller.

Neritina clenchi is a freshwater form occurring on rocks or among
algal growths. It seems to prefer swift waters or rapids and is found
in company with Neritilia succinea Reel, and Neritina pnnctulata

maximum

minimum

aperture

ength

width

length

18.0

11.4

14.3 mm.

18.9

19.0

12.0

16.1

19.4

19.7

12.7

15.9

16.8

17.0

10.6

14.4

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 379

Lamarck. The author has gathered specimens of this species from
algal covered rocks in the shallow rapids of the Rio Manjon, seven
kilometers southeast of Puerto Plata, Santo Domingo. They exist
under similar circumstances in the Cabaritta River, Jamaica, where
they have been found in a strong fresh to slightly brackish current.
Where there are no stones, they often occur on the stems of water
plants. They are sometimes associated with freshwater sponges, blue-
green algae, serpulid tubes and bivalves.

Specimens from the Great River Rapids, Jamaica, exhibit much
corrosion and an extensive growth of calcareous algae. The animal
was attempting to cover up the masses of these calcareous algae by
depositing shelly material over them. Neritid egg capsules containing
young specimens were found adhering to the shells which were col-
lected in July 1932 by Andrews. These may have been the egg cap-
sules of Neritina punctulata Lamarck, which was found at the same
place at the same time, but the young within the capsules were too im-
mature to be identified.

Range. United States : Florida — Tampa ; Fort Lauderdale ; Miami
Central America : Guatemala — Cavech River, Livingston
Greater Antilles: Cuba — Havana; Matanzas; Cienfuegos; Baracoa;
Sabana la Mar

Santo Domingo — Rio Manjon, 7 km. S. E. of Puerto Plata

Jamaica — Lucea ; Montego River, Montego ; Fern Gully, St. Ann
Virgin Islands — St. Croix, Fair plain stream
Lesser Antilles : Guadeloupe — Basse Terre
Martinique — Fort de Franc
Tobago — Courland River

Neritina meleagris Lamarck

pi. 3, fig. 3, 4

Neritina meleagris Lamarck 1822, Anim. Sans Vert. 6, p. 187.

Neritina pulchella W. S. Gray 1856, Index Test. Suppl. p. 232, pi. 8, fig. 18.

Theodoxus meleagris Lamarck, H. B. Baker 1923, p. 157, pi. 13, fig. 23.

Type locality, in rivers of Santo Domingo (Lamarck).

Description. Shell thin, globose, more so than Neritina virginea Linne.
Color variable, consisting of many subtriangular spots varying in size,
shape, number and arranged like imbricating scales with short, angular
lines between them. These spots vary in color in different localities from
a deep brown to olivaceous and to a bluish-grey to white. The "lead-
ing" or aperture edge of the spot is generally outlined with white, white
and black, white and red, or white and tan. The "trailing" or spireward

380

bulletin: museum of comparative zoology

margin of the spot is without the white or white combination color.
The spots may occur in more or less irregular bands and are irregularly
spaced. Between the spots are short, angular lines which are usually
arranged as follows : black, white and tan ; red, white and tan ; black and
tan; black and bluish grey, or merely white surrounding the tan or
olivaceous spots. The angular lines may occur as bands of color on the
shell. The spots usually take on a greyish cast where the bands are
grey and a deeper yellow where the bands are red. The shell is thinly
impregnated by pigment which scrapes off easily leaving a tan-bluish
grey surface exposed. Whorls 3 to 3^, rounded in cross section. Spire
low, rounded, occasionally slightly corroded, and cast at an angle of
127°. Aperture lunate, cast at an angle of 73° and slightly sinuous.
Palatal lip rather long (not as long as Neritina virginea L) sharp, thin,
bearing no teeth and colored bluish. Palatal hinge tooth weak. Parie-
tal area smooth, convex, dirty yellow. Columellar edge bearing small
irregular teeth varying in size and from about 4 to 10 in number.
Suture faintly impressed. Periostracum thin, finely pitted. Sculpture
consisting of many faint axial growth lines.

Operculum calcareous and blackish-grey with the exterior part bear-
ing many faint lines of growth radiating from the nucleus. Periostracal
layer along the palatal margin present. Peg and rib strong with the
peg smooth.

Radula. R-central width approximately equal to length.

A-central rather longer for its width than typical and with
very heavy rounded posterior lobe.

B-central typical.

C-central somewhat larger than typical.

D-lateral typical.

E-lateral typical but with somewhat larger and more
rounded major inner cusp. The elliptical reflection bears about 17-20
distinct cusps.

Inner marginals with claw-like anterior portion and bearing 9-10
cusps. Middle marginals bearing 6-8 rather longer, more curved cusps
arranged in a semicircular form. Outermost marginals appear to be
typical but bear no denticles.

Measurements.

length

maximum
width

minimum
width

aperture
length

Cuba : Punta de Los Colorados,

Cienfuegos

8.7

9.1

6.1

7.8 mm

Guatemala: Puerto Barrios

7.8

7.5

5.7

6.7

Nicaragua: Bluefields

8.0

8.1

5.3

7.2

Trinidad: Anse Trinquant

11.9

12.0

7.7

10.6

Brazil : Ceara

12.0

12.2

8.1

10.8

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 381

Remarks. Neritina meleagris Lamarck has been considered by some
as belonging with the European and western Asiatic genus, Theodoxus.
A comparison of meleagris with T. fluviatilis Linne, the type of Theo-
doxus, shows them to be quite different in shape. N. meleagris pos-
sesses a typical slate-colored Neritina operculum with a strong peg and
rib, while T. fluviatilis has lost the peg and the operculum is white and
the shape of the operculum of T. fluviatilis is more semicircular than
that of N. meleagris. The color patterns vary. N. meleagris has the
effect of imbricating scales with a white leading edge, usually arranged
on a brownish-yellow background, while that of T '. fluviatilis is made up,
largely, of white spots irregularly arranged on a varied background of
black, purple, or red. Ecologically, Theodoxus is a freshwater genus
while Neritina meleagris is a brackish water form. For the above
reasons I believe that Neritina meleagris Lamarck is not a Theodoxus.
This species has been considered a synonym of N. virginea by several
authors, Tryon, p. 40 and v. Martens 1877, p. 123, but a comparison of a
large series of both forms indicate them to be quite different. The
range of meleagris falls within that of virginea, though the former is far
more restricted.

No hybrid specimens have been seen between the two species.
Neritina meleagris Lamarck is a brackish water form found generally
in mangrove swamps. It is of interest to note that, like N. reclivata
Say, it does not occur on the Bahama Islands. Possibly these forms
cannot live here because the Bahamas are composed of eolianite which
is constantly being built up and broken down. Perhaps N. meleagris
has never reached the Bahamas. We know too little of its ecology even
to guess at the forces governing its relationship to its environment or
its distribution.

Range. Central America: British Honduras — Belize

Guatemala — Puerto Barrios

Honduras — Puerto Cortes

Nicaragua — Bluefields

Costa Rica — Matina

Panama — San Bias
South America: Colombia — Cartegena; Brazil — Fortaleza; Parna-
hyba, Nova Almeida; Nictheroy; Santos; Sao Paulo

Greater Antilles: Cuba — Cayo La Salina, Caibarien; Cienfuegos
Bay; Nuevitas

Haiti — Cap Haitien; Aquin; Port-au-Prince

Santo Domingo — Santa Barbara de Samana

Puerto Rico — Desecheo Island, Mayaguez

382 bulletin: museum of comparative zoology

Virgin Islands — St. Thomas; St. Croix
Lesser Antilles : Barbadoes — Bridgetown

Trinidad — Anse Trinquant, 9 mi. N. W. of Port of Spain

Neritina piratica Russell

pi. 3, fig. 5, 6

Neritina -piratica Russell 1940, Memorias de la Sociedad Cubana de Hist.
Nat. 14, no. 4, pp. 259-260, pi. 46, figs. 3, 4.

Type locality. Waunta Lagoon, Nicaragua.

Neritina listen Sowerby 1855, in part, Thes. Conchy. 2, p. 534,
pi. 116, fig. 249 (not fig. 250-251), not Nerita listen Pfeiffer 1840 ( = Ner-
itina virginea Linne) Weigmann's Arch. f. Naturg. p. 255; not Nerita
listen Recluz 1841 (= Nerita fulgur arts Gmelin) Rev. Zool. p. 177.

Description. Shell globose and rather thick. Color brownish-yellow
to olivaceous green with reddish brown or black zigzag axial lines form-
ing a reticulated network over the shell. These lines break up the main
color into triangular or diamond-shaped areas of lighter color which
vary in size. The pattern scrapes off easily leaving a yellowish surface
exposed. Whorls 2^-3, rounded in cross section. Spire rather low,
often somewhat corroded, and cast at an angle of 117°. Aperture lu-
nate and cast at an angle of 78°. It is slightly sinuous. Palatal lip
rather long and sharp. It tends to be slightly marginate but bears no
teeth. Palatal hinge tooth strong. Parietal area smooth, convex and
dirty-yellow. The columellar edge bears very small irregular teeth.
Suture faintly impressed and with a dark subsutural line. Periostra-
cum thin. Sculpture consisting of faint axial lines of growth.

Operculum opaque, calcareous and slate-colored. Periostracal layer
present along the palatal margin. Exterior bearing faint lines of
growth radiating from the nucleus. Rib and peg strong. Peg smooth.

Radula. R-central wider than long.

A-central with large, rounded posterior lobe and long, nar-
row shank towards R-central.
B-central typical.
C-central typical.

D-lateral typical but with very much reduced outer projec-
tion of inner arm of Y-thickening.

E-lateral typical, bearing 12 rather coarse denticles on the
anterior edge of the elliptical reflection. The major inner cusp is far
more prominent than typical.

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

383

maximum

minimum

aperture

length

width

length

20.5

21.5

13.8

18.6 mm

17.8

18.4

12.9

15.2

16.0

10.1

13.7

15.2

15.7

10.2

13.5

14.2

15.3

9.9

13.4

Inner marginals with claw-like anterior part bearing 8-9 denticles.
Middle marginals bearing about the same number of longer, more
curved ones arranged somewhat semicircularly. Outer marginals
typical and bearing no teeth.

Measurements.

Guatemala: Cavech River
Nicaragua: Waunta Lagoon
Cuba: Zapata, Santa Clara

Province
Santo Domingo: Sanchez
Venezuela: Porto Cabello

Location of holotype, M.C.Z. no. 115,702.

Remarks. Sowerby (Thes. Conchy. 2, pp. 534-535, 1855) describes
Neritina listeri and gives three figures of his species on pi. 116, figs.
249-251. There has been a confusion of species here, since two are
represented by his figures. Figures 250 and 251 are without question
Neritina zebra Bruguiere. This leaves figure 249 without a name. As
this figure is too poor to be designated as a type we have selected a
holotype based upon a recently collected series, M.C.Z. No. 115,702,
from Waunta Lagoon, Nicaragua.

This is a brackish water species inhabiting swampy areas. I am
indebted to Doctor Elizabeth Deichmann, M.C.Z., for ecologic notes
pertaining to this species. She found it on the stalks of brackish
marsh grasses in the Caroni Swamp, Trinidad. The swamp was tidal
with floating masses of fine, green algae. There were mangrove islands
present and a muddy bottom. The water was shallow and became
very warm at mid-day, under the tropical sun. It was filled with
crustaceans to such a degree that it was almost soupy. The author
has taken this species in brackish water at the head of Samana Bay,
Santo Domingo, near the mouth of the Yuna River. Here also it was
found on stranded logs and mangrove trunks above the mud and
water.

Range. Central America : British Honduras — Belize

Guatemala — Cavech River, Livingston; Quirigua, Maya Farm

Honduras — Punta Patuca

Nicaragua — Waunta Lagoon; San Juan del Norte
South America : Venezuela — Rio Yaracuy, 45 km. N. W. of Puerto
Cabello

Brazil — - Parnahyba

384 bulletin: museum of comparative zoology

Greater Antilles; Cuba — Rio San Juan and Rio Canimar, Matan-
zas Prov. ; Zapata

Santo Domingo — Sanchez ; Yuma
Lesser Antilles : Trinidad — Caroni Swamp

Neritina reclivata (Say)
pi. 3, fig. 7, 8

Natica reclivatus Say July 1822, Journ. Acad. Nat. Sci. Philadelphia, 2, pt. 2,
p. 257.

Neritina jamaicensis C. B. Adams 1851, Contrib. to Conch. 1, p. 175.
Neritina reclivata sphaera Pilsbry 1931, Nautilus 45, no. 2, pp. 67-68.
Type locality. St. John's River, Florida (Say).

Description. Shell globose and thin. Ground color deep purple
brown, olivaceous green, grey or horny yellow with many fine, close
angular, black, brown to lavender lines axially arranged. A sub-
sutural black line follows the suture from spire to the tip of the aper-
ture. Shell thinly impregnated by the pigment which scrapes off
easily leaving a greenish-grey surface exposed. Whorls 2^, rounded
in cross section. Spire very low to rather prominent, often more or
less corroded and cast at an angle of 90°-115°. Aperture lunate cast
at an angle of 81°. Palatal lip thin, sharp, slightly sinuous, long and
bearing no teeth. Color white to bluish white. Palatal hinge tooth
strong. Parietal area smooth, convex, bluish-white to dirty yellow and
bearing a variable number of fine irregular teeth on the columellar
edge. Suture faintly impressed. Periostracum thin. Sculpture con-
sisting of many faint axial lines of growth.

Operculum opaque, calcareous and black to slightly brownish.
Exterior bearing many fine lines of growth radiating from the nucleous.
Peg and rib strong; peg smooth. Periostracal layer present along the
palatal margin.

Radula. Specimens identified according to shells and opercula were
Neritina reclivata Say M.C.Z. No. 100295. The radulae vary as
follows :

Specimens from Hillsboro River, Tampa, Florida.

R-central square to rectangular.

A-central, central end of some teeth twice as wide as some
others in the same radula. Lateral end varies from possessing a
prominent posterior lobe to having merely a broadly rounded angle
at that part of the tooth.

B-central typical.

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

385

C-central typical.

D-lateral typical but usually with very much reduced
outer projection of inner arm of Y-thickening.

E-lateral typical in shape but bearing from 12-14 denticles
in one radula and from 16-20 in another on the edge of the elliptical
reflection and 34 in still another from Tampa.

Inner marginals with claw-like anterior portion and bearing 9-11
rather broad, curved, short denticles arranged chiefly on the anterior
edge. These become longer and arranged in an oval as we progress
laterally (see fig. 1). x\nterior tips of marginals taken from a row of
teeth and arranged from the innermost near the laterals on the left
towards the outermost at the margin of the radula on the right.

Fig. 2. Reading from left to right are the inner middle and two of the outer
marginals of Neritina reclivata Say.

Specimens from the Caloosahatchee River, Fort Myers, Fla. were
identified by the opercula and shells as Neritina reclivata Say, M.C.Z.
No. 56451. Radulae from this lot vary as follows:

R-central typical to narrow and rectangular.

A-central varying from longer and narrower than typical
and with a reduced posterior lobe to shorter, broader and with larger
to bluntly pointed posterior lobe than is typical. The position of the
posterior lobe also varies from near the lateral end of the tooth to
midway and ever slightly nearer the central end.

B-central typical.

C-central typical.

D-lateral typical.

E-lateral varies from those with a prominent major inner
cusp to those in which it is but little more than one of the denticles on
the anterior edge of the elliptical reflection. These denticles vary from

386 bulletin: museum of comparative zoology

8-9 strong, rather blunt ones in one radula to 17-22 fine ones in
another.

Inner marginals bearing about 10 rather blunt cusps as in fig. 2;
middle marginals with 10-12 longer, curved and more pointed denticles
as in fig. 2. Outermost marginals — some appear to bear 8-12 minute
denticles while others do not appear to bear any but one of the type
shown as fig. 2 No. 4 where the anterior tip of the tooth is curved over.

The above shows that there is considerable difference in the radula
of a species from the same locality and from place to place. The
specimens from these two Florida localities are particularly interest-
ing, not only because there is great variation in the radulae in the
individuals, but also because these colonies were pure with no other
species with which to interbreed. For this reason these lots were
selected for radula examination. If we wished to base our specific
determinations on radula differences alone, we could make several
new species from the individuals found at these two places. Other
considerations such as the shell and operculum would not permit of
such a division, however, and the author feels it far wiser to leave the
species intact. Radula variations or similarities could be explained by
a number of influences other than those of chromosomal change with
the resultant species formation. For example, with individuals feeding
upon the same type of food we should expect to find rather similar
weapons of a rasping nature for the preparation for digestion. This is
exactly what we do find. An examination was made of the radulae of
members of widely separated molluscan families that are plant and
lichen feeders. They are as follows: Cerionidae, Bulimulidae, Uri-
coptidac, Camaenidae, and Veroniccllidac. The localities that were
selected for the individual specimens are widely separated. For
example, Cepolis varians Menke is from the Bahamas. Crystallopsis
tricolor Pfeiffer is from the Solomon Islands. V eronicella occidentalis
Gldg. comes from Jamaica. Crystallopsis debilis Clapp is reported
from the Solomon Islands. Cerion incanum Binney is from Upper
Matecumbe Key, Florida. Liguus fasciatus viridis Clench comes from
Soledad, Cienfuegos, Cuba, and Urocoptis pruinosa Morelet is from
Sierra de Casas, Isla de Pinos, Cuba. There were also other members
of the above families that were studied, but this list of species serves
to cover the families examined. Most of these families are widely
separated; some of them in both classification and actual distance.
Nevertheless, they show a remarkable resemblance in the general type
of radula they exhibit. They all possess square to rectangular teeth
bearing a cusp or cusps. The teeth are placed in rows one next to

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

387

another and resemble tiers of blocks. The radula ribbon resembles a
coat of chain mail. The teeth in any one radula are of one type — the
block type — though they vary among themselves and from species
to species. The significant point is that these widely separated species
feeding upon similar food possess very similar weapons to deal with it.
A parallelism of diet apparently tends to develop somewhat similar
equipment for utilizing the material eaten. This is exactly what we
should expect. It might be argued equally that since these species
had a similar type of radula, they fed upon the same type of food.
The significant point here is that certain types of radulae have been
developed among the various groups of the mollusca.

Measurements.

maximum

minimum

aperture

length

width

length

Cuba: Zapata, Santa Clara

Province

16.7

17.4

11.4

14.8 mm

Florida : Caloosaha tehee

River, Fort Myers

17.2

16.8

10.8

14.0

Mexico

15.7

16.2

10.0

13.2

Nicaragua

16.0

15.4

10.3

14.0

Santo Domingo: Sanchez

17.4

11.1

14.5

The holotype is A.N.S.P. No. 37575.

Remarks. Neritina reclivata Say is a rather unique species and
differs sufficiently from other axially striate rather elongate to low
spired forms such as Neritina cumingiana Recluz or A7, roissayana
Recluz to be readily distinguishable. Though it becomes quite globose
in Cuba, Hispaniola and Nicaragua, it never becomes as globose as
Neritina zebra Bruguiere, which is the only other comparable form
which is axially striate, from the West Indian region. The more
globose form, a golden yellow color and broader, more widely spaced
black axial lines of N. zebra, separate it easily from N. reclivata.

A comparison of the ranges is equally interesting. N. reclivata Say
occurs with certainty from Florida south to Panama and is found on
certain of the larger West Indian islands (Cuba, Hispaniola, and
Jamaica), where there are large freshwater rivers. N. zebra Brug.
occurs in Venezuela and Bra? il and, so far as known, does not inhabit
the larger islands of the West Indies.

Specimens of N. reclivata from Cuba have been examined and it was
found that there was a gradual gradation of color pattern between that
of Neritina virginea and N. reclivata. Gradations of pattern from N.
reclivata to Neritina piratica Russell have been noted also. An injury

388 bulletin: museum of comparative zoology

to the shell of the former apparently sometimes tends to bring about a
reticulation of the pattern and this grades into the less and then the
more reticulated forms of the latter. N. zebra does not seem to grade
into either N. rcclivata, N. virginea or Ar. piratica though the last two
are found within its range. Further investigation might disprove
this, but specimens of N. zebra that I have examined do not seem to
grade into these other forms. This leads to the speculation that possibly
interbreeding takes place with the other three but that it does not take
place with N. zebra. Nothing definite can be said in regard to this in
the light of our present limited information.

Pilsbry, pp. 67-68 based his N. rcclivata sphaera upon a lower spire,
more globose form and relatively larger aperture than N. rcclivata.
Also the color of the variety is a grape green in contrast to the darker
green of Ar. rcclivata. An examination of Pilsbry's holotype (A.N.S.P.
No. 154935) and two paratypes in the collection of the M.C.Z. No.
(86377 and 83938) show that this form is not sufficiently different from
true Ar. rcclivata Say to be worthy of varietal rank. Pilsbry states that
"in many hundreds of N. rcclivata from many localities there are none
having the globose shape, short spire and relatively large aperture of
these shells." The globose form of N. rcclivata is found chiefly on the
larger West Indian islands. It is also known from other places in
Florida and in Central America. We have seen specimens of this
spherical form from the following localities :
M.C.Z. No. 132776 from Mobile, Alabama

" 56451 " Caloosahatchee River, Fort Myers,
Florida

" 53028 " Tampa, Florida

" 134224 " Palm River, Hillsboro Co., Florida

" 104307 " Gulfport, Florida

" 134222 " Livingston, Guatemala

" 134223 " Creek West of Livingston, Guatemala

" 135575 " Waunta Lagoon, Nicaragua.
The radula of N. rcclivata sphaera could not be studied since none of
the specimens possessed soft parts from which to extract it. The
operculum is similar to that of Ar. rcclivata.

Having seen Adams' cotype of Neritina jamaicensis in the collection
of the U.S.N.M., the author believes that it is a synonym of Neritina
reclivata Say. Also, specimens with a grey color have been observed
in various museums. The lavender axial lines, however, seem to be
limited to those specimens from the central block of WTest Indian
islands as this character has not been observed elsewhere. Many of

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 389

the specimens bearing the lavender axial lines were old, much worn,
and some stained with reddish earth. The cotype of Neritina jamaicen-
sis Ads. bears the most brilliant axial lavender lines that the author
has observed. It also is a much worn specimen, is rather small and
globose with low, rounded spire, and lacking all columellar teeth. It
possesses a lunate aperture, and does not vary much from specimens of
Neritina reclivata Say. Since specimens of Neritina reclivata from
Florida may possess a grey ground color and since true N. reclivata is
found in Cuba and Santo Domingo, then the only observable difference
between N. reclivata Say and N. jamaicensis Ads. is the presence of
lavender, axial lines in the latter species. Such a difference is hardly
worthy of even varietal rank and the author therefore believes that it
should be made synonymous with Neritina reclivata Say.

Neritina reclivata Say is a brackish to freshwater species. It is found
on any hard object in the water but not in mud. It is of interest to note
with respect to its distribution that it occurs chiefly on those West In-
dian islands that are large enough to support permanent freshwater
rivers and, apparently not further south than Santo Domingo or
Puerto Rico. It has not, to my knowledge, been reported from Guade-
loupe, the Barbados — islands that have permanent rivers. For what
reason they are not found on these islands, I cannot say. It is possible
that more extensive collecting might bring them to light here. This
species is not found on the Bahama Islands. There are no permanent
freshwater rivers on these islands. In Florida, Ar. reclivata is found on
the stems of reeds and other water plants growing in the shallow water
of drainage canals a few miles from the ocean. Apparently they feed
upon many species of algae according to H. A. Pilsbry (Nautilus 45,
p. 68, 1932).

Range. United States: Florida — St. Augustine; West Palm Beach;
Miami; Punta Rassa; Fort Myers; Tampa; Cedar Keys; Suwannee
River; Gulf port; Pensacola

Alabama — Mobile; Bon Secour

Mississippi — Biloxi

Louisiana — Grand Isle; Shell Island; Marsh W. end of Rigolet
Bridge

Texas: Port Arthur; Rockport; Corpus Christi
Central America: Mexico — Tampico; Vera Cruz

British Honduras — Belize

Guatemala — Cavech River, Livingston

Honduras — Puerto Cortes ; Punta Patuca

Nicaragua — Waunta Lagoon; Bluefields; San Juan del Norte

390 bulletin: museum of comparative zoology

Costa Rica — Limon

Panama — Puerto Bello
South America : Colombia — Maracaibo

Greater Antilles: Cuba — Rio Almendarez, Havana Province; Rio
Canimar, Matanzas Province; Zapata; Manzanillo

Santo Domingo — Sanchez

Jamaica — Montego Bay; Port Antonio; Black River Bay
Lesser Antilles: Trinidad

Neritina zebra (Bruguiere)

pl. 4, fig. 1, 2

Nerita zebra Bruguiere 1792, Actes de la Soc. d'Hist. Nat. Paris, p. 126.
Neritina lineolata Lamarck 1822, Hist. Nat. Anim. Sans vert. 6, pt. 2, p. 186.
Nereina lacustris Cristofori & Jan 1832, Mantissa, Pt. 2, Cat. test. p. 4, sp.
31-18.

Type locality. Cayenne (French Guinea, South America (Bruguiere).

Description. Shell globose and rather thick in adult specimens.
Ground color yellow to reddish brown with fine widely spaced or thick
zigzag axial black lines that are usually more or less parallel. A sub-
sutural black line usually is present but not always. Black lines scrape
off fairly easily leaving a chalky surface exposed. Whorls about 2 to
23/2> rounded in cross section. Spire low, rounded and often con-
siderably corroded, cast at an angle of 103°. Aperture lunate, slightly
sinuous and cast at an angle of 80°. Palatal lip thin and sharp and is
usually marginate. There is a low, rounded superior prominence on
the margination. Palatal lip shorter in proportion to maximum width
of the shell than in Neritina virginea L. and bears no teeth. Color grey
to bluish-white. Palatal hinge tooth strong. Parietal area smooth,
convex, chalky white to dirty yellow, bearing a variable number of
fine dentations on the columellar edge. Suture faintly impressed.
Periostracum thin. Sculpture consisting of many faint axial lines of
growth.

Operculum calcareous, opaque, bluish-black. Exterior bearing many
fine lines of growth radiating from the nucleus. Peg strong, smooth.
Rib tends to be reduced. Periostracal layer present along the palatal
margin.

Radula R-central width approximately equal to length — typical.
A-central typical. Some with more prominent posterior
lobe than others.

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 391

B-eentral typical.

C-central with slightly more vertical ridge than typically.

D-lateral typical but with reduced outer projection of inner
arm of Y-thickening.

E-lateral typical in shape and bearing 8-12 course denticles
on the anterior edge of the elliptical reflection.

Inner marginals with long claw-like anterior part bearing 8-9 rather
short pointed denticles. The middle marginals bear about 10-11 more
curved and slender denticles than those of the inner marginals and are
arranged semicircularly. The outermost marginals appear typical and
uncusped.

Measurements.

maximum

minimum

aperture

length

width

length

Brazil: Bahia

20.0

21.4

14.1

17.7 mm

Brazil: -

13.7

14.4

9.5

12.6

Remarks. The genus Nereina was described by Cristofori and Jan
with N. lacustris as its only species I.e. (1832). This we now consider
to be the same as Neritina zebra Bruguiere. As the name Nereina
antidates Vitta Morch (1852) it will therefore have to replace this
subgenus. As stated before, Ar. zebra is very closely related to N.
virginca Linne, which is the type species for the subgenus Vitta. There-
fore, Nereina C. and J. replaces Vitta Morch.

Neritina zebra is the most globose species occupying the West Indian
region. It resembles Neritina reclivata Say most closely but may be
distinguished from that species by its more globose form, more golden
yellow to dark brownish red color, the comparatively shorter, wider
palatal lip and the usually broader, more widely spaced axial lines.
These lines tend to zigzag more with Neritina zebra than with Neritina
reclivata. There are albinistic forms of N. reclivata occurring in Cuba
that appear most like Neritina zebra since the lack of the usual oliva-
ceous coloring of N. reclivata leaves the shell with a corn yellow color.
The characters as given above, however, will distinguish these species.

Neritina zebra Brug. is a brackish to freshwater form occuring
chiefly in Brazil. Beyond this, we know very little of its ecology.

Range. Central America : Honduras — Aguan

South America: Venezuela — Caracas; Pedernales; Curiapo
French Guiana — Cayenne
Brazil — Para; Bahia; Pernambuco; Itabapuana

392 bulletin: museum of comparative zoology

Subgenus neritina s.s. Lamarck
Neritina punctulata Lamarck

pl. 4, fig. 3, 4

Neritina punctulata Lamarck 1816, Encycl. Meth. 3, pl. 455, fig. 2a-b, list p.
11 (n. andf.).

Type locality. Rio Grande, Port Antonio, Jamaica (Russell).
Description. Shell subpatelliform and thin. Ground color varies
from a deep to a light brown or occasionally reddish pink. The ground
color is either solid or made up of fine, close, zigzag axial lines. Some-
times there will be two or three broad bands of reddish pink alternating
with one or two brown ones. Superimposed upon this background are
many suboval, light yellowish brown spots. These are outlined on the
"leading," aperture edge, by a heavy black line. The "trailing,"
spireward, edge is not thus outlined. Shell thinly impregnated by
pigment which is usually no more than a thin, brownish layer punc-
tulated by white spots. The pattern is carried mainly by the rather
thick periostracum. Shell is bluish-white beneath the periostracum.
Whorls 134 to \}/2 rounded in cross section. Spire very low, rounded,
and often much corroded leaving an irregular indentation. It is cast
at an angle of 127°. Aperture lunate and cast at an angle of 76°.
Palatal lip thin, sharp, and slightly sinuous. It is long, bearing no
teeth and bluish-white. Palatal hinge tooth strong. Parietal area
smooth and flat. Columellar edge bearing no teeth or very faint
irregular ones. Columellar area white to dirty yellow. Suture faintly
impressed. Sculpture consisting of many faint lines of growth axially
arranged.

Operculum opaque, calcareous and pinkish white to salmon pink.
Exterior bearing many fine lines of growth radiating from the nucleus.
Peg prominent and smooth. Rib very prominent. Periostracal layer
present along the palatal margin.

Radula. R-central almost square, slightly smaller at the anterior
end, resembling a kernel of corn.

A-central broad with large, smoothly rounded posterior
lobe which forms the whole lower corner of the tooth in that section.

B-central typical with "s"-shaped ridge.

C-central with ridge more vertical than normal.

D-lateral with a reduced outer projection of inner arm of
Y-thickening.

E-lateral with large crescent-shaped elliptical reflection

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

393

which is sharply pointed on the edge towards the center. The lateral
edge of the elliptical reflection bends sharply in the direction of the
crescent forming a loop which fades into the posterior body of the
tooth. This loop is characteristic of Neritina punctulata.

Inner marginals with large, claw-like anterior part which bears 14
short, blunt denticles on its anterior side. These denticles become
curved, longer and placed in an oval formation on the teeth as we
proceed laterally. The outermost marginals bear about 14 minute
denticles.

Measurements.

Guadeloupe Id.:

near St. Rose
Guatemala: Livingston,

Cavech River
Jamaica
Puerto Rico : El Yunque,

20 mi. E. of San Juan

maximum minimum aperture

length width width length

19.6 20.2 10.2 20.1 mm.

26.5
23.0

18.5

26.4
23.0

18.0

14.8
12.4

10.2

25.4
21.5

15.5

These specimens from the Cavech River, Guatemala, were not only
unusually large, but had very heavy shells and were in general a lighter
brown than usual.

Type locality here selected is Rio Grande, Port Antonio, Jamaica.

Remarks. The original locality is not given by Lamarck. The first
locality appeared as Indies and Mollucas, when Deshayes synonymised
Lamarck's species with Neritina pulligera L. in Encycl. Meth. 3, p. 625,
1830-32. Deshayes refers to Lamarck's plate 455, fig. 2a-b. This is
Neritina punctulata Lm., which is a valid species from the West Indies
and should not have been synonymised. Since a type locality was not
given by Lamarck, a type locality is here selected. It is Rio Grande,
Port Antonio, Jamaica.

Neritina punctulata stands alone in regard to the relationship of size
and shape among the species of the West Indian region. Its sub-
pa telliform shape approaches that oV Smaragdia viridis Linne and
Neritilia succinea Reel., but it is far larger than either of these. It is
similar in shape to Neritina pulligera Linne from the East Indies, yet
this species is usually larger. N. pulligera has a flat to concave colum-
ellar area and slightly convex operculum which is relatively wider for
its length than that of N. punctulata, and has a far greater relative
distance between the strong rib and peg than has N. punctulata. Also,
the color pattern of N. punctulata, which is made up of many light sub-

394 bulletin: museum of comparative zoology

circular olivaceous spots on a brown to dark olive background, would
prevent confusion of these two species. The color of N. pulligera is a
solid deep brown over the entire shell with a red columellar area, as
opposed to the dirty-orange to yellow columellar area of Ar. punetulata.

Those variants of Neritina clenchi Russell from Guadeloupe approach
N. punetulata in shape somewhat but are more highly spired and more
globose. Also, the interior of the aperture of N. clenchi is usually
bluish-white, though sometimes it contains yellow similar to those
specimens from Guadeloupe. The aperture is wider in proportion to
its length in N. punetulata. The color patterns of some specimens of
the two forms are rather similar but those from Guadeloupe are much
darker brown. There has been some confusion in the past between
Neritina virginea Linne and N. punetulata Lamarck, but the sub-
patelliform, low spired form of the latter should distinguish it im-
mediately from the globose, rather higher spired shape of the former.
Neritina latissima Broderip from the rivers of western Central America
somewhat resembles A', punetulata in general shape and color pattern
but the development of lateral wings on the palatal lip will distinguish
these two species. Also the aperture of Ar. latissima is relatively wider
in relation to its length than that of AT. punetulata.

In the Alan. Conch. 10, 1888, p. 60, it is stated that A", punetulata
"probably inhabits the sea as well as fresh water." It is also stated that
it occurs from Panama to Mazatlan, Mexico, on the west coast of
Central America. I believe that there has been a confusion here of the
two species N. punetulata from the West Indian region and N . latissima
Broderip from the west coast of Central America for the following
reasons :

1. Tryon, p. 60, p.l 20, fig. 41 makes Neritina turbida Morelet a
synonym of N. punetulata. Lot M.C.Z. No. 21170 from Rio Lajas,
Nicaragua, contains young specimens of Ar. latissima that are identical
with Tryon's fig. 41. It is therefore here considered that A", turbida
Morelet is a synonym of A", latissima.

2. On the same page as above Neritina bahiensis Recluz 1850, is
synonymised with N. punetulata, and is reported from Bahia, Brasil.
Recluz, in the above reference, p. 155, merely reports it as "Hab:
Bahia (M. Janelle)". Recluz' figure and description leave no doubt in
the author's mind that this is N. latissima. The reference to Bahia may
be an error or it may not mean Bahia, Brazil. For example, there are
many localities on the west coast of Central America that bear the
name Bahia. Also, the Spanish word "bahia" means "bay" and pos-
sibly Recluz' reference to it as a locality merely meant that his speci-

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 395

mens were collected from a bay. The reason for assuming that Bahia,
Bra7il, is an error is that nothing similar to Recluz' species, N. bahiensis
has been taken from Bahia, Brazil, since then.

In the Man. Conch. 1888, 10, p. 60, Neritina cassiculum Sowerby is
synonymised with A", punctulata. In the collection of the M.C.Z. are
specimens from Mazatlan, Mexico, and Realejo Bay, Nicaragua, that
are identical with Sowerby's figure. Since N. punctulata is a species
from the West Indian region and not from the west coast of Central
America, and since they are very different in shape and color pattern,
I do not believe that N. cassiculum is a synonym of N. punctulata.
Furthermore, N. punctulata is a freshwater species, while N. cassiculum
is very probably a brackish to salt water form. Ar. cassiculum is here
considered to be a valid species since it is quite different from any other
of the Ncritinas from the west coast of Central America.

N. punctulata is a freshwater species occurring on stones and boul-
ders in moderately swift to rapid water. Localities of this type where
specimens have been taken in Jamaica are the Great River Rapids,
Flint River, and at "Mount Pleasant" in Hanover. This species is
found in company with AT. clenchi Russell and Neritilia succinea
Recluz. It is interesting to note that the specimens of N. punctulata
from the Cavech River, Guatemala, are much larger and have far
thicker shells than those from any other locality from which the M.C.Z.
possesses specimens. This may be due to a less acid condition of the
water, a greater quantity of lime available for the formation of shells,
or a very vigorous race of individuals. Neritina piratica Russell from
the same locality reaches its largest size here, also, and possesses the
thickest shells found for this species throughout its range. From the
point of view of the relationship between the individual and its en-
vironment, these two species could be studied here to advantage.
Range. Central America ; Mexico — ■ Vera Cruz

Guatemala — Cavech River, Livingston

Nicaragua — Waunta Lagoon
Greater Antilles : Cuba — Baracoa

Haiti — Jeremie; Port de Paix

Santo Domingo — Yaguada, Cape San Rafael

Jamaica — Lueea; Montego Bay; St. Ann's Bay; Annotta Bay;
Rio Grande, Port Antonio; Buff Bay River; Morant Point; Kingston

Puerto Rico — San Juan; El Yunque, 20 mi. E. of San Juan
Lesser Antilles : Guadeloupe — St. Rose

Dominica — Laudet

Martinique

Grenada

396 bulletin: museum of comparative zoology

Genus Smaragdia Issel

Smaragdia viridis weyssei Russell

pi. 4, fig. 5, 6; pi. 7, fig. 2

Nerita viridis Linne 1758, [in part], Syst. Nat. ed. 10, p. 778.

Neritina (Smaragdia) viridis L., G. W. Tryon Jr., 1888, Man. Conch. 10, pp.

54-55, pi. 18, fig. 88.

Smaragdia viridis Linn6, H. B. Baker, 1923, Proc. Acad. Nat. Sci. Philadelphia

70, pp. 173-174.
Smaragdia viridis weyssei Russell 1940, Mem. Soc. Cubana Hist. Nat. 14,

p. 257, pi. 46, fig. 5-6.

Type locality. Miami, Florida.

Description. Shell subpatelliform and thin. Color grass green, with
irregular white spots varying in shape, size and number. Shell thinly
impregnated by pigment. Pattern scrapes off easily leaving a green

Fig. 3. Right half of a radula row of Smaragdia viridis weyssei Russell.
R-central, A-central, B-central, C-central, E-lateral, broad inner and slender
outer marginals.

surface exposed. Whorls l}/i-\]4, and rounded in cross section. Spire
very low, rounded, not corroded and cast at an angle of 138°. Aper-
ture lunate and cast at an angle of 76°, very slightly sinuous. Palatal
lip long, sharp, thin and bearing no teeth. Palatal hinge tooth faint to
lacking. Parietal area convex, smooth and whitish-green; columellar
edge bearing 6-8 small, irregular teeth. Suture faintly impressed.
Periostracum thin, very finely pitted. Sculpture consisting of many
faint axial lines of growth.

Operculum opaque, calcareous and greenish-white. Periostracal

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

397

maximum

minimum

aperture

length

width

length

5.1

5.2

2.6

4.9 ram.

5.7

6.0

2.8

5.2

layer present along the palatal margin. Exterior bearing many faint
lines of growth radiating from the nucleus. Rib and peg strong. Peg
smooth.

I have been unable to locate the denticles on the marginals that
Baker mentions on p. 174, nor do I find any on the E-lateral that he
mentions on p. 173. The edge of the elliptical reflection bears no
denticles in the seven complete radulae examined under high and low
powers of the microscope.

Measurements.

Florida — Miami

West Indies — Barbados

Virgin Ids., Guana Id.,

Tortola 4.8 5.1 2.6 4.1

Location of type M.C.Z. No. 88815.

Remarks. Linne in his Syst. Nat. ed. 10, p. 778, 1758, bases his
description of Nerita viridis upon and refers to the work of Patrick
Brown, "The Civil and Natural History of Jamaica," Pt. 2, p. 399,
1756. Linne does not refer to any work containing figures of his
species. He describes it as coming from the Balearic Islands, Mediter-
ranean Sea, and Jamaica, West Indies. There appear to be certain
constant differences between the forms from these localities. Since
Linne based his description upon specimens from these two localities,
it leaves the matter open as to which name shall be employed for
these two forms. It is proposed here to use the name Smaragdia viridis
L. for the European form and limit S. viridis weyssei to the western
Atlantic region. The European form possesses solid black axial, more
or less parallel lines on the whorls. Whether the specimen be young or
old, these lines are found on almost 100% of the specimens. These
narrow lines are located on the spireward or "trailing" edge of the
white spots. The West Indian form never possesses these solid black
lines, but rather, when present, a brownish-red "trailing" edge to the
white spots. Also, these brownish-red edges occur in only from 3-15%
of the specimens from any one lot amLseem to be linked with the age
of the specimen in some way since no adults could be found that
possessed them. It is interesting to note according to Woodring, p.
427, that the fossil forms that have been found in the Cercado forma-
tion of the Dominican Republic may indicate that the West Indian
and European forms may have been the same during the Middle
Miocene, but the living one from the West Indies possesses differences
that mark it off very definitely from the European. Therefore it is

398 bulletin: museum of comparative zoology

felt to be wiser to keep the West Indian form as a variety of the
European rather than raise it to specific rank.

Smaragdia viridis weyssei is a saltwater species living on eel grass
and under broken rocks. The author has collected it alive on several
West Indian islands, but never abundantly at any one place. It is an
easily recognizable species and cannot be confused with any other of
the West Indian forms. For this reason it would make an excellent
one for ecologic or distributional studies. It also seems to have no
other forms with which to interbreed and must, therefore, be a com-
paratively pure strain.

Range, (see pi. 7, fig. 2). The range extends from Palm Beach,
Florida, and the Bermudas through the West Indian islands as far
south as Bonaire. It has also been reported from Vera Cruz, Mexico,
Progreso, Yucatan and Belize, British Honduras, Central America.
No records from South America have been seen.

Genus Neritilia von Martens

Neritilia succinea (Recluz)

pi. 4, fig. 7, 8

Nerita succinea Recluz, 1841, Rev. Zool. 77, p. 343.

Nerita pygmaea C. B. Adams 1845, Proc. Bost. Soc. Nat. Hist. 2, p. 7.

Neritina succinea Recluz, v. Martens 1879, Conchy. Cab. 2, pt. 10, pp. 242-243,

pi. 23, fig. 23, 24.

Neritilia succinea Recluz, Tryon 1888, Man. Conch. 10, p. 54, pi. 17, fig. 83;

H. B. Baker 1923, Proc. Acad. Nat. Sci. Philadelphia, 75, p. 172, pi. 16,

fig. 41.
Neritilia succinea guatemalensis Pilsbry 1919, Proc. Acad. Sci. Philadelphia
70, p. 172.

Type locality. Madagascar and Guadeloupe (Recluz).

Description. Shell subpatelliform, thin, and uniformly horn colored.
Color deeply impregnated in shell which does not scrape off. Whorls
234 to 23^2, rounded in cross section. Spire very low, rounded, not
corroded and cast at an angle of 152°. Aperture lunate, slightly sinu-
ous and cast at an angle of 90°. Palatal lip long, sharp, thin, and
bearing no teeth. Palatal hinge tooth lacking. Parietal area smooth,
flat to slightly convex, bluish-white to horn colored. Columellar edge
bearing no teeth. Suture very faintly impressed. Periostracum lack-
ing. Sculpture consisting of many very faint axial growth lines.

Operculum calcareous, white to horn colored. Periostracal layer

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE

399

present along the palatal margin. Exterior of operculum with many
fine concentric lines of growth. Inner side bearing a ridge on the
parietal margin and basal third of the palatal edge. This ridge rises to
form a smooth, prominent peg at the inferior palatoparietal junction.
There is a horizontal "V"-shaped depression at the base of the superior
palato-parietal junction. The angle of the "V" points in the direction
of the peg.

Fig. 4. Radula teeth from a row in the left half of the radula of Neritilia
succinea Recluz. The R-central is lacking. A-B-C-centrals together, B-cen-
tral, C-central, E-laterl, an inner and one of the outer marginals.

The teeth of the radula of Neritilia succinea Reel, are very similar to
those of Neritilia rubida Pease as given by Baker, pi. 16, fig. 42. N.
rubida is an East Indian species, while N. succinea is a West Indian
one, yet their radulae are surprisingly alike. There is little more
difference than one would expect to find between the individuals of a
species. The number of denticles on the E-lateral vary slightly and the
B and C centrals vary in shape somewhat also, but, in general, the two
are remarkably similar.

400

bulletin: museum of comparative zoology

Measurements.

length

maximum
width

minimum
width

aperture
length

Jamaica : Great River

Rapids

5.3

5.7

2.9

4.4 mm.

Hispaniola : Mouth of Sosua

River, Puerto Sosua

4.6

5.2

2.7

4.4

Guadeloupe Id.

4.0

4.4

2.5

3.8

Neritilia succinea has not been reported from Madagascar since
Recluz described it from there in 1841. He also reported it from
Guadeloupe. We have records which show that this species exists in
the West Indian Region. Therefore, the latter locality is here con-
sidered to be the correct one.

Remarks. Neritilia succinea Reel, is closely allied to N. rubida Pease
and N. manoeli Dohrn from the East Indies and Princes Island, Gulf
of Guinea. It differs from N. rubida in possessing a somewhat smaller
callous, a larger palatal lip and the peg of the operculum is usually
much more pronounced. The size of the two is about equal and the
radulae are very similar. There is greater difference between N. suc-
cinea and N. manoeli from across the Atlantic Ocean than there is
between N. succinea and N. rubida from the Pacific Islands. N.
manoeli is a very much smaller specie?, attaining only about x/i the size
of Ar. succinea and with a much reduced peg on the operculum.

Recluz distinguishes N. succinea from Smaragdia viridis and S.
rangiana Reel, as follows: This species (N. succinea) cannot be con-
fused with Nerita (Smaragdia) viridis L. and Nerita (Smaragdia)
rangiana Reel, because it is deprived of color spots and crenulations
at the edge of the columellar area. Also in general these two species
are larger.

Pilsbry's holotype of Neritilia succinea guatemalensis proves on
examination to be AT. succinea Recluz. Pilsbry's variety is, therefore,
here synonymised with it as above.

Neritilia succinea Reel, is a freshwater species inhabiting streams
both swift and slowly flowing. It is found on rocks among algal tufts
in company with Neritina clenchi and Neritina punctulata. Specimens
sent to the M.C.Z. by Prof. E. A. Andrews carried a limy algal deposit
covering the shell. The author has found specimens of this species
clinging to rocks in a small stream at the mouth of the Sosua River on
the north coast of Santo Domingo.

Range. Central America : Guatemala — Livingston

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 401

Greater Antilles: Haiti — Jeremie, Coteau; Aux Cayes; Port-au-
Prince

Santo Domingo — Little stream, mouth of Sosua River,
Puerto Sosua

Jamaica — Great River Rapids and Flint River, near Mon-
tego Bay; Mount Pleasant near Kingston

Guadeloupe Island

SPURIOUS WESTERN ATLANTIC NERITIDAE

Lepyrium showalteri (Lea)

Neritina showalteri Lea 1861, Proc. Acad. Nat. Sci. Philadelphia 13, p. 55;

Tryon 1888, Man. Conch. 10, p. 53, pi. 17, fig. 81, 82.
Lepyrium showalteri Dall 1896, Nautilus 10, pp. 13-15.
Lepyrium showalteri cahawbaensis Pilsbry 1906, Nautilus 20, p. 51.

As indicated above, this species was 'described as a Neritina, but the
horny operculum, and the radula place it with the family Hydrobiidae.

The type locality for this species is the Coosa River above Fort
William, Alabama. It has not been reported since from that locality
and it is very possible that there is an error in Showalter's record. The
reasons for believing this, are that we possess several records of
Lepyrium showalteri from the Cahawba River, Alabama, A. A. Hinkley
and Smith did not find it in the Coosa River and though as careful a
collector as R. E. Call collected in both these rivers he only found it in
the former. Dr. Henry Vander Schalie of the Museum of Zoology, Ann
Arbor, Michigan, and Mr. W. J. Clench of the Museum of Comparative
Zoology, at Harvard University, found it at Lily Shoals, Cahawba
River and not in the Coosa River. It is, therefore, reasonably certain
that the Cahawba River is the correct locality.

Lepyrium showalteri Lea has been discussed by the above authors.
It is considered to be a Neritina by Lea. This cannot be, for the reasons
given above. Tryon p. 53 states, "It-has been suggested that this is a
young Anculosa, but it has not the characters of that group; on the
contrary, it more nearly approaches in general Neritina crepidularia.
The coloring of the epidermis more nearly resembles Anculosa how-
ever than the other fluviatile species of Neritina." Tryon's figure
no. 79, pi. 17 representing Lepyrium showalteri, however, neither by
shape nor color markings is this form, but rather very probably
Theodoxus fluviatilis Linne.

402 bulletin: museum of comparative zoology

Dall, Nautilus 1896, 10, p. 15, states, "The Oligocene of Southern
United States contains several species of Neritina, but none, so far as
known, having a close resemblance to Lepyrium, which is, however,
probably an offshoot from Neritina."

Thus far Lepyrium remains a part of, or very closely allied to the
genus Neritina.

Pilsbry's subspecies, Lepyrium showaltcri cahawbaensis, I believe to
be a synonym of Lepyrium showalteri Lea, based upon immature
specimens. This would explain the smaller size, the "straighter
eolumellar edge" and the lack of "a raised outer margin of the colum-
ellar area" upon which Pilsbry based his variety.

From what has been stated above it is obvious that Lepyrium
showalteri Lea is not a Neritina. The author believes that it is closely
allied to the Hydrobiidae, though there are marked radular differences.

Neritina reclivata palmae Dall

Neritina reclivata palmie Dall 1885, Proc. United States Nat. Mus. p. 259, 8,
no. 17.

An examination of Dall's type specimen of N. reclivata palmae proves
it to be Neritina turriia Gmelin of the East Indies. The specimen
was supposedly collected in a brook near Palma Sola, Florida. Very
likely a mixture of specimens had taken place, causing Dall's error.

Neritina jayana Recluz

Neritina jayana Recluz 1850, Journ. Conch. I, pp. 157-158, pi. 7,
fig. 13. Recluz states in his description of this species, that it is from
Dr. Jay of New York and that it cannot be confused with Neritina
(Theodoxus) fluviatilis Linne of Europe because its coloration is con-
stant, the spire is conical and the suture channelled. It is reported
from North America by Recluz. M.C.Z. lot 56628 contains specimens
of Theodoxus fluviatilis from Birmingham, England. These possess a
color pattern similar to that given by Recluz in his figure pi. 7, fig. 13.
They are about the same size as his hair line for it and have an elevated,
conical spire. Some have a channelled suture. I agree with Tryon,
p. 53, that Neritina jayana is doubtfully ascribed to North America
and that it is an Old World species, but I go even further than this and
believe that it is a Theodoxus and very probably T. fluviatilis which
is a very variable species as to color pattern and form.

RUSSELL: MOLLUSKS OF THE FAMILY NERITIDAE 403

Neritina reclivata striolata von Martens

Neritina reclivata striolata von Martens 1877, Conchy. -Cab 2, pt. 10,
p. 120, pi. 13, fig. 12 proves, on examination, to be a synonym of
Neritina smithi Sowerby and not a synonym of Neritina reclivata Say
as it is made by von Martens.

Neritina cassiculum Sowerby
Neritina cassiculum Sowerby 1836, Conch. Ill, fig. 55 under section "Neritina".

Though N. cassiculum has previously been considered (Man.
Conch. 10, p. 60, 1888), a synonym of N. punctulata Lamarck, it is
here considered to be valid and of specific rank for the following
reasons :

Its closest relative among the species of the west coast of Central
America is N. latissima Broderip, which it resembles only in the very
young stages. The color patterns are practically identical being
composed of lighter brown, suboval spots upon a darker background.
These spots are enclosed by brown or black angular lines. Some speci-
mens are banded with two or three alternating stripes of reddish-
orange and dark brown. Here the resemblance ceases. N. cassiculum
is a globose species, usually much smaller than N. latissima, N.
latissima is not globose, but subpatelliform and rather thin shelled
where N. cassiculum is comparatively thick-shelled. The opercula
vary also. That of N. latissima is usually black and rather long for its
width as compared with the brownish-white, operculum of N. cassicu-
lum which is rather thick and wide for its length. Also the latter does
not possess the wide aperture, the sides of which develop wing-like
enlargements such as is the case with the former.

Note concerning Nerita ascensionensis Gmelin

According to von Ihering, p. 534, Nerita ascensionensis occurs on
Fernando Noronha island about two, hundred miles off the coast of
Brazil. As yet we have no evidence that this species has reached con-
tinental America. I have not seen specimens of this species from
Fernando Noronha.

404 bulletin: museum of comparative zoology

BIBLIOGRAPHY

Baker, H. Btjrrington. Notes on the Radula of the Neritidae. Proc. Acad.
Nat. Sci. Philadelphia 75, pp. 117-178, pis. 9-16, 41 figs. 1923).

(a) Clench, William J. Origin of the Land and Freshwater Mollusk Fauna of

the Bahamas, etc. (Bull. Mus. Comp. Zool. 80, pp. 481-541, pis. 1-3,
1938).

(b) Clench, William J. Land and Freshwater Mollusks of Grand Bahama and

the Abaco Islands, Bahama Islands. (Mem. Soc. Cubana de Hist. Nat. 12,
pp. 303-33, pis. 24-25, 1938).

Eyerdam, Walter. The Distribution of Melania and Neritina in the British
Solomon Islands. (Nautilus 50, pp. 44-46, 1936).

Ihering, H. von ,1907, Anales de Mus. Nacional, Buenos Aires, ser. 3, 7,
d. 534.

Martin, A. C. & F. M. Uhler. Food of Game Ducks in the United States and
Canada (Tech. Bull. no. 634, pp. 1-156, pis. 1-153, text figs. 1-137, 1939.

Martini, Friedrich H. W. & Johann H. Chemnitz, Conchylien-Cabinet.
Note. These references are to the New Series. (2, pt. 10, pp. 1-303, pis.
A-23a. Neritina, Nurnberg, 1863-1879).

Moerch, Otto A. L. Catalogus Conchyliorum etc. Comes de Yoldi (fasc. 1,
pp. 1-170, Hafniae, 1852).

Pfeiffer, Lud-wig. Uebersicht der im Januar, Februar und Marz 1839 auf
Cuba gesammelten Mollusken. (Wiegmann's Arch. f. Naturg. 6, pi. 1, pp.
250-261, Berlin, 1840).

Pilsbry, Henry A. A New Race of Nerita reclivata Say. (Nautilus 45, pp.
67-68, pis. 3, fig. 3, 1931).

Roger, E. J. Observations sur les Variations de la Dent Laterale de la Radula des
"Theodoxia" Mollusques Gastropodes Neritides. (Journ. de Conchyl. 78,
pp. 78-79, 1934).

Thiele, Johannes. Handbuch der Systematischen Weichtier Kunde. (pt. 1,
pp. 71-78, Jena, 1929).

Tryon, George W. Manual of Conchology. (10, pp. 18-160, pis. 1-29, 1888).

Woodring, Wendell P. Contribution to the Geology and Palaeontology of the
West Indies, (part 2, Gast. & discussion & results. Carnegie Inst. Wash-
ington, no. 385, pp. 423-427, pi. 35,, figs. 7-13, 1928).

EXPLANATION OF PLATES

PLATE 1

Russell — Mollusks of the Family Neritidae.

PLATE 1

Fig. 1 & 2. Nerita peloronta Linne (Nat. size).

Fig. 3 & 4. Nerita versicolor Gmelin (Nat. size).

Fig. 5 & 6. Nerita fulgurans Gmelin (Nat. size).

Fig. 7 & 8. Nerita tessellata Gmelin (Nat. size).

BULL. MUS. COMP ZOOL.

Russell. Mollusks of the Family Neritidae. Plate 1.

PLATE 2

Russell — Mollusks of the Family Neritidae.

PLATE 2

Fig. 1 & 2. Fluvinerita tenebricosa C. B. Adams (2x).

Fig. 3 & 4. Puperiba pupa Linne (2x).

Fig. 5 & 6. Puperita tristis D'Orbigny (2x).

Fig. 7 & 8. Neritina virginea Linne (Nat. size).

BULL. MUS COr/P. ZOCL.

Russell. Mollusks of the Family Neritidae. Plate 2.

PLATE 3

Russell — Mollusks of the Family Neritidae.

PLATE 3

Fig. 1 & 2. Neritina clenchi Russell, holotype (2x).

Fig. 3 & 4. Neritina meleagris Lamarck (2x).

Fig. 5 & 6. Neritina -piratica Russell, holotype (2x).

Fig. 7 & 8. Neritina reclivata Say (Nat. size).

BULL. MUS. COMP. ZOOL.

Russell. Mollusks of the Family Neritidae. Plate 3.

rju

PLATE 4

Russell — Mollusks of the Family Neritidae.

PLATE 4

Fig. 1 & 2. Neritina zebra Bruguiere (Nat. size).

Fig. 3 & 4. Neritina punctulata Lamarck (Nat. size).

Fig. 5 & 6. Smaragdia viridis weyssei Russell, holotype (6x)

Fig. 7 & 8. Neritilia succinea Recluz (4x).

BULL. MUS. COMP. ZOOL.

Russell. Mollusks of the Family Neritidae. Plate 4.

PLATE 5

Russell — Mollusks of the Family Neritidae.

PLATE 6

DISTRIBUTION MAPS

Fig. 1. Nerita peloronta Linne.
Fig. 2. Nerita versicolor Gmelin.

BULL. MUS. COMP. ZOOL.

Russell. Mollusks of the Family Neritidae. Plate 5.

0 AMLES 300

Fig. 2

PLATE 6

Russell — Mollusks of the Family Neritidae.

PLATE 6

DISTRIBUTION MAPS

Fig. 1. Neritafulgurans Gmelin.
Fig. 2. Nerita tessellata Gmelin.

BULL. MUS. COMP. ZOOL.

Russell. Mollusks of the Family Neritidae. Plate 6.

WEST INDIAN REGION

■23

^ittS 300

Fig. 2

PLATE 7

Russell — Mollusks of the Family Neritidae.

PLATE 7
DISTRIBUTION MAPS

Fig. 1. Neritina virginea Linne.

Fig. 2. Smaragdia viridis weyssei Russell.

BULL. MUS. COMP. ZOOL.

Russell. Mollusks of the Family Neritidae. Plate 7.

WEST INDIAN REGION

0 -MILES 300

Fig. 1

**•*

•23

WEST INDIAN REGION

<s>*

•23

0 -MILES 300

Fig. 2

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE

Vol. LXXXVIII, No. 5

THE CRANIAL ANATOMY OF
ERYOPS MEGACEPHALUS

By H. J. Sawin

With Twelve Plates

CAMBRIDGE, MASS., U.S.A.
PRINTED FOR THE MUSEUM
September, 1941

No. 5 — The Cranial Anatomy of Eryops Megacephalus1
By H. J. Sawin

TABLE OF CONTENTS pAGE

Introduction 407

Materials and Methods 408

General Features of the Skull 412

The Separate Skull Elements 418

Dermal Bones of the Skull Roof 418

Dermal Bones of the Palate 422

The Primary Palatoquadrate Arch 425

The Mandible 427

The Braincase 431

1. Anterior Division 432

2. Posterior Division 437

3. Dermal Elements Associated with the Braincase. . . 440

4. Endocranial Cast 442

Angiology 444

Myology 446

Discussion 448

Summary 456

Bibliography 457

Explanation of Abbreviations 462

INTRODUCTION

Eryops megacephalus, occupying a central position in the Rachito-
mous group of amphibia, is an important representative of the Permo-
carboniferous labyrinthodonts. It is an extremely abundant fossil in
the redbeds of Texas and hence is frequently encountered in the larger
museum collections. Many investigators have been attracted to this
easily available amphibian and, as is shown here, numerous descrip-
tions of the skull or braincase have appeared in the literature. Unfor-
tunately these descriptions, although of great value, are in many
respects incomplete or inaccurate and consequently there is a lack of
agreement regarding a number of morphologically important skull
structures. It is the aim of this study to present a revision of the
cranial anatomy of this animal based on material which has been
prepared by modern methods with the hope that the results will be
of use to students of early tetrapods. It is believed that the fairly
completely ossified skull of Eryops will provide a more sound basis
for the restoration of the nervous and vascular systems than the more

1 Published with the aid of a special gift from Mr. George R. Agassiz.

408 bulletin: museum of comparative zoology

chondrified skulls of Lyrocephalus (Soderbergh, '36) or Benthosuchus
(Bystrow, '39).

Eryops megacephaluswas first described by Cope (1877). An excellent
figure of the skull of the type specimen in a volume of Cope's plates
was edited and published by Matthew ('15). The first attempt at an
analysis of the dermal elements of the skull was made by Broili (1899) ;
Branson ('05) and Case ('11) continued this task, the latter investigator
determining many of the relationships of the bones of the palate.
Broom ('13) first correctly determined the position of the dermal
bones of the entire skull, describing the braincase as well. Von Huene
('13) pioneered in the study of the braincase and cranial foramina.
Watson ('16) provided much accurate information concerning the
cranial morphology of this animal and later ('19) discussed Eryops
in his important study on the evolutionary trends among the Rhachi-
tomi and Stereospondyli. Williston ('18) critically reviewed the con-
clusions of previous workers on the morphology of the braincase and
presented a separate analysis based on newly prepared specimens.
Sushkin ('27) in a general discussion of the palatoquadrate and stapes
included pertinent data concerning various structures of the cranium
of Eryops. More recently Dempster ('35) reinterpreted the braincase
and, by means of a series of endocranial casts, in part determined
the relationships of the brain and inner ear. Muscles of the jaws have
been restored by Adams ('19), while Miner ('25) and Olson ('36) have
restored other portions of the muscular system which impinge on
the skull.

This work has been conducted under the supervision of Professor
Alfred S. Romer, for whose aid and encouragement I am extremely
grateful. I am indebted to Mr. L. I. Price for his indispensable aid in
the preparation of the illustrations. The many favors granted by the
staff of the American Museum of Natural History are to be acknowl-
edged, particularly on the part of Professor W. K. Gregory and Dr.
Barnum Brown. I also here wish to thank Dr. E. C. Olson of the
University of Chicago for the loan of a valuable specimen. The prep-
aration of the sections was achieved by means of apparatus provided
by the Elizabeth Thompson Science Fund.

MATERIALS AND METHODS

A number of the specimens prepared especially for this study were
collected by expeditions of the Museum of Comparative Zoology at
Harvard College in the Wichita Group of the Permo-Carboniferous
Red Beds of Texas; these and most of the other specimens used as
reference material were found in the Belle Plains formation of that

sawin: cranial anatomy of eryops megacephalus

409

Group. As a supplement to these, the collections of the American
Museum of Natural History were examined and one braincase from
the collections of the Walker Museum of the University of Chicago was
studied. The separate specimens are listed as follows with details
concerning their collection and mode of preparation :

PMX

SMX

■A

INF

PRF

POF

\—c

POP

Fig. 1. Eryops megacephalus. Skull in dorsal view, x 3/10.

410

bulletin: museum of comparative zoology

M. C. Z.1 No. 1129. Eryops megacephalus

Collected by L. I. Price during the summer of 1934 at a locality
northeast of the Woodrum ranch-house in Archer County, Texas.
This is in the Wichita Group, the Belle Plains Formation.

Measured in
Centimeters

to
oo

*
o

T— 1

si
.a>

>>"3 oo

a-*

Length of
cranium from
extremity of
quadrate

52.5

48.6

46.8

45.0?

43.7

43.3

42.2

41.2?

38.6

Length of
cranium
along midline

44.0

39.3

40.5

35.8
7.9

36.4

33.5

35.6

34.6

30.7

Length from
end of muz-
zle to
nostril

9.5?

7.3

7.8

7.3

8.9

7.6

7.2?

Width of
cranium be-
tween
quadrates

42.5

35.0?

33.0?

29.5?

31.9

30.6

30.2?

30.5?

25.9

Width be-
tween
orbits

11.0

9.1

9.7

9.0

8.2

8.6

8.2?

9.0?

7.0

Diameter of
orbits anter-
oposteriorly

5.2

5.0

5.8

5.1
5.6

5.0

4.8

Diameter of
orbits trans-
versely

5.8

5.7

5.2

3.4?

* American Museum of Natural History.

This specimen consist of an excellently preserved skull with one
ramus of the mandible missing. The descriptions are based on this
skull with the exception of those concerning certain portions of the
braincase. A comparative series of measurements shows that this
skull conforms closely in dimensions with the type skull as described

1 Museum of Comparative Zoology at Harvard College.

sawin: cranial anatomy of eryops megacephalus 411

by Cope (1877). Corresponding measurements taken from other
entire skulls indicate that M. C. Z. No. 1129 might be described as a
young adult. This diagnosis is indicated also by the state of ossifica-
tion of the skull, the specimen being less well ossified in certain regions
than in the corresponding regions of the skulls of larger specimens.
This lack of ossification, however, is slight and does not interfere with
the interpretation of the fundamental structural relationships.

The preparation of this specimen was done by means of dental
apparatus, a standard dental engine and heavy carborundum drills
being used for the removal of the superficial matrix. This device was
supplemented by hand and mechanical chisels for the coarser work.
Two to 4% phosphoric acid solutions were found to be useful in cor-
roding away the more superficial matrix, the action of the acid being
carefully controlled by protecting the exposed areas of bone with a
thin coating of cellulose acetate. The braincase required special
treatment: the entire preparation was carried on under a binocular
microscope with delicate hand chisels or a high speed hand drill with
small carborundum points. Casts of the endocranial cavities were
made of plasticine and vulcanized latex.

M. C. Z. No. 1407. Eryops sp.

Collected by H. J. Sawin near Electra, Texas, on the east side of
Rough Creek. This locality is in the Clyde Formation of the Clear
Fork Group.

A horizontally cracked sphenethmoid of this specimen clearly
reveals the channels in that bone.

M. C. Z. No. 1458. Eryops megacephalus

Collected by R. V. Witter and party at Slippery Creek, Archer
County, Texas. Wichita Group; Belle Plains Formation.

Serial sections were made from the braincase of this specimen
employing the parlodian peel technique. The posterior portion of the
skull was divided near the sagittal plane; one half was sectioned longi-
tudinally at half millimeter intervals, the other half transversely at
millimeter intervals. The anterior portion of the basisphenoid and the
entire sphenethmoid were sectioned as a separate unit transversely.
The posterior portion of this mass was sectioned at half millimeter
intervals in the region of the basisphenoid and at millimeter intervals
along the sphenethmoid. Casts of these parts were used for the orien-
tation of the sections. The "peels" containing the sections were

412 bulletin: museum of comparative zoology

stained in 2% methylene blue, by which process the bone is clearly
differentiated from the matrix, the latter taking the stain more deeply.
These sections served as a basis for wax plate reconstructions of
parts of the braincase and endocranial cavities.

M. C. Z. No. 1553. Eryops sp., probably megacephalus

Collected by L. I. Price at a locality one and one-half miles north-
west of the Woodrum ranch-house in Archer County, Texas. Wichita
Group; Belle Plains Formation.

One half of this specimen, which consists of the posterior portion
of the braincase, was sectioned in the sagittal plane at millimeter
intervals. This was used as a check on the other sectioned material.

M. C. Z. No. 1651. Eryops sp.

Collected by L. I. Price, west of C. Williams' ranch in Archer
County, Texas, Wichita Group; Belle Plains Formation.

This fragment of the skull is naturally eroded and clearly shows
details in the region of the sella turcica.

M. C. Z. Nos. 1213, 1214, 1223, 1447. Eryops sp.

Collected by T. E. White and L. I. Price in the Putnam, Admiral,
and Belle Plains Formations of the Wichita Group.

These specimens consist of various fragments of the braincase which
have been cleaned by natural agencies and show certain details which
are lacking or obscure in the more complete skulls.

Walker Museum No. 1260. Eryops sp.

Briar Creek, Archer County, Texas. Admiral Formation; Wichita
Group.

This specimen, which consists of the posterior end of a skull, was
figured by Williston ('18) and Sushkin ('27). The stapes is complete
and in position, although somewhat displaced due to crushing.

GENERAL FEATURES OF THE SKULL

The skull is broad and moderately flat. The greatest length is
43.7 cm. from the muzzle to a line between the quadrates; the greatest
width, found in the region of the quadratojugals, is 34 cm. The
greatest height of the skull without the mandibles is 11.3 cm., this

SAWIN : CRANIAL ANATOMY OF ERYOPS MEGACEPHALUS

413

last measurement being taken from parallel lines projected from the
tops of the orbits and the tips of the descending flanges of the ptery-
goids. The ratio of length to width is therefore 1.28+ while the ratio
of height to width is .33+. In lateral view (Plate 4) and particu-
larly in sagittal section (Text fig. 2) the flat character of the skull
is apparent.

Fig. 2. Eryops megacephalus. Sagittal section of skull, x 3/10

A dorsal view (Text fig. 1,' Plate 1) shows the general outline of
the skull which decreases but slightly in width from back to front
until the region about the nares is reached. There a slight indentation
is seen in back of the nares and the muzzle is constricted in front of
the nares, the anterior portion of the muzzle ending bluntly. There
is a noticeable emargination at the widest portion of the skull, here
bounded by the quadratojugals. The quadrates and associated dermal
bones extend well behind and lateral to the occiput. Also prominent
in this view are the openings for the nares, orbits, and the median
parietal organ. The naris, largely closed by the septomaxillary bone,
is 3.6 cm. long and 2.6 cm. high. Facing dorsolaterally, the orbits
measure 5.0 cm. from front to back and 5.8 cm. in width. These
are posteriorly placed, the distance between them and the nares indi-
cating that the region of "intensive growth" of the skull occurred
anteriorly between the orbits and nares. In general Eryops appears
to resemble Cyclotosaurus as described by Bystrow ('35) in its regions
of secondary skull growth, the elongation of the skull occurring in
but one area as noted above. The median parietal foramen is situated
somewhat behind a line across the posterior rims of the orbits and is
0.5 cm. in diameter.

An inspection of the contours of the skull roof (Plate 1) demon-
strates that a somewhat horizontal skull table may be recognized in
contrast to the sloping lateral walls ; this rather flat dorsal area extends
on to the rostral region where it is altered by local changes in contour.

'Seep. 409.

414

bulletin: museum of comparative zoology

Definite longitudinal ridges bound this horizontal area, these extending
from each of the tabular "horns" anteriorly to a point 2 cm. on either
side of the midline at the muzzle. A comparison of dorsal (Plate 1)
and lateral views (Plate 4) enables one to trace the general outline
of these ridges. The skull table slopes gently downward posterior to
the orbits; anterior to the orbits it slopes more abruptly to the muzzle.

Fig. 3. Eryops megacephalus. Transverse sections of skull. Position of sec-
tions shown in Text fig. 1 (p. 409) at levels A, B, C. x 3/10.

The surface enclosed by these ridges is concave as seen in trans-
verse sections of the orbital and narial regions (Text figure 3); a
transverse ridge connects the longitudinal ridges midway between the
orbits and nares (Plate 4) and a definite ridge crosses the concave area
between the anterior rims of the orbits. Less perceptible ridges
radiate from the parietal foramen, the most prominent forming an

SAWIN : CKANIAL ANATOMY OF ERYOPS MEGACEPHALUS 415

"X" which centers on the foramen and extends to the dorsal rims of
the orbits anteriorly and the lateral surfaces of the supratemporal
bones posteriorly. A low but sharply defined crest borders the post-
parietals posteriorly and continues with diminished height along the
tabulars. The bones of this dorsal region are in general thicker pos-
teriorly at the skull table and along the course of the ridges. They
are quite thin anteriorly, the central portions of the nasals and the
front tip of the internasofrontal being especially fragile.

The lateral portions of the skull roof slope downward on either side
from this horizontal region. From the anterior end of the jugal back
to the otic notch the angle of slope is about 50° (Text fig. 3, b and c).
Anteriorly the lateral wall slopes more gently as it approaches the
muzzle; posteriorly it curves mesially to meet the palatoquadrate
complex.

The bones of the lateral portion of the skull are involved in the for-
mation of the narial and orbital openings and in addition bear struc-
tural ridges anteriorly. The most prominent ridge (Plate 4) extends
from a point on the prefrontal element immediately in front of the
orbit forward and outward to a point about 4 cm. in back of the narial
opening. An elevation leads from the anterior portion of this ridge
to the margin of the skull over the widest portion of the maxillary. A
depression occurs in the lateral wall between the dorsal longitudinal
ridge and the ridge described above, and a prominent depressed region
is also present anteroventral to the orbits. The bone is quite thin
in these depressed regions but is thicker posteriorly and at the margins.

Rough pitted sculpture occurs over the dorsal surface of the skull.
These pits are quite small about the margin and are larger posteriorly
than anteriorly, the largest being situated in the jugal region. The
sculpture is, in some cases, oriented in relation to the ossification
centers of the elements (Plates 1, 4, 5, 6) much as has been noted
in a number of other fossil forms by Bystrow ('35).

A ventral view of the skull (Plate 2) shows the dentition, palatal
openings, and the general structure of the palate. Marginal teeth are
present on the premaxilla and maxilla and, in addition to these, pairs
of teeth are borne in "craters" on the pre vomers, palatines, and ecto-
pterygoids. Tiny denticles, not apparent in the figures, are to be
found over much of the palatal surface except on the parasphenoid.
In some small specimens, doubtfully referable to the species mega-
cephalus such denticles appear even on this element and are especially
numerous on that surface of the bone between the basipterygoid
processes.

416 bulletin: museum of comparative zoology

Three pairs of openings are seen in the palate. Anteriorly the inter-
nal narial openings are prominent. Posterior and medial to these are
situated the medium-sized interpterygoid vacuities. These are bounded
laterally by the pterygoids and in front by the prevomers, which
widely separate the pterygoids anteriorly. The parasphenoid provides
the median boundary between the two vacuities and probably joined
the prevomers anteriorly. The posterior ends of the prevomers and
the anterior end of the sphenethmoid are fractured in every specimen
examined and no evidence exists as to the manner of junction of these
elements. Posteriorly and laterally the adductor fenestrae are promi-
nent. These are bounded mesially by the pterygoids, anteriorly by
the ectopterygoids, and anterolateral^ and laterally by the quad-
ratojugals; the posterior boundary is formed by the quadrate. The
fenestra is narrow anteriorly and expands abruptly behind the tip
of the descending flange of the pterygoid.

The bones of the palate are more or less horizontally situated anterior
to the basipterygoid processes. Thick bone is found anteriorly under-
lying the surface in front of a line across the posterior rims of the
prevomerine tooth craters, and quite thick structural ridges extend
from the prevomerine craters anteriorly toward the marginal tooth
row and posteriorly along that portion of the prevomer which extends
between the palatine and pterygoid. The inner margin of the internal
narial opening is also quite stout. A shallow basin of thin bone (2 mm.)
is formed by the prevomers behind a ridge connecting the posterior
rims of the prevomerine tooth craters. Anteromesially the border of
the pterygoid is quite thin; it becomes thicker as it approaches the
basipterygoid process. Its descending flange (Plate 4, PT) is quite
thick. The articular surface of the quadrate is prominent in ventral
view. This bone is quite small and is overlapped both laterally and
mesially by dermal elements. The parasphenoid is the prominent
median element which underlies most of the ventral surface of the
braincase, probably reaching the prevomers anteriorly, and forming
a stout sutural connection with the pterygoids laterally.

Cartilage bone elements of the braincase visible in ventral view
(Plate 2) include the sphenethmoid anteriorly and the otic ossification
and the occipital bones posteriorly. The sphenethmoid is the most
massive element of the entire skull. Its anterolateral walls are recessed
and in continuity with channels leading from the cranial cavity. The
otic elements, each of which occupies the region usually taken by
pro- and opisthotics, are expanded laterally. These lateral expansions
extend to the skull roof beneath the tabulars posteriorly and the

sawin: cranial anatomy of eryops megacephalus 417

supratemporals anteriorly. The fenestra ovalis is immediately above
the posterodorsal suture of the parasphenoid. The occipital bones
are best described in posterior view.

If the superficial bones of the palate be removed (Plate 3) certain
hitherto undescribed features of the under side of the skull roof become
visible. These include the obverse of the ridges noted on the dorsal
surface of the skull and also bony emplacements for the cartilaginous
nasal capsules. The counterparts of the preorbital portions of the
dorsal longitudinal ridges appear as deep grooves leading forward
from the region anterior to the orbits to terminate anteriorly and
mesially at pronounced depressions which are situated behind the
septomaxillae. These grooves (Plate 3, TG) were probably occupied
by trabecular cartilages in life. The oval depressions anterolateral to
these grooves (NAC), doubtless housed the dorsal portions of the carti-
laginous nasal capsules. Each depression is bordered mesially and
posteriorly by a sharp rim. Posterolaterally the region presumably
occupied by the nasal capsule extended above the anterior portion
of the palatine tooth crater, being limited dorsally and laterally by
the maxillary. The ventral portion of the capsule probably rested on
the dorsolateral surface of the prevomer and the anterodorsal surface
of the palatine.

The counterpart of the ridge described above as extending from the
prefrontal to a point in back of the narial rim appears as a groove
(Plate 3, LG). This courses anterolateral^ to the posterior wall of
the capsular emplacement. The lacrimal duct probably occupied this
groove, since no channel can be found for it through the lacrimal bone.

A definite groove (Plate 3, AG) runs from a position immediately
posterior to the orbital rim to foramina in the jugal. These foramina
probably extend to enter a channel in the maxillary bone. This chan-
nel, which is not figured, is enclosed in the maxillary bone above and
mesial to the tooth row. Posteriorly it is continuous with a channel
in the quadratojugal which leads to the paraquadrate foramen.1
Anteriorly it continues in the premaxillary bone. Small foramina
which open from the maxilla into the postero-lateral portion of the
space assigned to the nasal capsule probably communicate with this
channel.

A posterior view is shown in Plate 6A. The foramen magnum is
continuous ventrally with a space which was probably occupied by
the notochord in life. These conjoined openings have the outline of
an arrowhead, the base of which was occupied by the notochord.

1 See description of quadratojugal, page 420.

418 bulletin: museum of comparative zoology

Dorsally and laterally are the post-temporal fenestrae, spaces bounded
by the tabulars and postparietals dorsally, and the exoceipitals
mesially. The paroccipital process of the otic ossification and the
tabulars limit the fenestra ventrally. The rather wide surfaces of the
tripartite occipital condyle are composed largely of the exoceipitals,
the basioccipital forming the ventral third of the articular area. The
stapes (Plate 6C) is only partly in view, extending outward and
upward from the ventral portion of the braincase to a position beneath
the tabular bone close to the otic ossification.

The mandible is narrow and high (Plate 5). It is 43 cm. in length,
the height is 10 cm. at the coronoid process and the width is 6 cm.
at the articular, these being the maximum dimensions. The larger
teeth are borne on the dentary bone, only tiny denticles being present
on the coronoids. Five apertures are visible in the jaw. The largest
of these is the meckelian fossa, which is bordered anteriorly by the
coronoid, laterally by the coronoid and surangular, posteriorly by the
articular, and mesially by the prearticular. The other openings are
on the mesial surface and include a mental foramen situated anteriorly
at the suture between the splenial and precoronoid bones, a small
mandibular foramen in the postsplenial, an inframeckelian fenestra
between the angular and prearticular, and a small dental foramen
located posteriorly in the prearticular. The sculpturing of the lateral
surface of the mandible consists of interrupted ridges and pits ante-
riorly, but resembles that of the skull roof posteriorly.

THE SEPARATE SKULL ELEMENTS

Dermal Bones of the Skull Roof

In the following sections the various bones of the skull will be con-
sidered individually. The dermal elements were for the most part
briefly but accurately described by Broom ('13); the present account is
essentially an amplification of Broom's as regards these bones. These
dermal elements may be divided into two groups, consisting of a
dorsal horizontal series, for the most part bounded laterally by the
dorsal longitudinal ridges, and a paired lateral series which form the
sloping lateral walls and the marginal elements.

The lateral series will be first discussed. Ten bones (Text fig. 1, p. 409,
Plates 1,4), the premaxillary, maxillary, septomaxillary, lacrimal, jugal,
quadratojugal, squamosal, postorbital, prefrontal, and nasal comprise
this series on either side. Of these the premaxillae, maxillae, and quad-
ratojugals also form portions of the palate while the postorbitals,

sawin: cranial anatomy of eryops megacephalus 419

prefrontals, and nasals extend on to the dorsal horizontal portion
of the skull.

The premaxillae form the muzzle and meet at a common suture
at the midline. Posterolaterally each forms the anterior border of the
external narial aperture and joins the maxillary beneath this opening.
Posteriorly the premaxillary joins the nasal, sending a sharp short
process over that bone mesial to the narial opening. That part of the
premaxillary which underlies the narial aperture forms a base for the
anterior portion of the septomaxillary. The latter (Plate 4, SMX)
is a thin bone which in part rests on the premaxillary and maxillary
bones at the ventral rim of the naris. It curves upward in back,
closely following the posterior half of the narial rim1 in such a manner
as to restrict the flow of air to the anterior half of the naris. This
bone extends mesially well into the narial opening, its basal area
(Plate 3) articulating with that portion of the prevomer directly
above the tooth "crater."

The maxillary is a long bone, forming most of the lateral margin
of the skull. Anteriorly it meets the premaxillary, forms the postero-
ventral border of the naris and is overlain by the septomaxillary. It
meets the lacrimal and jugal mesially and diminishes in size posteriorly,
tapering off to the margin after gaining contact with the quadrato jugal.

The lacrimal extends posteriorly from the naris two-thirds of the
distance to the orbit. It bears a definite ridge over most of its length,
the under side of which forms a groove for the lacrimal duct. Laterally
it is bounded by the maxillary and jugal, the sutural borders being
approximately equal in length. Mesially it meets the nasal and
posteromesially the prefrontal. Anteriorly it forms the posteroventral
rim of the naris. This bone is quite thin posteriorly at its junction
with the jugal and prefrontal.

The jugal is the largest bone of the skull roof. It forms the ventral
rim of the orbit, joins the prefrontal anterodorsally and the lacrimal
anteriorly. Ventrally it meets the maxillary in a long suture; it tapers
to a point anteriorly between that^bone and the lacrimal. Postero-
ventrally it joins the quadrato jugal and posteriorly the squamosal.
Posterodorsally it meets the postorbital at an irregular suture. The
jugal is quite thin in the preorbital region but becomes thicker ventrally
and posteriorly, where it is deeply sculptured.

The quadratojugal forms the posterolateral margin of the skull.
Dorsally it meets the squamosal and the jugal, the sutures being ir-

1 In other specimens, e.g., A.M.N.H. no. 4189, the septomaxillary actually meets the
posterior rim of the naris.

420 bulletin: museum of comparative zoology

regular. Anteriorly it is separated from the margin by the maxillary.
Posteriorly it is quite stout and strongly overlaps the ventrolateral
surface of the quadrate. Immediately anterior to this junction with
the quadrate the margin expands laterally and becomes somewhat
arched (Plate 4). A foramen, paraquadratum proprium,1 0.6 mm. in
diameter enters the inner surface of the quadratojugal just above the
stout portion which abuts against the quadrate. It communicates
with marginal channels of the skull as described above.

The squamosal is a complex bone. It meets the postorbital, the
supratemporal, and, to a slight extent, the tabular dorsally. Anteriorly
it joins the jugal and ventrally the quadratojugal. Posteroventrally
it strongly overlaps the quadrate. The posterodorsal portion of the
squamosal turns mesially at more than a right angle, and after an
interval of about two centimeters devoid of sculpture, meets the
pterygoid. This unsculptured area of the bone faces posteriorly, and
above curves posterodorsally to form a considerable portion of the
otic notch.

Internally (Plate 3) the squamosal sends a descending process
lateral to the quadrate flange of the pterygoid. Ventrally this descend-
ing process diverges noticeably from the quadrate ramus of the ptery-
goid, the space between these probably being filled by cartilage.

The postorbital is small, forms the posterior rim of the orbit and is
equally divided between the lateral wall and the skull table. The post-
frontal bounds it mesially, the supratemporal posteromesially and the
jugal ventrally, by an irregular suture. It is limited posteroventrally
by the squamosal.

The prefrontal forms the thick anterior rim of the orbit and extends
anteriorly to a point midway between the orbit and naris, the antero-
ventral portion of the bone being quite thin. It lies between the nasal
and lacrimal anteriorly, meets the frontal dorsally, and the post-
frontal above the orbit by a short suture. It joins the jugal ventrally.
The dorsal longitudinal ridge is prominent on this bone, dividing the
surface unequally into a small horizontal area which is on a plane
with the skull table, and a much larger area which slopes down with
the lateral wall. The contour of this lateral area is interrupted by the
posterior portion of the ridge which overlies the lacrimal groove
and joins the dorsal longitudinal ridge on the prefrontal just in front
of the orbit.

The nasal is large ; of the elements of the skull roof it is exceeded in
area only by the jugal. This bone extends almost three quarters of

1 Term employed by Bystrow ('39) in descriptions of Dwinosaurus and Benthosuchus.

sawin: cranial anatomy of eryops megacephalus 421

the distance between the naris and orbit, the most anterior portion
meeting the premaxillary slightly in front of a line across the anterior
borders of the nares. Anterolaterally it forms the dorsal border of
the naris, closely approaching or meeting the septomaxillary. It
meets its fellow at the midline for a distance of more than half its
length, this junction being interrupted posteriorly by the internaso-
frontal. In back, the nasal tapers between the prefrontal and frontal.
Laterally it joins the lacrimal. The dorsal longitudinal ridge divides
this bone into dorsal and lateral portions, the dorsal area consisting
of two-thirds of the surface of the bone.

The remaining area of the skull roof includes the dorsal and generally
horizontal portion between the dorsal longitudinal ridges. This may
be divided into a group of six paired bones and a median element.
Of the paired bones, the frontals, parietals, and postparietals form a
series which meets in the midline while the postfrontals, supratem-
porals and tabulars constitute a series posterolateral to these.

The median internasofrontal1 is situated between the nasals anteri-
orly and the frontals posteriorly. It is 7.5 cm. long and oval in shape.
The central and posterior portion of the bone overlies the anterior
portion of the sphenethmoid region of the braincase.

The frontal is larger than the internasofrontal. It meets its fellow
over the course of half its length posteriorly at the midline and extends
anteriorly around the posterior half of the internasofrontal to terminate
between that bone and the nasal. It meets the prefrontal laterally,
the postfrontal posterolaterally and the parietal posteriorly. The
frontals are convex between the orbits and bear a noticeable trans-
verse ridge which follows the contour of this convexity, this ridge
being situated on a line across the anterior rims of the orbits. These
bones overlie most of the more massive portion of the sphenethmoid
region of the braincase.

The parietal is much smaller than the frontal. It joins its fellow
at the midline over its entire length, a parietal foramen opening
posterocentrally between the two. This bone meets the frontal ante-
riorly, the postfrontal anterolaterally, the supratemporal postero-
laterally, and the postparietal posteriorly. The parietals form a portion
of the roof of the braincase and overlie the anterior portions of the
otic elements posteriorly and the posterior portions of the spehenth-
moid and laterosphenoid regions anteriorly.

The postparietals (or dermal supraoccipitals) form most of the
posterior rim of the skull table. These bones are ridged on their

1 This term as noted by Broili ('16) has priority over the term interfrontal.

422 bulletin: museum of comparative zoology

posterodorsal margins and turn down onto the occipital surface. On
the occiput (Plate 6A) each bone forms the mesial half of the dorsal
border of the post-temporal fenestra and just mesial to the fenestra
sends a short process over the exoccipital. A noticeable indentation
occurs in the bone mesial to this process and at the midline it meets
its fellow in a second ventral extension which reaches the dorsal
border of the foramen magnum. The postparietals join at the midline,
extend to the parietals anteriorly, the supratemporals anterolaterally,
and the tabulars posterolaterally.

The supratemporal is thick, roughly hexagonal in shape and overlies
the otic and in part the supraoccipital region of the exoccipitals. It
joins the postfrontal and postorbital anteriorly, the parietal mesially,
the squamosal laterally, the tabular posteriorly, and the postparietal
posteromesially. This bone is elevated laterally, forming a portion
of the dorsal longitudinal ridge. Another ridge which is smaller
courses centrally from this toward the parietal foramen.

The tabular occupies the posterior corner of the skull table and is
prominent in occipital view (Plate 6A) where it forms the lateral
border of the post-temporal fossa, joining the postparietal above the
fossa and the otic element below. In dorsal view (Text fig. I,1 Plate 1)
it is seen as an element which meets the postparietal mesially, the
supratemporal anteriorly, and to a slight degree the squamosal antero-
laterally. It extends freely posterolaterally to form a "horn" which is
deeply sculptured and somewhat compressed laterally. The antero-
ventral portion of this bone bears a definite surface which presumably
forms an abutment for the dorsal process of the stapes (Plate 3, S TPD).
An unossified gap exists between this region of the tabular and the
dorsal border of the otic in M. C. Z. 1129. In larger and better ossified
skulls this region is completely ossified and the sutural relations are
not clear.

Dermal Bones of the Palate

(Plate 2)

The premaxillary carries the anterior portion of the marginal tooth
row and has in addition a noticeable palatal exposure which extends
posteriorly from the tooth row to the prevomers. A definite elevated
area with a rough surface occurs at the common suture of the pre-
maxillae behind the tooth row; the function of this rugose area is
unknown. This bone is ridged laterally in front of the prevomerine
tooth crater. Posterolaterally the premaxillary meets the maxillary
just behind the anterior border of the internal naris. Each bone accom-

1 See page 409.

sawin: cranial anatomy of eryops megacephalus 423

modates thirteen teeth, very nearly half this number being present
and functional. These teeth, ovoid at the base and conical in shape,
alternate quite regularly with unoccupied pits. Their distribution
and relative sizes are shown in Plates 2 and 4.

The maxillary extends from the premaxillary diminishing in size
in back. Anteriorly the maxillary forms the lateral border of the inter-
nal naris and mesially it meets the palatine, ectopterygoid, and a small
anterior extension of the quadratojugal. Three small foramina are
present in this bone along the suture with the palatine behind the
posterior rim of the internal naris. These foramina (not shown in the
figures) evidently communicate with the channel in the maxillary
bone described in the general section above. Accommodations for 36
teeth appear on the left maxillary and 38 on the right. Two-thirds
of the possible number of teeth are visible on each bone, these occurring
in three or four groups interrupted by a series of alternating teeth
and pits. It is probable that teeth were replaced in "waves" along
the maxillae rather than by simple alternation as in the case of the
premaxillae. As will be noted below, the dentition of the dentary
correlates well with this mode of succession, the teeth underlying the
premaxillary alternating regularly in the dentary, while the teeth
underlying the maxillary correspond closely to those of that bone in
their distribution.

Two groups of larger teeth are present in the upper jaw; counting
both teeth and pits from the midline, it is seen that tooth number 10
on each side marks the center of a group of larger teeth on the pre-
maxillae, while on the maxillae, tooth number 19 or 20 marks the center
of another group of large teeth on each side.

The remainder of the palatal elements, except for the parasphenoid,
which will be described later with the braincase, may be divided into
two series of bones. The first consists of dermal elements and includes
the prevomers, palatines, ectopterygoids, and pterygoids. The second
series, including the remnants of the primary palatoquadrate arch,
the epipterygoids, and the quadrates, will be considered in a separate
section. This complex of cartilage and dermal bones is firmly joined
to the rest of the skull.

The prevomers constitute most of the central palatal area anterior
to the tip of the sphenethmoid. These, with the premaxillae, form a
horizontal shelf, which extends back to a line across the posterior
rims of the prevomerine tooth craters. This shelf is bounded on either
side by a longitudinal elevation which is surmounted by the tooth
crater. This elevation extends posterolateral^, following that portion

424 bulletin: museum of comparative zoology

of the prevomer which is between the palatine and ectopterygoid.
The prevomer is quite stout laterally where it forms the rim of the
internal naris and is also thick in the region of the "shelf" described
above. Posteriorly and centrally the prevomers are thin and form a
shallow basin which is convex dorsally. Each bone meets its fellow
in the midline over most of its length but separates posteriorly and
tapers into a process which leads back toward the parasphenoid. The
posterior edge of each process of the prevomers is fractured in M. C. Z.
1129, as is the anteriormost edge of the parasphenoid. The manner of
connection of these elements is a matter of conjecture; a possible
restoration is shown in dotted lines in Plate 2.

The palatine borders the internal naris posteriorly. It bears the
largest tooth crater of the palate near its lateral border just posterior
to the internal naris. This bone is solidly articulated with the maxillary
beneath the posterior portion of the tooth crater and is quite stout
at the rim of the naris. Above and in front the bone is excavated to
accommodate a portion of the cartilaginous nasal capsule, as previously
described. Laterally the palatine joins the maxillary, mesially the
prevomer, and posteriorly it extends in a narrowing process which
turns mesially and ends between the ectopterygoid and pterygoid.

The ectopterygoid is a narrow bone posterior to the palatine between
the maxillary and quadratojugal laterally and the pterygoid mesially.
A large crater occupied by a tooth and pit is situated anteriorly just
inside the suture with the maxillary. The bone is thick beneath this
crater, and is also stout along its suture with the maxillary and in
back where it joins the pterygoid mesially and the quadratojugal
laterally to form the anterior border of the adductor fenestra.

The pterygoid is the largest and most complex bone of the palate.
For descriptive purposes it may be divided into four regions:

1. A proximal portion consisting of the basipterygoid articulation
and that part of the bone dorsal to it which is transversely oriented.1

2. A palatal ramus which is generally horizontal and extends ante-
riorly to meet the dermal palatal elements.

3. A posterior quadrate ramus which is more or less vertically
oriented.

4. A descending flange which constitutes the ventralmost portion
of the pterygoid.

These arbitrary divisions will be considered in the order given.
The basipterygoid articulation of the proximal region is in part com-
posed of a substantial sutural connection of the pterygoid with the

1 "Lamina postquadrata" of Sushkin ('27).

sawin: cranial anatomy of eryops megacephalus 425

ventrolateral portion of the parasphenoid.1 This connection is very
firm, as is shown by the interdigitating articular surfaces (Plate 10A)
of the parasphenoid. Above this articulation the pterygoid forms a
cup (Text fig. 3C, Plate 3) which accommodates the distal end of the
basipterygoid process of the basisphenoid. Dorsal to the basipterygoid
region and extending laterally, the pterygoid forms a transverse
flange, the anterior face of which is mainly occupied by the epiptery-
goid (Plate 9ab.) This transverse flange extends to the skull roof,
which it either approaches closely or meets beneath the supratemporal-
squamosal suture (Text fig. 3C). It descends mesially from this point
to follow the contour of the expanded portion of the epipterygoid.
Laterally it makes an abrupt turn posteriorly and becomes continuous
with the quadrate flange.

The horizontal palatal ramus 2 extends forward from the proximal
portion to meet the ectopterygoid, palatine, and prevomer. It under-
laps the prevomer, tapering to a point over the lateral ridge of that
bone. The mesial edge of the palatal ramus is thick proximally, but
distally becomes much thinner and is quite thin and fragile ante-
riorly. The base of the expanded portion of the epipterygoid overlaps
the dorsal surface of the proximal portion of this process.

The quadrate ramus extends from the lateral portion of the trans-
verse region to the squamosal and quadrate. This ramus meets the
surface of the descending process of the squamosal mesially and over-
laps that bone to a considerable extent posteriorly (Plates 1, 3, 9).
Below, the pterygoid curves ventrally and laterally to form the mesial
border of the adductor fenestra. Posteriorly it overlaps the postero-
mesial surface of the quadrate.

That region of the pterygoid which projects downward and forms
the anteromesial border of the adductor fenestra is the descending
flange (Plate 4b, PT). This flange is thick, and dorsally provided the
surface of origin for an adductor muscle of the jaw.

The Primary Palatoquadrate Arch

(Plates 6, 9)

The ossifications of the primary palatoquadrate arch include the
quadrate and epipterygoid. The latter is leaf-shaped ventrally and
produces a free dorsal "rod" which extends up to abut against the

1 The suture between the parasphenoid and pterygoid is clearly denned on both sides in
the specimen. Cf., Dempster ('35, pp. 176-177).

2 A groove on the dorsal surface of the palatal ramus may be related to a cartilaginous
palatal process.

426 bulletin: museum of comparative zoology

parietal immediately lateral to the side wall of the braincase slightly
behind a line across the parietal foramen. A large part of the leaf-like
expanded region of the epipterygoid is firmly applied to the trans-
verse portion of the pterygoid, but dorsomesially that portion which is
beneath the rod is thicker than the rest and articulates with the basis-
phenoid, fitting into a depression on the anterolateral face of that bone
(Plate 10a). Posterolateral to the rod the dorsal margin of the epi-
pterygoid (Plate 9) curves upward on to the dorsal portion of the
pterygoid. There it loses its finished surface and closely approaches
an unfinished area on the upper portion of the otic ossification. This
dorsal region of the leaf-like portion is in continuity with the otic
ossification in older specimens, and has been termed the prootic
process of the epipterygoid by Sushkin ('27). In the specimen here
described, the prootic process was evidently completed in cartilage.
The region of the epipterygoid lateral to the prootic process is quite
thin and bears unfinished surfaces at its dorsal and lateral margins.
It is possible that these surfaces were joined by cartilage which ex-
tended to the quadrate — if one assumes that the gap between the
descending process of the squamosal and the pterygoid was occupied
by cartilage. There is no certain indication of a cartilaginous palatal
process,1 the anteroventral edge of the epipterygoid bearing a smooth
sutural relationship with the pterygoid beneath.

The quadrate (Plates 3, 6, 9) is a wedge-shaped bone, the lateral
and mesial surfaces of which are covered by overlapping dermal ele-
ments. In posterior view (Plate 9) it is only partly exposed and tapers
dorsally between the pterygoid and squamosal. A prominent tubercle2
occurs ventrolaterally on this surface. Dorsolaterally the quadrate is
overlapped by the squamosal and ventrolaterally it meets the quadra-
te jugal, which forms a strong abutment against it, enclosing its
anteroventral margin. Posteromesially it is very widely overlapped by
the pterygoid. This considerable overlap is apparent if the posterior
exposure (Plate 6A) of the quadrate be compared to the anterior ex-
posure (Plate 9a). In the latter view the quadrate is seen to extend
mesially. The proximal edges of this bone which overlie the pterygoid
are unfinished as is the surface of the pterygoid. This rugose surface of
the pterygoid extends dorsally, approaching the region of the descend-
ing flange of the squamosal. As was mentioned above, the space be-
tween the descending flange of the squamosal and the pterygoid was
originally occupied by cartilage which may have extended to the lateral

1 A shallow groove on the dorsal surface of the proximal portion of the palatal process of
the pterygoid is probably a structural feature rather than an emplacement^ for cartilage.

2 This is identified as the "tuberculum supratrochleare" by Bystrow ('39).

SAWIN : CEANIAL ANATOMY OF ERYOPS MEGACEPHALUS 427

and dorsal surface of the expanded portion of the epipterygoid. If
cartilage existed over the rough surfaces of the pterygoid and quadrate,
it is possible that the quadrate and epipterygoid were more or less com-
pletely connected by cartilage.

The ventral surface of the quadrate forms the articular area for the
mandible. Its form is shown in Plate 3. This surface is rugose, prob-
ably having been finished in cartilage.

The Mandible
(Plate 5)

Ten bones form the mandible including the dentary, splenial, post-
splenial, angular, surangular, prearticular, articular, and a series of
three coronoid elements. The dentary is the largest of these, extending
back from the symphysis over five-sixths of the length of the jaw. It
forms most of the symphysis with the exception of a small ventral
region and appears on the mesial surface of the jaw anterior to the
mental foramen. Most of the lateral surface of the mandible beneath
the tooth row is composed of this bone. It meets the splenial ante-
roventrally, joins the postsplenial by a long suture ventrally and slopes
upward in contact with the angular to taper to a point far back on the
surangular. Medially it is bordered by the coronoids.

Accommodations for 47 teeth appear on the dorsal surface of the
dentary and an additional pair of small teeth are situated behind the
regular tooth row near the symphysis. The teeth alternate with pits in
a manner similar to that seen in the upper jaw, those beneath the pre-
maxillary alternating regularly, and the teeth beneath the maxillary
appearing in groups which compare closely in distribution with those
on that bone above. Not conformable, however, are the groups when
compared as to size. They occur as follows in the dentary, if the teeth
be numbered from front to back: numbers 1-4 are large; 11-14 are the
largest of the jaw, and teeth 25 and 26 represent the center of a third
group which is distinctly larger in size than the others in that region
of the jaw. When matched with those of the upper jaw it is seen that
the larger groups of teeth of the dentary alternate with those above,
roughly occurring in front of, between, and behind the two larger
tooth groups on the premaxillary and maxillary.

The dorsal surface of the dentary includes a narrow, horizontal,
shallow trough inside the tooth row. This is possibly the site of origin
of the replacement teeth. Mesially the coronoid series forms the

428

bulletin: museum of comparative zoology

SPP

ANG

Fig. 4. Eryops megacephalus. Sections through mandible, x 1/2. Mandi-
ble x 3/20.

SAWIN : CRANIAL ANATOMY OF ERYOPS MEGACEPHALUS 429

boundary to this trough, extending above the border of the dentary to
form a rim, as seen in sections (Text fig. 4). This trough becomes
narrower and disappears posterior to the tooth row. The rim formed by
the coronoids is interrupted at points beneath the large teeth which
occur in the tooth craters of the palatine and ectopterygoid bones.

The splenial is situated at the ventral margin of the anterior portion
of the jaw. Laterally it is barely visible beneath the dentary; antero-
mesially it forms a small part of the symphysis and is in contact with
the dentary for a short distance at the inner surface of the jaw. It is
most largely exposed on the mesial surface of the jaw where it is
bounded by the precoronoid dorsally. A small mental foramen is
located anteriorly at the junction of the sutures between precoronoid,
dentary, and splenial. This is in communication with a mesial diver-
ticulum of the meckelian canal.

Also on the ventral margin of the jaw is the long postsplenial, lying
between the dentary on the lateral surface and the prearticular and
precoronoid mesially. It is exposed only slightly in lateral view but
has a large exposure in medial view. Anteriorly it meets the splenial
and posteriorly is in contact with the angular along an irregular suture.
A small mandibular foramen in communication with the meckelian
canal occurs anteriorly on its mesial surface. This bone forms the
floor of the meckelian canal over most of its length.

The angular constitutes the entire posteroventral area of the jaw as
seen laterally, but is restricted to the ventral portion of that region of
the mesial side. It meets the postsplenial anteriorly and the sur-
angular and to a slight degree the articular posteriorly. Antero-
dorsally it joins the dentary on the outer side of the jaw and postero-
dorsally the surangular. On the mesial side it is limited dorsally by
the prearticular and forms the ventral rim of the inframeckelian fossa.
Internally it floors the posterior part of the meckelian canal and forms
the ventrolateral wall of the meckelian fossa.

The surangular is mainly exposed on the lateral surface of the jaw.
It joins the coronoid dorsally with ah irregular suture, is overlapped
below this by a long tapering process of the dentary and extends anter-
iorly between that bone and the angular. Behind, it overlaps the pos-
terior portion of the articular and forms a strong abutment on the
anterolateral surface of that bone, extending over a considerable por-
tion of its surface inside the meckelian fossa. The rim of the surangu-
lar, anterior to the articular, bounds the meckelian fossa laterally, and
most of the dorsolateral wall of the fossa is made up of this bone. No
retroarticular process is to be found on the jaw but small ridges occur

430 bulletin: museum of comparative zoology

on the posterodorsomesial surface of the surangular, which are possibly
scars related to the insertion of the depressor muscle.

The prearticular is the largest bone on the mesial surface of the jaw,
extending from the precoronoid to the articular. It is limited ante-
rodorsally by the coronoid series, tapering anteriorly between the pre-
coronoid and postsplenial. Its free thin dorsal surface forms the mesial
rim of the meckelian fossa, the bone itself bounding the fossa mesially.
Posteriorly it overlaps the mesial surface of the articular. The angular
and postsplenial constitute its ventral boundary. A small dental fora-
men occurs posteriorly in this bone, but cannot be traced to the inside
of the jaw. Ventrally, near the anterior end of the angular, this bone
forms the dorsal rim of the inframeckelian fossa.

The articular is a small wedge-shaped bone considerably overlapped
by the dermal jaw elements and exposed above in an elongately oval
articular area, the form of which is shown in Plate 5. In front, this
bone forms the posterior wall of the meckelian fossa, and extends
downward between the surangular and prearticular probably to rest
anteriorly on the angular. This anterior process of the bone bears an
unfinished surface distally; this was evidently continued in cartilage
which probably extended anteriorly into the meckelian canal as a
remnant of the meckelian cartilage. On the mesial surface of the jaw it
is slightly exposed between the prearticular and surangular and barely
meets an attenuated process of the angular below (Plate 5, fig. E).

The three coronoids form the dorsal portion of the inner surface of
the mandible. The distal two of these are quite thin, and are applied
largely to the dentary which bone entirely underlies the pre-and inter-
coronoid as well as the anterior portion of the coronoid. The last is the
posteriormost and largest of the series. It produces a flange which
provides a part of the surface for insertion of the adductor muscles.
The coronoid meets the dentary laterally and forms an interdigitating
connection with the surangular posteriorly. It bounds the anterior por-
tion of the meckelian fossa and extends anteriorly, sloping mesially to
the inner surface of the jaw. Anteriorly, it meets the intercoronoid
and ventrally it is bounded by the prearticular. Small denticles occur
on its anteromesial surface.

The pre- and intercoronoids are vertically placed, the latter being the
smallest bone of the jaw. It lies above the prearticular and the poste-
rior third of the precoronoid and meets the coronoid behind. The pre-
coronoid extends from a point on the dentary behind the symphysis
to the prearticular posteriorly, tapering between that bone and the
intercoronoid and meeting the splenial and postsplenial ventrally. A

sawin: cranial anatomy of eryops megacephalus 431

fragment is missing at its posterior boundary in the specimen which is
figured. Denticles were missing on the two anterior coronoids in this
specimen, but it is probable that these were removed in the course of
preparation. In other specimens they are present on the exposed sur-
faces of these bones.

The general shape and extent of the meckelian canal is shown by
means of sections in Text fig. 4. Two channels are present in the ante-
rior end of the jaw, one of irregular shape leading toward the symphysis
and ending blindly, the other leading mesially to the mental foramen.

The Braincase

For the most part the neural cranium of Eryops is formed of endo-
chondral bones which are neither completely nor clearly distinguished
by sutures. These elements may conveniently be divided into anterior
and posterior groups. The anterior division includes a massive sphen-
ethmoid region in front, a basisphenoid region posteriorly, and a later-
osphenoid region laterally. The posterior division includes a single
otic ossification, here termed the otic, a basioccipital and the exoccipi-
tals.

In sagittal view (Plate 10b) it is apparent that the internal portion of
the braincase is unossified centrally between the basisphenoid region
and the basioccipital, and laterally there is a large gap between the otic
ossification and basisphenoid ventrally; the principal junction of the
latter elements occurs laterally (Plate 10a). The unossified regions
were evidently completed in cartilage, but suggest a fundamental
point of division of the braincase into anterior and posterior portions
as noted above. There is also a small unossified gap in front of the sella
turcica (SET, Plates 10b, 7B) which incompletely separates the basi-
sphenoid from the sphenethmoid. Otherwise the sphenethmoid, basi-
sphenoid, and laterosphenoid regions are essentially a unit with only
indefinite indications of other lines of demarcation existing between
them.

The posterior portion of the braincase is externally divisible into
occipital and otic ossifications. The otic ossification, which occupies the
regions usually formed by pro- and opisthotics in tetrapods, is exter-
nally limited from the basisphenoid region in front and the occipital
region behind by sutures (Plates 6B, 9a). The basioccipital is also a
distinct element, but the exoccipitals are unusual in meeting at the
midline above the foramen magnum where they apparently occupy the
position usually taken by the supraoccipital. No sutures can be seen
between these bones within the braincase.

432 bulletin: museum of comparative zoology

A third and minor division of the braincase consists of dermal bones.
Ventrally the parasphenoid enters into such intimate relationships
with the endochondral bones above that it is best described with the
braincase. Dorsally the parietals form a considerable portion of the
cranial roof.

The stapes, although not properly a braincase element is also con-
sidered in this section.

1. Anterior Division. The sphenethmoid region (SE, Plates 2, 3,
7AB, 8, 10) of the anterior division is arbitrarily defined as the
massive anterior portion of the braincase which ventrally (Plate 10b)
extends back to an unossified region in front of the sella turcica, and
dorsolateral^ it may be considered to terminate in thin lateral walls
at the region immediately behind the foramen for the optic nerve
(N II, Plates 10a, 7B). It is to be noted that no suture is visible be-
tween the sphenethmoid and the posterior regions in lateral view, only
the deficiency in the floor of the braincase serving as a possible indica-
tion of a boundary between the sphenethmoid and basisphenoid. The
sphenethmoid, as a region rather than a distinct element, underlies a
large portion of the internasofrontal and frontal bones and overlies the
anterior portion of the parasphenoid. It flares out laterally in front of
the orbits (Plate 3) then tapers anteriorly to end in an unfinished sur-
face. An elongate anterolateral recess is present on either side of this
tapering portion (Plate 10a). This recess is in communication with
channels leading internally through the bone into the cranial cavity.
Internally (Plate 10b) the posterior portion of the sphenethmoid
houses the cerebral region of the brain and anterior to this, three paired
canals lead to the elongate recess mentioned above. The relations of
these are seen to best advantage in a horizontal section of the braincase
(Plates 7B, 8b). The ventralmost pair of canals is smallest and prob-
ably conducted the vomeronasal nerve, as suggested by Dempster
('35). Dorsal and lateral to this pair another and larger pair diverge.
These probably carried the olfactory nerve. The dorsal and lateralmost
pair of canals, which possibly conducted blood vessels, slant down
anteriorly to join the olfactory canals as shown in the figures. The
general shape of the cerebral cavity and the relations of the canals as
seen in the sectioned specimen are shown in Text figure 5, G-N. The
opening for the optic nerve leads from the floor of the cranial cavity in
front of the unossified portion of the braincase between the spheneth-
moid and basisphenoid regions (N II, Plate 10b).

In all of the unsectioned specimens the dorsal wall of the spheneth-
moid was observed to end about 1 cm. anterior to the pineal foramen

sawin: cranial anatomy of eryops megacephalus

433

(Plate 10b). M. C. Z. No. 1458 as reconstructed shows a posterodorsal
extension, possibly of the sphenethmoid, which forms a chamber iso-

Fig. 5. Eryops megacephalus. Transverse sections through anterior portion
of braincase. See Fig. 4 for position of sections, x 1/2.

lating the region beneath the parietal eye (Text fig. 2).1 It is probable,
however, that this represents a portion of the lateral wall which be-

'See page 413.

434 bulletin: museum of comparative zoology

came displaced, and hence this is omitted in the rest of the figures
(ef. Dempster, '35, pp. 182-183).

Thin lateral walls extend posteriorly from the sphenethmoid region,
in back of the foramen for the optic nerve, to the otic ossification and
the basisphenoid region. A large part of these walls is missing in the
specimens examined and has been restored as indicated by dotted
lines in Plates 10 and 7. Its sutural relations with both the otic ossi-
fication and the basisphenoid regions are not at all certain and hence
have been indicated in dotted lines in the figures. Above, (Plate 8b)
the lateral walls may be traced from the sphenethmoid to the otics
and are firmly joined to the parietals dorsally. Since this thin lateral
region is situated in part in the morphological position of the latero-
sphenoid of archosaurs it is designated as the laterosphenoid region.
Its ventral relationships are uncertain. Posteriorly it apparently
joined the dorsum sellae of the basisphenoid, and anterior to this the
lateral walls of the anterior portion of the sella turcica (Plates 7 and
9a). At best it can be indicated as a region which is very doubtfully
separable from the basisphenoid and is certainly continuous with the
sphenethmoid. The restoration of foramina for nerves III and IV
is purely arbitrary, but nerve VI has been observed in one specimen
leading into a small channel close to the junction of the thin latero-
sphenoid region with the basisphenoid about 0.8 cm. from the midline
(Plate 9a). It is a matter of opinion as to whether the abducens nerve
penetrates the dorsum sellae of the basisphenoid region or the latero-
sphenoid region.

The basisphenoid region is in part distinct internally from the rest
of the braincase (Plate 10b), being bounded anteriorly in front of the
sella turcica by an unossified area and posteriorly by a gap between it
and the basioccipital. Anterolateral^ there is a suture between the
otic and the basisphenoid region (Plate 9a), but the relations of this
bone with the lateral walls are obscure as noted above. The basi-
sphenoid lies dorsal to the parasphenoid and produces basipterygoid
processes laterally which, together with the processes of the para-
sphenoid, form an articulation with the pterygoid and the epiptery-
goid. The distal end of the basipterygoid process of the basiphenoid
is cartilaginous (Plate 3). Above the process, anterolateral^ (Plate
10a), the surface of the basisphenoid is depressed for the articulation
of the epipterygoid. This relation is diagrammatically shown in sec-
tion in Text fig. 4B. Anteriorly, just behind the sphenethmoid, the
dorsal surface of the basisphenoid bears a definite sella turcica (SET,
Plate 7) which is pear-shaped in outline, 0.9 cm. wide at the widest

sawin: cranial anatomy of eryops megacephalus

435

portion, 1.2 cm. long and 0.35 cm. deep anteriorly, becoming shallower
posteriorly. Openings for the internal carotid arteries penetrate the
floor of the sella anterolaterally. The anterolateral margins have frac-
tured surfaces which were probably continuous with the laterosphe-

C DEF ,

J K L M N

FEN.OV.

Fig. 6. Eryops megacephalus. Transverse sections through braincase.
A. Otic region. B. Basipterygoid region, x 1/2.

noid region, but the rim of the posterior portion appears to be finished
bone. A thin dorsum sellae, V-shaped in section, extends above and
slightly behind the posterior portion of the sella. The dorsum sellae
is very fragile and its relations with the lateral walls are not definite.
This region is restored in the figures, no clear evidence existing as to
its extent laterally.

The thick central surface of the basisphenoid posterior to the sella
is V-shaped in section (Text fig. 6). Posteriorly the bone slopes

436 bulletin: museum of comparative zoology

abruptly to the floor of the braincase immediately in front of the basi-
occipital. The posterior surfaces of the bone are rough and were evi-
dently continued in cartillage.

Posterolateral to the sella and beneath the dorsum sellae are paired
recesses1 (REM, Plates 7, 9a) which probably provided the surfaces
of origin for the rectus muscles of the eye. The ventral portion of the
dorsum sellae incompletely separates these recesses and leaves an
opening for the hypophysial vein ventrally.

Channels for the internal carotid arteries enter foramina in the basi-
sphenoid anterolateral^ (FCA, Plate 10a). These arteries occupied
deep grooves in the parasphenoid which lead toward the foramina on
either side (Plate 8a). Within the basisphenoid each carotid leads
up to the sella turcica, there presumably sending a branch into the
cranial cavity. According to data from the reconstructed specimen,
the artery then proceeds under the anterior end of the sella where it
again branches, one branch entering the cranial cavity near the fora-
men for the optic nerve, the other member turning mesially and
anastomosing with the corresponding vessel of the other side. Paired
nutrient branches appear to lead into the sphenethmoid from this
anastomosis.

The above description of the anterior portion of the braincase is in
several respects not in agreement with accounts given by previous
workers. I agree in part with Dempster's ('35) general analysis of this
region of the braincase. He states (p. 177), "The heavily ossified
sphenethmoid of Eryops is an uncommon feature among amphibians.
This bone together with the lateral sphenoids and the basisphenoid,
which are fused with it, forms a complete osseous envelope for the mid-
brain and the more anterior brain parts."

It is evident, however, in the young adult specimen which forms
the basis of my descriptions, that the basisphenoid is to be distin-
guished in part from the sphenethmoid by the ventral unossified
region. Otherwise the elements may be distinguished only as regions,
and as a more or less united series may be regarded as being compa-
rable with the greater part of the ethmosphenoid of Crossopterygians.
The possibility remains, however, that more complete and somewhat
younger specimens will reveal sutures in this anter or region. This
possibility renders a description by regions desirable until it be defi-
nitely proved either that the anterior portion of the braincase is a
single ossification or that it represents a fusion of the several elements.

1 "Myodome' ' of Dempster ('35).

sawin: cranial anatomy of eryops megacephalus 437

The fragmentary laterosphenoid region and the usual poor preser-
vation of the anterior part of the basisphenoid has led to highly diver-
gent accounts of this portion of the cranium. Broom ('13, fig. 14)
figured the sella in the proper position. Williston ('18) also figured it
correctly but his descriptions are somewhat ambiguous. Watson
('16, fig. 1), however, assigned the pituitary gland to the portion of
the basisphenoid which appears to be in the position of the recesses
for the eye muscles, an interpretation readily understandable because
of the frequent loss of the thin floor of the braincase above the recesses.
Dempster ('35) placed the gland in an even more posterior position
in the unossified region between the basisphenoid and basioccipital.
The latter two accounts are not to be reconciled with the evidence
gained from the excellently preserved material which it was my good
fortune to investigate. The evidence from the sectioned specimens
is definite as to the presence of six canals in the sphenethmoid. This
differs from the accounts of Broom ('13) and Dempster ('35) who
describe four canals but appear to neglect the ventralmost canal here
shown in Plate 7B.

2. Posterior Division. A single otic ossification on either side houses
most of the inner ear structures. This element is not divisible into
prootic and opisthotic regions by sutures in either the entire or the
sectioned specimens. This is in general agreement with the findings
of others. This bone overlies the basisphenoid anteriorly and the
exoccipital posteriorly. It is underlapped ventromedially by the para-
sphenoid. Anteriorly it meets the laterosphenoid region, forming
with this the large prootic foramen. Dorsolaterally the otic expands
and abuts against the tabular and supratemporal elements.

An anterior view (Plate 9a) shows the prootic foramen (A7 V) and
behind and lateral to it a smaller foramen1 which is the anterior open-
ing of a channel communicating with the posttemporal fossa. The otic
bears an unfinished surface above this foramen. This was doubtless
completed in cartilage which was in continuity with the prootic process
of the epipterygoid as described above. Ventrally there is a suture
between the otic and basisphenoid.

In lateral view (Plate 10a) two openings are to be seen. The most
evident of these is the large fenestra ovalis which was occupied in life
by the footplate of the stapes. The ventral border of the fenestra is
formed by edges of the parasphenoid which, in other specimens de-
scribed by Watson ('16) and Sushkin ('27), is suturally connected to

1 This foramen was assigned to the facial nerve by Sushkin ('27, fig. 6). This is obviously
erroneous.

438 bulletin: museum of comparative zoology

the base of the stapes. A small foramen occurs about 1 cm. in front
of this fenestra on a level with its ventral rim. This is probably an
opening for the hyomandibular branch of N. VII. In this view are
visible the sutures between the otic and parasphenoid below and be-
tween the otic and exoccipital posteriorly. The latter suture extends
through the foramen for N. X. These sutures are indistinct or incom-
plete in most specimens.

Posteriorly (Plate 6B) a suture is visible between the otic and exoc-
cipital. This fades out dorsally in the mesial surface of the post-
temporal fossa. Ventrally it extends to the vagal foramen and be-
comes continuous with the suture described in lateral view.

Posterodorsally (Plate 7A) the otic forms the ventral and lateral
surfaces of the posttemporal fossa. More anteriorly it completely sur-
rounds the channel leading from this fossa to the foramen on the
anterior surface of the otic (V, Plate 9a). A thin plate of bone which
is presumably a lateral extension of the exoccipital overlies the fossa
and joins the dorsal margin of the otic. No suture is to be seen between
these bones dorsally. The postero-dorso-lateral surface of the otic is
rough, indicating the presence of cartilage between that region of the
otic and the overlying tabular.

Internally (Plates 7, 8, 10) no sutures are visible to distinguish the
otic from the occipital. A large aperture beneath and posterior to the
ventral margin of the prootic foramen leads from the endocranial cav-
ity to the vestibular region of the inner ear (Plate 7B). In hori-
zontal section the prootic foramen (AT V, Plates 7B, 8b), is seen as a
rather extensive channel, the proximal portion of which extends well
into the braincase. This foramen is directed outward at a 45° angle to
the longitudinal axis. Ventrally the thin walls of this channel overlie an
unossified space which is anterior to the vestibular recess and in com-
munication with it. Dorsally the wall of this foramen, — the parietals
above, and the laterosphenoid region laterally, — bound a small cranial
recess {R, Plates 8b, 10b) which was possibly occupied by a portion of
the cranial endolymphatic system.1 This recess leads ventrally to a
groove which is posterodorsal to the prootic foramen and slopes ven-
trally to the opening between the braincase and the inner ear region.
The walls of the otic which overlie the basisphenoid are thin (4 mm.).
Apparently no portion of the otic forms a floor to the braincase, the
parasphenoid being exposed ventrally.

'Watson ('16) and Dempster ('35) find a foramen for the endolymphatic duct in the cranial
wall. This may well exist in better ossified specimens in which the cranial wall is more com-
plete. There is no confirmation in the material available to me for the endocranial sutures
described by Watson ('16). (Cf. Williston, '18; Dempster, '35.)

sawin: cranial anatomy of eryops megacephalus 439

The vestibular apparatus1 of the inner ear was housed ventrally in
the otic in the space designated as the otic recess (ROT, Plate 7B) but
the rough surface of the bone in this region defines no specific structural
areas, that portion of the inner ear evidently having been completely
housed in cartilage. This space is continuous ventrally with the large
cavity beneath the rim of the prootic foramen described above. The
larger portion of this anterior space, with the exception of channels for
the facial nerve, was probably also occupied by cartilage.

A nerve, probably the hyomandibular branch of N. VII, made its
exit through the wall of the otic in the most lateral portion of this region
(N. VII, Plate 10b) and a depression in the otic lateral and ventral to
this may indicate the point of departure for the palatine branch of
N. VII although the foramen cannot be traced through the bone. The
relations of this anterior cavity are best shown by the brain cast which
will be discussed in a later section.

Dorsally (Plate 8b) the positions occupied by the anterior vertical
and posterior vertical semicircular canals are seen as grooves in a
cancellous bone with an unfinished surface. There is no definite in-
dication of a horizontal canal ; that portion of the labyrinth evidently
was lodged either in cartilage or an extremely cancellous bone which
was not preserved.

The exoccipitals form the lateral portions of the occipital condyle
and extend dorsally on either side of the foramen magnum to join
above this opening. There is an external suture between the exocci-
pital and otic as noted above in the description of the otic. Ventrally
and posteriorly (Plates 8a, 6B) there is a clear suture with the basioc-
cipital. The anteroventral surface of this bone is overlapped by the
parasphenoid. A dorsal view of the exoccipitals (Plate 7 A) shows that
these bones meet above the cranial cavity to occupy the region usually
formed by the supraoccipital. The sectioned specimen shows no
median suture nor any structure which may be interpreted as a sepa-
rate supraoccipital. Dorsolaterally the exoccipitals join the otics above
the posttemporal fenestra as described above.

The external opening of the vagal foramen is on the suture between
the otic and exoccipital. The foramen extends through a mass of
bone which, as noted below, may be part of the exoccipital. Its in-
ternal opening (N X, Plate 10b) is above and slightly anterior to two
smaller foramina situated in the postero ventral surface of this bone.

1 The more detailed analysis of the ear region given by Dempster ('35) can not be corro-
borated by the material available to me. It is possible that his specimens were more highly
ossified although this is not indicated by his figures.

440 bulletin: museum of comparative zoology

These (N XII, Plate 10b) lead laterally to openings on the outer
surface (N XII, Plate 10a) and probably conduct branches of the
hypoglossal nerve. The channels connecting these foramina have been
carefully checked in the sectioned specimen, and the foramina them-
selves are constant in all carefully prepared specimens. An inconstant
foramen in some specimens appears posterior to the hypoglossal fora-
mina. This cannot be traced in the sectioned specimens and was
probably a nutrient opening.

It is likely that the entire dorsal and lateral surface of the braincase
posterior to the prootic foramen is a part of the exoccipital. In the
sectioned specimens it is impossible to determine where the internal
boundary occurs between the otic and exoccipital, and consequently
the extent of the participation of the exoccipital in the housing of ear
structures is a matter of conjecture.

The basioccipital forms a keystone-like wedge between the ventral
portions of the exoccipitals. This bone is exposed posteriorly as the
central region of the condyle. Posterodorsally it produces small
shelves1 which overlie the dorsolateral margin of the space presumably
occupied by the notochord. Anteriorly the basioccipital forms a por-
tion of the floor of the braincase (Plates 7B, 10b). The entire endo-
cranial surface of this bone, with the exception of the shelf-like portion
which overhangs the notochordal space, has an unfinished surface.
The central portion is elevated and extends nearly to the posterior
portion of the basisphenoid. Posterolaterally the bone meets the
exoccipitals. Posteroventrally it is exposed on the lower surface
(Plate 8a); unfinished surfaces at the sutural junction of this bone
with the parasphenoid and exoccipitals probably formed the base for
cartilaginous tubera. Anteroventrally this bone is underlain by the
parasphenoid.

3. Dermal Elements Associated with the Braincase. The parasphenoid
covers most of the ventral surfaces of the braincase. This bone is
quite thick in the region of the basipterygoid processes (Plate 10) and
becomes thinner anteriorly under the sphenethmoid and posteriorly
under the basioccipital, as may be noted in various sections through
the braincase (Text figs. 5, 6.) In ventral view (Plate 8a) it is seen to
be, quite extensive posteriorly, covering the anterior portion of the
basioccipital and laterally underlying the anteroventral surface of the
exoccipitals and the ventral surfaces of the otics. It underlies the

1 Watson ('16) described these plates associated with the exoccipitals. Dempster ('35)
however, states that the basioccipital forms most of the lower region of the foramen mag-
num (p. 176), an interpretation confirmed by our material.

sawin: cranial anatomy of eryops megacephalus 441

basisphenoid almost completely except for the distal ends of the basip-
terygoid processes.

Pronounced grooves exist on the ventral surface of the bone mesial
to the basipterygoid processes. These, as noted by Watson ('16) were
occupied by the internal carotid arteries. Each groove extends from
the region anteroventral to the fenestra ovalis to a foramen, (CG,
Plate 8a), described above, on the suture between the parasphenoid
and basisphenoid. Anterior to the basipterygoid processes the para-
sphenoid rather extensively underlies the posterior third of the sphe-
nethmoid. Anteriorly, it becomes less expanded and thinner, and in
the specimen figured, ends in a fractured edge near the unfinished
surface at the anterior edge of the sphenethmoid.

In anterior view (Plate 9a) other details of the parasphenoid be-
come visible. It extends dorsally to meet the epipterygoid beneath
the articulation of that bone with the basisphenoid and covers the
anterior surface of the basipterygoid processes of the basisphenoid to
form a strong connection with the pterygoids. The grooves for the
internal carotid artery are seen to course upward and slightly inward
in this view as they lead to the lateral surface of the basisphenoid.
Constant paired foramina occur in this groove on the anterior surfaces
of the basipterygoid processes which are with some doubt assigned
to the palatine branch of N. VII. These cannot be traced in the sec-
tioned specimens.

The extent, to which the parietals enter into the formation of the
braincase is indicated in Plate 8b, and the position of the parietal
foramen is noted in other figures. These bones are removed in Plate
7 to show the deficiency in the braincase bridged by them.

The Stapes : Somewhat more than a half of the right stapes is present
in M. C. Z. 1129, and in the reconstructed specimen most of the ele-
ment is present. The base of the stapes of the prepared specimen is
figured (Plate 6C) the dorsal portion being restored from other speci-
mens. It is apparent that the proximal end of the stapes was cartilag-
inous. No satisfactory details of the mode of insertion of the stapes
into the fenestra ovalis may be gained from studies of the specimens.
It is obvious, however, that the base as figured was applied only in
part to the fenestra and that the ventrolateral portion (OP, figs. A, C
Plate 6C) was finished in cartilage which may have been homologous
to the operculum of recent forms. The reconstructed specimen, which
is better ossified, shows definitely that the parasphenoid was suturally
united to the base, an observation made previously by Watson ('16)
and Sushkin ('27). The stapedial foramen is close to the base and

442 bulletin: museum of comparative zoology

extends obliquely upward through the bone from the posterior aspect
anteriorly. The stapes extended dorsally and from all available evi-
dence apparently abutted against a depression on the tabular (S TPD,
Plate 3). The portion abutting against the tabular has been termed the
dorsal branch by Sushkin ('27) and compared to the dorsal suprasta-
pedial process of reptiles by Goodrich ('30, p. 483). I have carefully
examined the specimen which Sushkin figured (Walker Museum 1260)
but, compared with other material, this stapes is evidently crushed
anteroposteriorly and consequently the lateral branch which he de-
scribes may well be an artifact.

4. Endocranial Cast. A composite "brain cast" of Eryops was
described by Dempster ('35) which differs both in contours and
interpretation from that described below. The most fundamental
differences, as indicated above in the description of the braincase,
are concerned with the position of the pituitary gland, the hypo-
glossal nerve, canals of the sphenethmoid, and the position of the
endolymphatic system.

The general shape of the cast is best seen in Plate 12. The inclusion
of the unossified area surrounding the parietal foramen and the space
assigned to the endolymphatic system prevent an exact delineation of
brain regions dorsally. The inclusion in the cast of the notochordal
space, the unossified area between the basioccipital and basisphenoid,
and the region beneath the ventral rim of the prootic foramen obscure
the ventral outline of the brain. Anterior to the cast of the sella tur-
cica {PIT, fig. B) the unossified space between the sphenethmoid and
basisphenoid regions shows as an irregular transverse plate. The
region of the inner ear is appended to the brain and shows but little
detail ventrally. As suggested in the description of the otic, the ves-
tibular apparatus of the inner ear was probably entirely housed in
cartilage. The foramen for the hyomandibular branch of the facial
nerve (AT VII) is indicated in that part of the cast which underlies the
prootic foramen. A protuberance ventromesial to this may doubt-
fully be referred to the proximal end of a foramen for the palatine
branch of the facial nerve. The dorsal surface of the otic region shows
only the general outlines of the anterior and posterior vertical semi-
circular canals. The space possibly occupied by the endolymphatic
duct is seen as a ridge leading from the dorsal and proximal portions of
the semicircular canals to a dorsal prominence above the cast of the
prootic foramen (END, Figs. A, C). This may be interpreted as a
portion of the cranial endolymphatic system comparable to that of
Ambystoma maculatum as described by Dempster ('30, Plate 2). De-

sawin: cranial anatomy of eryops megacephalus 443

tails as observed in the sectioned specimen do not entirely encourage
this interpretation, the duct as reconstructed in this apparently lead-
ing more directly mesially. Unfortunately the sectioned specimens are
somewhat distorted in the otic region.

The approximate locations of the forebrain (PRO), midbrain (MES)
and cerebellum (CBL) are bracketed in Plate 12, Figure A. The fore-
brain cast is convex dorsally and ventrally, where housed by the
sphenethmoid, and the forebrain probably extended posteriorly some-
what behind the cast of the parietal foramen dorsally and to a point
just behind the pituitary gland ventrally. Anteriorly the forebrain
region is in continuity with three paired channels enclosed in the an-
terior portion of the sphenethmoid. The largest of these diverges
laterally from the brain and is dorsal and lateral to the rest. It is
possible that blood vessels occupied this channel as suggested by
Broom ('13). A smaller channel is ventral and mesial to this and
probably carried the olfactory nerve. The third and smallest is situ-
ated close to the midline and is the ventral most of the group. Ac-
cording to Dempster's ('35) analysis with which I agree, this carried
the vomeronasal nerve.

Ventrally the size and shape of the pituitary region is well shown
by the cast of the sella (PIT, Fig. B). The portion of the arteries lead-
ing anteriorly from the pituitary region are also shown in the cast.
The form of the diencephalon is not known; the lateral walls of the
braincase are restored in the area above and posterior to the pituitary
gland.

The midbrain probably extended from the region posterior to the
parietal eye to a point marked by the dorsal prominences which have
been assigned above to the anterior portion of the endolymphatic
system. The cast of the roof of the braincase in this region gives little
indication of the shape of the midbrain, which was probably situated
well beneath the skull roof. The lateral surfaces of the midbrain region
have been restored. Behind the presumed endolymphatic prominences
the dorsal surface of the cast slopes downward to the general level
of the medullar region. This sloping area was probably occupied by
the cerebellum. Anterolateral to the prominences the cast of the
prootic foramen appears as a cylindrical structure diverging from the
axis of the brain at a 45° angle.

The ventral contours of the brain are not revealed behind the
pituitary gland due to the absence of finished bony surfaces and the
extensive gap between the basisphenoid and basioccipital. The ventral
surface of the medulla was probably only slightly below the proximal

444 bulletin: museum of comparative zoology

portions of the roots of the hypoglossal nerve (N XII) posteriorly.
The abducens nerve (N VI) has been indicated in its proper position
and coincides in level with the hypoglossal, but no other data can be
obtained concerning the ventral contours of the medulla. If cartilage
is assumed to be present centrally and the notochord given space
posteriorly, the brain would be flat and, in any event could not possess
the rather pronounced reptilian sigmoid flexure described by Demp-
ster ('35, p. 193). Otherwise the lack of definite boundaries prevents
any judgment as to the amphibian or reptilian characters of the brain.

ANGIOLOGY

The following structures present data for the erection of a portion
of the pattern of the cranial arterial system.

1. Pronounced grooves on the ventral surfaces of the basipterygoid
processes of the parasphenoid lead to foramina posterolateral to the
sella turcica (CG, Plate 8a; FCA, Plate 10a). These foramina open
into channels within the basisphenoid which send branches to the
cranial cavity through the floor of the sella turcica and continue
anteriorly in the bone to a point beneath the anterior portion of the
sella. There the channels again branch, each sending laterally one
member to the region of the optic nerve foramen and mesially another
which presumably anastomoses with its fellow of the other side. At
the anastomosis these last produce nutrient branches which enter the
base of the sphenethmoid.

It is obvious that the above structures are concerned with the in-
ternal carotid artery which gives rise within the basisphenoid region
to cerebral arteries leading through the floor of the sella, opthalmic
arteries which pass through the foramen for the optic nerve, and
nutrient branches to the sphenethmoid.

2. A groove leading anteriorly from the foramen for the carotid
artery (PAG, Plate 11) presents evidence for the presence of a palatine
artery which branched from the internal carotid before that vessel en-
tered the foramen in the basisphenoid region.

3. A foramen perforating the base of the stapes (STPF, Plate 6B,
C.) is 2.5 mm. in diameter and was doubtless occupied by a stapedial
artery. This vessel probably lead anterodorsally in the cranioquadrate
passage between the epipterygoid and lateral wall of the cranium and
divided into supraorbital and infraorbital branches as in recent primi-
tive tetrapods. The size of the stapedial foramen indicates the possi-

sawin: cranial anatomy of eryops megacephalus 445

bility that the mandibular branch was anastomosed by the external
carotid artery as in the Crocodilia (cf. Goodrich, '30, fig. 547).

4. The groove on the ventral surface of the postorbital and jugal
(AG, Plate 3) bones leading to foramina which communicate with a
channel in the maxillary bone was probably occupied by a vessel which
was a factor of the infraorbital branch of the stapedial artery similar
to the arteria maxillaris of Sphenodon. After entering the maxillary
bone this vessel branched, sending vessels anteriorly and posteriorly
along the margin of the upper jaw. The posterior vessel may have
anastomosed with a posterior branch of the internal carotid artery
similar to the mandibulo-jugalis (cf., Bystrow, '39, figs. 3, 15) of the
Axolotl at the foramen paraquadratum proprium. It is equally possible
that an anastomosis was formed with another factor of the stapedial
artery or temporal artery as in the frog.

Structures related to the venous system are as follows:

1. The large prootic foramen (N V, Plate 9a). The principal endo-
cranial vessels probably drained into a large vein, the vena cercbralis
medialis which made its exit from the cranial cavity by way of the
prootic foramen. Outside the cranium, this vessel was probably
joined by orbitonasal,1 ophthalmic and palatine veins, and lead pos-
teriorly as the vena capita lateralis (internal jugular).

2. .4 foramen in the otic region posterior to the prootic foramen (V,
Plate 9a). A vein from the posttemporal fossa possibly emerged
from this foramen to join the vena capita lateralis. The latter vessel
probably led through the cranioquadrate passage between the epi-
pterygoid and cranium and continued posteriorly, dorsolateral to the
stapes.2

3. The aperture behind the sella turcica (VH, Plate 10b). A hypo-
physial vein probably connected the anterior portions of the venae
capitis lateralcs. Since the remainder of the cranial foramina are quite
small, there is but little possibility of a major drainage occurring
through them.

Mandibular and lingual regions were probably served by an external
jugular vein. It is evident from the known data concerning the vascu-
lar system that the basic pattern in Eryops was in general that of a
primitive tetrapod. The only feature in which Eryops appears to de-
part markedly from the pattern is indicated by the small size of the
stapedial foramen which suggests that here, as in a number of other

1 The venous channel in the sphenethmoid possibly drained a portion of this region, leaving
the braincase somewhere in the restored area of the lateral wall.

2 Bystrow _('39) assigns the posttemporal fossa to this drainage. To my knowledge there is
no evidence in recent forms for this position of the vena capita lateralis.

446 bulletin: museum of comparative zoology

tetrapod groups, the mandibular blood supply may have been derived
by secondary anastomoses from the other sources.

MYOLOGY

The various muscles discussed fall into groups which are listed below.
The terminology of Edgeworth ('35) is followed :

a. Mandibular muscles.
Constrictor dorsalis.
Adductores mandibulae.
Intermandibular is .

b. Hyoid musculature.
Depressor mandibulae.

c. Ocular muscles.

d. Narial muscles.

It seems reasonable in the face of evidence brought out in the
discussion that the distribution of the jaw muscles in Eryops
would approach that known in recent reptiles. Unfortunately no
muscle scars are present on the skull except for feeble indications on
the mandible. The absence of these renders an analysis of the sepa-
rate jaw muscles largely conjectural as to precise areas of origins and
insertions.

Since the skull of Eryops is akinetic, it is unlikely that the primary
members of the constrictor dorsalis group, which function in moving
or bracing the palatal complex, would be functional. If present these
were vestigial. It is possible that a derivative of these, the depressor
palpebrae inferioris (cf. Edgeworth, '35, p. 57; Lakjer, '26, pp. 15 and
25) was retained in Eryops in the following position:

Origin. Along the mesial surface of the expanded portion of the
epipterygoid, possibly extending anteriorly over the mesial edge of the
palatal process of the pterygoid.

Insertion. Lower eyelid and the fleshy part of the palate.

The adductores mandibulae are usually divided into three series of
muscles, the adductores mandibulae extemus, medius and internus. In
Eryops the adductor mandibulae extemus (cf. Edgeworth, '35, p. 59;
Lakjer, '26, p. 31 ; Adams, '19, pp. 128-129) possibly had the following
relations :

Origin. Anterolateral surface of the quadrate, mesial surface of the
squamosal and quadratojugal and inner surface of the jugal.

SAWIN : CRANIAL ANATOMY OF ERYOPS MEGACEPHALUS 447

Insertion. Mandible; principally on the eoronoid process, possibly
extending posteriorly to the dorsomesial surface of the surangular.

The adductor mandibulae externus is separable into three portions
in the majority of recent reptiles. These consist of superficialis,
medialis and profundus divisions. In Eryops space is too limited be-
tween the jaw and skull laterally to permit of the function of a super-
ficialis division. The other subdivisions of the externus are inde-
terminable as separate muscles.

The adductor mandibulae medius (cf. Edgeworth, '35, pp. 59-60;
Lakjer, '26, p. 53; Adams, '19, p. 129) possibly had the following rela-
tions :

Origin. Ventrolateral portion of the epipterygoid, the central and
posterior portion of the lateral surface of the quadrate ramus of the
pterygoid, and the descending process of the squamosal.

Insertion. Mandible; bones forming the internal surfaces of the
meckelian fossa.

It seems improbable that the surfaces of origin of this muscle would
extend "far into the parietal region" as described by Adams ('19) since
this would interfere with the function of the ocular muscles.

The presence of a pseudotemporalis as a division of the adductor
mandibulae medius is a matter of conjecture. If it were present, it
would be represented by that portion of the medius which is given a
surface of origin on the ventrolateral portion of the epipterygoid.

The adductor mandibulae interims (cf. Edgeworth, '35 p. 60; Lakjer,
'26, pp. 58-61 ; Adams, '19, p. 129) may be restored in Eryops as follows :

Origin. Dorsal surface of the descending process of the pterygoid.

Insertion. Dorsal surface of posterior portion of the prearticular.
This muscle as described by Adams ('19) is apparently assigned to the
ventral surface of the pterygoid for its area of origin. Since denticles
are present over much of this surface, this interpretation seems to be
improbable.

The inter mandibular is (cf. Edgeworth, '35, p. 61) probably formed a
sheet of muscle between the jaws as in known forms.

The depressor mandibulae (cf. Edgeworth, '35, pp. 106-108; Miner,
'25) possibly existed as a slip of the levator hyoidci. This muscle
possibly took origin on the fascia of the mid-dorsal line close behind
the occiput and inserted on that part of the surangular which lies
posterior to the articular. It was probably weakly developed, the jaw
being depressed largely by its own weight.

The rectus group of occular muscles, including the bursalis and
retractor oculi, if present, probably centered in origin near and in the

448 bulletin: museum of comparative zoology

recess for eye muscles (REM, Plate 9a). The superior and inferior
obliquii may have originated on the sphenethmoid dorsally at its
lateralmost portion in front of the orbit. There is no definite indication
of this region as to the point of origin.

Narial muscles were doubtless present in the nasal region, originat-
ing on the septomaxillary bone and inserting on the fleshy rim of the
narial opening. These probably functioned in opening and closing the
narial aperture.

DISCUSSION

A number of specific morphological problems arising from the study
of the skull of Eryops have been discussed in the descriptive sections.
In the following sections an attempt is made to fit Eryops into the
general scheme of early tetrapod evolution through a consideration of
its general structural features as compared with those found in other
early amphibians and reptiles.

Watson ('19, p. 50) noted that the central skull shape among the
Labyrinthodonts was represented by that of Capitosaurus. He traced
this skull form from the Lower Carboniferous embolomere Anthraco-
saurus through various amphibians including Eryops to Cyclotosaurus
of the Upper Triassic and described the general shape of these skulls
as follows:

"The characteristic features are the wide muzzle, posterior position
and nearness to the middle line of the small orbits, and small otic
notches not very widely separated."

It is interesting to note that Cyclotosaurus is described as a type
illustrating growth regions of the skull by Bystrow ('35). The "zone
of intensive growth" in this form lies between the nasal and orbital
openings as described by that author, this animal being designated as a
member of a group which has but one zone of intensive growth in the
skull. Elongation of the skulls in the preorbital region is a primitive
tetrapod feature to be correlated with the greater demand placed on
the upper and lower jaws as a result of increase in size. This primary
elongation was probably initiated early in amphibian history as sug-
gested by Romer ('37, p. 47) who makes the following statement:

"The elongation of the face of early amphibians as contrasted with
their piscine relatives is probably related in part to changed food
habits and elongation of the jaws, with a necessary elongation of the
braincase. In great measure, however, this elongation is probably
related to size differences in the forms compared. Crossopterygians

sawin: cranial anatomy of eryops megacephalus 449

investigated are fishes of modest size; the Embolomeri whose brain-
cases are known are mostly far larger.

"The results of change in size upon proportions of animals are so
obvious that they are usually overlooked (but cf. Watson, '30). If, for
example, an animal doubles in length, its necessary food intake is
(roughly) cubed, and a disproportionate growth of mouth parts tends
to result . . ."

This secondary growth of the cranium, as shown by Romer ('37,
Fig. 15) affected the braincase of amphibians, resulting in an elonga-
tion of the ethmoid region. In relation to this situation, the nasal
capsules are widely divorced from the rest of the braincase and pre-
sumably were connected with it merely by extensions of the anterior
portions of the trabeculae.

Posteriorly, on the skull roof, the pattern of the dermal bones is
comparable to that of other non-"anthracosauroid" Labyrinthodonts
in that the tabulars are separated from the parietals by the post-
parietals and supratemporals. The presence of a well developed
posttemporal fossa is another general non-"anthracosauroid" feature.
Failure of the lacrimals to meet the orbits is a situation widespread in
labyrinthodonts. The lacrimal enters the orbits only in the "anthra-
cosauroids" and one or two "loxommids"; in all other forms it is
excluded. Primitively the presence of the septomaxilla as a superficial
dermal element prevents the lacrimal from reaching the narial rim.
In many amphibians, such as the typical sterepspondyls, the lacrimal
fails to enter the narial margin, even though the septomaxilla is re-
duced. In spite of the well developed septomaxilla in Eryops the
lacrimal does, however, form the posterior narial rim.

The palatal region of Eryops shows a fixed basipterygoid articula-
tion, medium sized interpterygoid vacuities, a much reduced quadrate,
and an epipterygoid with a prootic process. This condition is probably
closer to that known in stereospondyls than to that recognized in more
primitive Rhachitomi such as Dendrcrpeion or Edops. The former, as
described by Steen ('34), is intermediate in a number of characters
between embolomerous and rhachitomous forms. Edops1, occurring
much later in the Carboniferous, also shows a number of primitive
characters. Comparison with these two forms shows that Eryops is,
in palatal structure, a fairly advanced member of the Rhachitomi.

Dendrerpeton, like Eryops, has medium sized interpterygoid vacuities
but retains a movable basipterygoid articulation, while in Edops the

1 Data concerned with this form communicated by Professor A. S. Romer of the Museum
of Comparative Zoology, at Harvard College.

450 bulletin: museum of comparative zoology

interpterygoid vacuities are much smaller and a movable articulation
exists in the basipterygoid region. Certain other features in the
primary palatoquadrate arch of Eryops are also of phylogenetic
interest. Watson ('19) assumed that the prootic process of the epip-
terygoid was developed in the group Stereospondyli, but the presence
of this process, not merely in Eryops but also in Edops and perhaps in
Dendrerpeton (cf. Steen, '34, fig. 2F), shows that this was instead a
relatively primitive feature. It is possible to interpret the prootic
process as a true otic process which ossifies in continuity with the
epipterygoid. It falls into the proper position for an otic process in
respect to the inferred positions for the vena capita lateralis, the
stapedial artery and the hyomandibular nerve, and is, in spite of
Sushkin's ('27) objections, probably homologous with the otic process
of reptiles. DeBeer ('37) supports this point of view in his analysis
of homologues in living forms.

The quadrate is reduced in Eryops and the probability that this
element and the epipterygoid were joined by cartilage has been shown
in the descriptive section.

Soderbergh ('36) has restored the primary palatoquadrate arch in
Aphaneramma, Lyrocephalus and Platystega. This arch, as perhaps in
other of the more advanced stereospondyls, appears to have been
well developed, with a reduced ossified epipterygoid and a quadrate
connected with it by cartilage. In addition, a well developed carti-
laginous palatal process was presumably present. It has been noted
that there is no evidence of such a highly developed palatal process in
Eryops. On the basis of the limited existing knowledge of this region
it may be assumed that the stereospondyl condition is the retention of
an essentially larval type of palatoquadrate. Whether or not the an-
cestors of the stereospondyls passed through an eryopid stage with a
reduced palatal ramus in the adult is uncertain.

A comparison of the braincase of Eryops with those of other forms
may be facilitated by an attempt to interpret the embryonic condition
of the neurocranium. Within limits this also permits a comparison of
certain structural features with those of recent amphibia. In most
cases it has been possible to compare definite regions of the cranium
of Eryops with corresponding portions of the chondrocrania of recent
forms. Divisions of the braincase are listed below in relation to em-
bryonic structures. The terminology followed, which is concerned with
embryonic structures is that of DeBeer ('37).

Basioccipital. This element doubtless ossified in the posterior part
of the parachordals.

SAWIN : CRANIAL ANATOMY OF ERYOPS MEGACEPHALUS 451

Exoccipitals. The fact that the exoceipital ossifications appear to
include the supraoccipital area has been noted. This implies that the
exoccipitals represent an ossification not only of the occipital arches
but also of the synotic tectum, a condition not known in amniotes.
The presence of two hypoglossal foramina suggest that the number of
metotic segments approached or equaled that of reptiles. Dorsally
the synotic sectum must have been extensive as is testified by the broad
dorsal expansion of the exoccipitals in the supraoccipital region. The
foramen for the vagal group of nerves marks the location of the me-
totic fissure between the occipital arch and the otic capsule.

Otic. The region of the otic capsule was probably largely ossified by
the otic although the posterior portion of the capsule may have been
invaded by the exoceipital. The capsule was probably joined to the
parachordals by means of a basicapsular commissure or commissures,
and the presence of a foramen identified with N. VII anterolateral^
in the otic region suggests the presence of a prefacial commissure
which extended from the parachordal cartilage to the otic capsule
between the prootic foramen and the foramen for AT. VII.

Basis phenoid region. The dorsum sellae probably ossified in an
acrochordal cartilage while the remainder of the basisphenoid ossified
in the region of the posterior portions of the trabeculae. Lateral
basitrabecular processes were articulated with the primary palato-
quadrate arch, which probably had a palatal process in addition to
ascending and otic processes.

Later os phenoid region. This region probably ossified in the area
usually occupied in developing crania by the taenia marginalis dor-
sally and by the pilae antotica and metoptica. The taenia marginalis
probably connected an orbital cartilage with the otic capsule, and the
pila antotica extended between the acrochordal cartilage and the taenia
marginalis in front of the prootic foramen and behind the oculomotor
foramen. The pila metoptica probably joined the trabeculae in front
of the pituitary gland, and behind the optic foramen and extended up-
wards to the taenia marginalis. The area presumably derived from the
pila metoptica is difficult of determination due to poor preservation of
the lateral wall in the region immediately posterior to the sella turcica.

Sphenethmoid region. This large area probably represents an ossifi-
cation of the orbital cartilages laterally and a portion of the trabeculae
ventrally, the orbital cartilages being joined to the trabeculae by an
extensive preoptic root.

Anterior portions of the braincase. The rest of the braincase of Eryops
remained cartilaginous in the adult form and probably consisted of

452 bulletin: museum of comparative zoology

cartilaginous extensions of the anterior portions of the trabeeulae which
connected the dorsolateral part of the sphenethmoid with the nasal
capsules as is shown by the architecture of the ventral portion of the
skull roof.

The structure as described is generally like that of recent anurans
with the exception of the rather narrow interorbital region. This is
semi-trophitrabic and is intermediate in nature between the anuran
condition and that characteristic of reptiles and the primitive amphib-
ians. The increased number of metotic segments is definitely reptilian
as inferred from the Xllth nerve foramina. If the foramen on the
anterior surface of the basipterygoid process (N VII, Plate 9a) is for
the palatine branch of the facial nerve, it may be that Eryops developed
a frog-like larval pseudobasal process of the palatoquadrate. The true
basal process, as DeBeer ('37) has shown, is in front of the palatine
branch of the facial nerve. It must be reiterated that the interpretation
of the basal articulation made above is a doubtful one. If, however,
it is pseudobasal it would be in agreement with evidence indicating
the derivation of the Anura from labyrinthodont ancestors (Piveteau,
'37; D. M. S. Watson, in press).

If the above analysis of the braincase be compared with that of
Romer ('37) for Megalichthys, it is seen that Eryops differs from this
crossopterygian in that the nasal capsule and the anterior portion of
the trabeeulae, because of facial elongation, are far removed from the
primitive position. In Megalichthys and all other known crossoptery-
gians the most prominent feature of the braincase is a specialized and
kinetic bipartite condition which Romer suggests as "a retention in the
adult of an essentially embryonic condition."

It is improbable that Eryops conforms to this type of division. There
are points of weakness in the adult neurocranium of Eryops in the
region corresponding to the crossopterygian joint. However, that re-
gion in the embolomeres Paleogyrinus and Eogyrinus is stoutly con-
structed and it seems likely that in primitive tetrapods this bipartite
cranial structure was unknown. There is no doubt that the braincase
of Eryops is structurally weak between the regions corresponding to
the otico-occipital and ethmosphenoid segments of known crossoptery-
gians. In Eryops this feature is to be noted in the thin bridge between
the laterosphenoid and otic regions, in the gap in ossification between
the basisphenoid and basioccipital ventrally, and dorsally in the ab-
sence of a roof of endochondral bone between the sphenethmoid and
the supraoccipital region.

The definite space for an intra-cranial notochord of Eryops is inter-

sawin: cranial anatomy of eryops megacephalus 453

esting as a point of comparison with the notochordal space in crossop-
terygians. In older specimens of Eryops the notochordal notch as
seen in occipital view is considerably overgrown by the exoccipitals
or in some cases apparently totally obscured. If the inverse be true,
this suggests that in the larval condition the notochordal space would
be quite large, although it did not attain the aberrant size seen in
crossopterygians.

Very little is known concerning the braincases of Embolomeres
except in Paleogyrinus and other forms described by Watson ('26).
Eryops differs from Paleogyrinus in the absence of an ossified roof in
the middle portion of the braincase (as noted above) and in the wider
and less trophitrabic interorbital portion of the basicranial region.
Certain differences appear to be present in the lateral cranial wall.
Watson describes a very large prootic foramen which appears to con-
trast markedly with the smaller opening in Eryops. However, in
Eryops the thin bone in this region is often destroyed, making the
opening appear much larger, and it is not impossible that a similar
condition existed in the single known specimen of Paleogyrinus. The
foramen above the basal articulation in Paleogyrinus is said to be a
point of entry for the internal carotids and also the aperture through
which the optic and eye muscle nerves left the cranium. In Eryops
the corresponding region appears to be a foramen for the hypophysial
vein and the area of origin for the rectus eye muscles. The anterior
massive portion of the sphenethmoid is quite similar in these forms,
with the exception that in Paleogyrinus it is narrower and has only
two channels anteriorly for the olfactory nerve. An obvious difference
between these forms is in the occipital condyles, that of Paleogyrinus
being single and circular.

The braincase of Paleogyrinus is a primitive type from which that
of Eryops could easily be derived.

It has long been recognized, largely on the basis of more superficial
structures, that Eryops is a generalized rhachitomous form morpholog-
ically ancestral to stereospondyls. Watson ('19) has compared Eryops
with stereospondyls and advanced Rhachitomi, and Case ('33) has
summarized the existing evidence, adding new data concerned with
Buettncria. The braincases of various stereospondyls have been com-
pared with Eryops by the above investigators and Soderbergh ('36).
Much of the new data given in this study is concerned with the ossified
portion of the braincase, and there is obviously nothing that can be
added to comparisons with stereospondyls which are notoriously lack-
ing in ossified neurocranial structure.

454 bulletin: museum of comparative zoology

In most cases only the dermal elements of the skull are known among
members of the group Rhachitomi. Most of these data were available
to Watson ('19) who made as adequate comparisons as possible con-
sidering these limitations. Relatively little significant data has since
appeared regarding rhachitomous forms. New information is available,
however, concerning a limited number of genera including Dendrerpe-
ton, Edops, Trimerorhachis and Dwinosaurus.

Dendrerpetou acadianum, as described by Steen ('34) exhibits many
primitive cranial characters. The skull is high, and the palate, as
pointed out above, is primitive though definitely rhachitomous. Poster-
iorly in the braincase the occipital condyle is single and the supra-
occipital is unossified. Openings for the vagal group and the Xllth
cranial nerves are present and in much the same position as in Eryops
except that N. XII leaves the braincase by a single opening and the
foramen for the vagal group is much larger. The basioccipital and
exoccipital exist as a single unit separated suturally from the otic
capsule.

This animal, as placed by Steen, is intermediate in structure between
embolomerous and rhachitomous groups. It is more primitive than
Eryops in regard to the higher skull and circular condyle. There is
conflicting evidence concerned with the state of ossification of the
braincase. The otic region is much more complete than in Eryops; on
the other hand, the supraoccipital is unossified.

Edops appears to differ from Eryops in a number of features which
in general indicates that it is more primitive than the latter. The
braincase is deeper and narrower and the condition of the palate as
noted above is generalized. In the relations of the pituitary gland and
carotid artery as well as in the general endocranial contour, Edops
resembles Eryops.

New data are available on Trimerorhachis through descriptions by
Case ('35) and verbal communication by Mr. J. B. Wilson, who is
restudying this form. Bystrow ('38) has recently given a full descrip-
tion of Dwinosaurus. These two genera are examples of neotenous
rhachitomous forms. The considerable time interval which exists be-
tween Trimerorhachis of the Permo-Carboniferous and Dwinosaurus
of the upper Permian and the indisputable fact that these were derived
from different adult forms, as is shown by the pattern of the dermal
elements, does not obscure the parallelisms introduced by neoteny.
Both possess well developed branchial arches, and the reduction of
the elements of the braincases has reached a similar state in each form.
Trimerorhachis evidently has advanced less far in the reduction of

sawin: cranial anatomy of eryops megacephalus 455

neurocranial elements in that the exoeeipitals, basioeeipitals and a
portion of the opisthoties are present. In Dwinosaurus only the occip-
ital series is ossified including, however, the supraoccipital. It is of
interest that both genera show the presence of double openings in the
position of those assigned to N. XII in Eryops.

Although these neotenous types arose from rhachitomous forms,
Eryops can not be considered as an ancestor. T rimer orhachis retains
an intertemporal element which is lost in Eryops, and the basiptery-
goid joint is movable. These features hint of an origin from a more
primitive form. In both T rimer orhachis and Dwinosaurus the lacrimal
extends from orbit to naris and in neither case is an internasofrontal
element present as in Eryops.

Seymouria has been compared with the Rhachitomi by Watson
('19a) who has noted a number of resemblances in the skull. As de-
scribed and interpreted by White ('39), this form bears closer affinities
with the "anthracosauroids" than with other Paleozoic reptiles or
amphibians. The skull is more trophitrabic than that of Eryops and
the condyle is single. The interorbital region of the braincase is de-
cidedly more reptilian in the extreme reduction of the sphenethmoid
and laterosphenoid regions. These are represented in Seymouria by a
slender Y-shaped presphenoid and a paired orbitosphenoid. The pitui-
tary fossa and the dorsum sella are in much the same position relative
to paired depressions assigned to the rectus muscles of the eye as in
Eryops. As would be expected, the cranial foramina are disposed simi-
larly in the two forms except for the exit for N. XII which is single in
Seymouria. The position of the palatine nerve foramen, which has
been doubtfully assigned to a position in front of the epipterygoid for
Eryops, is in back of that process in Seymouria. This is the expected
relation in a typical tetrapod not having a complication of that region
due to the establishment of a pseudobasal process. The path of the
internal carotid artery is also modified in that it is largely enclosed in
the basisphenoid. The foramen for the vagal group of nerves is quite
large, indicating the possibility of a reptilian type of venous drainage
of the endocranial region from a posterior cerebral vessel in contrast to
the anterior drainage by way of the prootic foramen in Eryops. The
descending flanges of the pterygoids in Seymouria are quite similar,
the only exception being in the assumed presence of levators and pro-
tractors of the pterygoid in connection with the movable palato-
quadrate arch in Seymouria, which is another primitive feature of the
skull of this reptile. It seems safe to assume that the muscular system
of the latter and of Eryops approached that of the recent reptiles,

456 bulletin: museum of comparative zoology

which, according to Brock ('39), is more generalized than that of recent
amphibia.

As shown by the characters listed above, the skull of Scymouria
is much more primitive and reptilian than that of Eryops. It is clear
that these are divergent forms about equally advanced from a primitive
tetrapod condition.

SUMMARY

Modern methods of preparation have permitted a relatively com-
plete description of the osseous cranial structure of Eryops megacephalus,
a Permo-Carboniferous rhachitomous labyrinthodont. The analysis
of the skull yielded data regarding the following morphological fea-
tures among others :

1. Relation of the cranial nerves to the brain as interpreted from
an endocranial cast.

2. Structure of a part of the inner ear and the endolymphatic
system.

3. An attempted analysis of the chondrocranium.

4. The possibility of the existence of a larval pseudobasal process.

5. Relations of the anterior portion of the trabeculae and of the
nasal capsules.

6. Extent of the cartilaginous palatoquadrate.

7. Existence of an intracranial notochord.

8. Structure of the basipterygoid region.

9. Extent of the lacrimal duct.

10. Presence of a prootic process of the epipterygoid.

11. Structure and position of the sella turcica.

12. Additional metotic segments of the cranium.

13. An analysis of the cranial muscular and vascular systems.

A morphological and systematic study of the skull of Eryops mega-
cephalus indicates that this genus occupies a central position among the
labyrinthodonts. Evidence presented shows that it may be derived
from the primitive morphological stage represented by embolomeres,
but that it is well off the line leading to the Reptilia. The structural
features seen in Eryops may be antecedent to those of the stereo-
spondyls.

sawin: cranial anatomy of eryops megacephalus 457

BIBLIOGRAPHY

Adams, L. A.

1919. "A Memoir on the Phylogeny of the Jaw Muscles in Recent and
Fossil Vertebrates." Ann. N. Y. Acad. Sci., 28, pp. 51-166.

Beer, G. R. de

1937. "The Development of the Vertebrate Skull." Oxford.

Branson, E. B.

1905. "Structure and Relationships of American Labyrinthodontidae."
J. Geology, 13, pp. 568-610.

Brock, G. T.

1939. "The cranial muscles of the Gecko." Proc. Zool. Soc. London,
Series B, 108.

Broili, F.

1899. "Ein Beitrag zur Kenntniss von Eryops megacephalus (Cope)."

Palaeontographica, 46, pp. 61-84.
1917. "Unpaare Elemente im Schadel von Tetrapoden." Anat. Anz.,
49, pp. 561-576.

Broom, R.

1913. "Studies on the Permian Temnospondylous Stegocephalians of
North America." Bull. Am. Mus. Nat. Hist., 32, pp. 563-595.

Bystrow, A. P.

1935. "Schadel der Stegocephalen." Acta. Zool., 16.

1938. "Dvinosaurus als neotenische Form der Stegocephalen." Acta Zo-
ologica, 19.

1939. "Blutgefassystem der Labyrinthodanten." Acta. Zool., 20.

Case, E. C.

1911. "Revision of the Amphibia and Pisces of the Permian of North

America." Carnegie Inst. Washington, Publ. No. 146.
1931. "Description of a new species of Buettneria, with a discussion of

the brain case." Contrib. Mus. Paleont., Univ. Mich., 3, pp. 187-

206.
1933. "Progressive Chondrification in the Stegocephalia." Proc. Am.

Phil. Soc, 72, 4.

1935. "Description of a collection of associated skeletons of Trimeror-
hachis." Contrib. Mus. Paleont., Univ. Mich., 4, pp. 227-274.

Darrah, W. C.

1936. "The Peel Method in Paleobotany." Bot. Mus. Leaflets, Harvard
Univ., 4, no. 5.

458 bulletin: museum of comparative zoology

Dempster, W. T.

1930. "The Morphology of the Amphibian Endolymphatic Organ."

Jour. Morph., 50, pp. 71-120.
1935. "The braincase and endocranial cast of Eryops megacephalus
(Cope)." J. Comp. Neur., 62, 1, pp. 171-196.

Ecker, A., and Wiedersheim, R.

1896. "Anatomie des Frosches." Revised by E. Gaupp.

Edgeworth, F. H.

1935. "The Cranial Muscles of Vertebrates." Cambridge, The Univer-
sity Press.

Edinger, T.

1929. "Die Fossilen Gehirne." Ergeb. Anat. Entwick., 28, S. 1-221.

Francis, E. T. B.

1934. "The Anatomy of the Salamander." Oxford, The Clarendon Press.

Goodrich, E. S.

1930. "Studies on the Structure and Development of Vertebrates." Mac-
millan.

Gregory, W. K.

1917. "Second Report of the Committee on the Nomenclature of the
Cranial Elements in the Permian Tetrapoda." Bull. Geol. Soc.
Am., 28, pp. 973-986.
1920. "Art. II. Studies in Comparative Myology and Osteology; No.
IV. A review of the evolution of the lacrimal bone of vertebrates
with special reference to that of mammals." Bull. Am. Mus. Nat.
Hist., 42, pp. 97-263.

Htjene, F. von

1912. "Beitrage zur Kenntnis des Schadels von Eryops." Anat. Anz., 41,
pp. 98-104.

1913. "The Skull Elements of the Permian Tetrapoda in the American
Museum of Natural History, New York." Bull. Am. Mus. Nat.
Hist., 32, pp. 315-386.

Lakjer, T.

1926. "Studien iiber die Trigeminus-versorgte Kaumuskulatur der
Sauropsiden." Kopenhagen.

Lapage, E. O.

1928. "The Septomaxillary. 1. In the Amphibia Urodela." J. Morph.,

45, pp. 441-472.
1928. "The Septomaxillary of the Amphibia, Anura and of the Peptilia.

II." J. Morph., 46, pp. 399-430.

SAWIN : CRANIAL ANATOMY OF ERYOPS MEGACEPHALUS 459

Miner, R. W.

1925. "The Pectoral Limb of Eryops and other Primitive Tetrapods."
Bull. Am. Mus. Nat. Hist., 51, VII, pp. 145-312.

Matthew, W. D.

1915. "Hitherto unpublished Plates of Tertiary Mammalia and Permian
Vertebrata." Prepared under the direction of Edward Drinker
Cope. Am. Mus. Nat. Hist., Monograph Series no. 2.

McCotter, R. E.

1917. "The Vomero-nasal Apparatus in Chrysemys punctata and Rana
catesbiana." Anat. Rec, 13, pp. 51-67.

O'Donoghue, C. H.

1920. "The Blood Vascular System of the Tuatara, Sphenodon punc-
tata." Phil. Trans. Roy. Soc. London, Series B, 210, pp. 175-252.

Olson, E. C.

1936. "The Dorsal Axial Musculature of Certain Primitive Permian
Tetrapods." J. Morph., 59, 2, pp. 265-311.

PlVETEAU, J.

1937. "Paleontologie de Madagascar. XXIII. Un Amphibien du Trias
Inferieur." Ann. Paleont., 26, pp. 135-177.

Price, L. I.

1935. "Notes on the Brain Case of Captorhinus." Proc. Boston Soc.
Nat. Hist., 40, 7, pp. 377-386.

Ramaswami, L. S.

1936. "The Cranial Morphology of the Bufonid Head." Proc. Zool. Soc.
London, Part IV, pp. 1137-1169.

Retzius, G.

1881. "Das Gehororgan der Fische und Amphibien." Pt. 1, Stockholm.

Romer, A. S.

1928. "Vertebrate Faunal Horizons in the Texas Permo-Carboniferous

Red Beds." Univ. Texas Bull., 2801, pp. 67-108, July.
1933. "Vertebrate Paleontology." Univ. Chicago Press.
1935. "Early History of Texas Redbeds Vertebrates." Bull. Geol. Soc.
Am., 46, pp. 1597-1658.

1937. "The Braincase of the Carboniferous Crossopterygian Megalich-
thys nitidus." Bull. Mus. Comp. Zool., 82, 1, pp. 1-73.

Save-Soderbergh

1932. "Preliminary Note on the Devonian Stegocephalians from East
Greenland." Meddelelser om Gr0nland, Bd. 94, no. 7, 1932.

460 bulletin: museum of comparative zoology

1935. "On the Dermal Bones of the Head in Labyrinthodont Stegocepha-
lians and Primitive Reptilia." Meddelelser om Gr0nland, Bd.
98, no. 3, pp. 1-211.

1936. "On the Morphology of Triassic Stegocephalians from Spitzbergen
and the Interpretation of the Endocranium in the Labyrinthodon-
tia." Kungl. Svenska Vetenskapsakad. Handlingar, Tredje Serien,
16, 1, pp. 1-181.

SCHMALHAUSEN, J. J.

1923. "Der Suspensorialapparat der Fische und das Problem der Gehor-
knochelchen." Anat. Anz., 56, pp. 534-543.

Stadtmuller, F.

1936. "Kranium und Visceralskelett der Stegocephalen und Amphibien."
Handbuch der vergleiehenden Anatomie der Wirbeltiere, 4,
501-687.

Steen, M. C.

1934. "The Amphibian Fauna from the South Joggins, Nova Scotia."
Proc. Zool. Soc. London, Part 3.

(Mrs. J. Brough)

1938. "On the Fossil Amphibia from the Gas Coal of Nyrany and other
Deposits in Czechoslovakia." Proc. Zool. Soc. London, Series B,
108, 2, pp. 205-283.

Stickler, L.

1899. "Ueber den microscopischen Bau der Faltenzahne von Eryops
megacephalus Cope." Palaeontographica, 46, pp. 85-94.

Sushkin, P. P.

1927. "On the modifications of the mandibular and hyoid arches and their
relations to the brain case in the early Tetrapoda." Palaeontolo-
gischen Zeit., 8, H. 4.

Watson, D. M. S.

1916. "On the Structure of the Brain Case in Certain Lower Permian

Tetrapods." Bull. Am. Mus. Nat. Hist., 35, no. 31, pp. 611-636.
1919a. On Seymouria, the most primitive known reptile. Proc. Zool. Soc.

London, Parts III and IV, pp. 267-301.
1919b. "The Structure, Evolution and Origin of the Amphibia, — The

"Orders" Rachitomi and Stereospondyli." Phil. Trans. Roy. Soc.

London, Series B, 209, pp. 1-73.
1926. "Croonian Lecture. — The Evolution and Origin of the Amphibia."

Phil. Trans. Roy. Soc. London, Series B, 214, pp. 189-257.
1929. "The Carboniferous Amphibia of Scotland." Palaeontologia

Hungarica, 1, pp. 221-252.

SAWIN : CRANIAL ANATOMY OF ERYOPS MEGACEPHALUS 461

Westoll, T. S.

1938. "Ancestry of the Tetrapods." Nature, 141.

1938. "The origin of the Tetrapods and their relation to the bony fishes."
J. British Assoc, pp. 59-60.

White, T. E.

1939. "Osteology of Seymouria baylorensis Broili." Bull. Mus. Comp.
Zool. Harvard Coll., 85, no. 5.

WlLLARD, W. A.

1915. "The Cranial Nerves of Anolis carolinensis." Bull. Mus. Comp.
Zool. Harvard Coll., 59, no. 2.

WlLLISTON, S. W.

1918. "(2) The Osteology of Some American Permian Vertebrates III."
Contrib. Walker Mus., 2, no. 4.

462

bulletin: museum of comparative zoology

EXPLANATION OF ABBREVIATIONS

AG = arterial groove
ANG= angular
ART = articular

BO =basioccipital

BPTP = basipterygoid process

BSPH=basisphenoid region

CA = carotid artery

CBL = cerebellum

CC = cerebral cavity

CG = groove for carotid artery

CO = coronoid

CO I=intercoronoid

CO II = precoronoid

D = dentary

DF = dental foramen

DLR = dorsal longitudinal ridge

DS = dorsum sellae

EO =exoccipital

END = endolymphatic organ

EPT = epipterygoid

FCA = foramen of internal carotid

artery
FD = dental foramen
FEN O V = fenestra ovalis
FR= frontal

IMF = inframeckelian fossa
INF =internasofrontal

JU = jugal

L = lacrimal

LG = lacrimal groove

LS = laterosphenoid region

M
MES = mesencephalon
MF = mandibular foramen
MTF = mental foramen
MX = maxilla

N
N A = nasal
N AC = space for cartilaginous nasal

capsule
NCH =notochord
NF = nutrient foramen
N I to N XII = cranial nerves
NVN = vomeronasal nerve

0 = otic

OA = ophthalmic artery
OP = operculum

OT PR EPT = otic process of the epi-
pterygoid

P = parietal

PA =prearticular

PAL = palatine

PAG = groove for palatine artery

PAS =parasphenoid

PC = chamber for parietal eye

PF = parietal foramen

PIT = pituitary gland

PMX =premaxilla

PO=postorbital

POF=postfrontal

POP =postparietal

PRF = prefrontal

PRO = prosencephalon

PS = parasphenoid

PT = pterygoid

PTF = posttemporal fossa

PVO = prevomer = Vomer

sawin: cranial anatomy of eryops megacephalus

463

Q = quadrate

QJ = quadratojugal

R = cranial recess for endolymphatic

organ
REM = recess for rectus eye muscles
ROT = otic recess

SMX =septomaxilla

SP = splenial

SPP = postsplenial

SQ = squamosal

ST = supratemporal

STP = stapes

STPD= abutment for dorsal portion

of stapes
STPF = stapedial foramen

S ANG = surangular

SCC AV = anterior vertical semicircu-
lar canal

SCC PV= posterior vertical semicir-
cular canal

SE=sphenethmoid region

SET = sella turcica

T = tabular

TG = trabecular groove

TP = ectopterygoid (transpalatine)

V = vein

VH= hypophysial vein

EXPLANATION OF PLATES

PLATE 1

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 1
Eryops megacephalus. Dorsal view of skull. M.C.Z. No. 1129. x 3/10.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 1.

PLATE 2

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 2
Eryops megacephalus. Ventral view of skull, x 3/10.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 2.

PLATE 3

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 3

Eryops megacephalus. Ventral view of skull with palate largely removed.
The right basipterygoid process is shown in horizontal section, x 3/10.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 3.

PLATE 4

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 4

Eryops megacephalus. a. Lateral view of skull with mandible in position.
x 3/10. b. Lateral view of skull without mandible. M.C.Z. 1129. x 3/10.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 4.

PLATE 5

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 5

Eryops megacephalus. A. Jaw, lateral view, x 1/4. B. Jaw, mesial view,
x 1/4. C. Jaw, dorsal view, x 1/4. D. Jaw, ventral view, x 1/4. E. Jaw,
posterior view, x 1/2.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 5.

SANG

PLATE 6

Sawin — ■ Cranial Anatomy of Eryops megacephalus

PLATE 6

A. Skull, posterior view, x 3/10. B. Braincase, posterior view, x 1/2.
C. Stapes, x 1/2. A. posterior view of right stapes; B. anterior view of right
stapes; C. base of stapes.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 6.

PTF

BPTP

PLATE 7

Sawin — ■ Cranial Anatomy of Eryops megacephalus

PLATE 7

Eryops megacephalus. A. Braincase in dorsal view. B. Ventral portion of hor-
izontally sectioned braincase in dorsal view. The plane of the section is shown
by an interrupted line in the figure below. C. Detail of the basisphenoid. The
otics, the laterosphenoid region and the dorsum sellae have been removed to
expose the area of origin of some of the rectus muscles of the eye. x 1/2.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 7.

N VN

PLATE 8

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 8

Eryops megacephalus. a. Braincase in ventral view. b. Dorsal portion of
horizontally sectioned braincase in ventral view. The plane of the section is
indicated in the explanation of Plate 7B. x 1/2.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 8.

STP

5CC

PLATE 9

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 9

Eryops megacephalus. a. Anterior view of braincase and posterior portion of
the right palato quad rate complex. The sphenethmoid region has been cut away
and the braincase is shown in a transverse section which passes through the
anterior part of the sella turcica. The transversely sectioned part of the ptery-
goid is cut just in back of its descending process, b. Lateral view of braincase
and posterior part of the right palatoquadrate complex, x 1/2.

BULL. MUS. CO MP. ZOOL Sawin. Cranial Anatomy of Eryops megacephalus. Plate 9.

OT PR EPT

PLATE 10

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 10

a. Braincase in lateral view. b. Left half of braincase as seen from the
median section.

BULL. MUS. COMP. ZOOL Sawin. Cranial Anatomy of Eryops megacephalus. Plate 10.

BSPH

PLATE 11

S.vwin — Cranial Anatomy of Eryops megacephalus

PLATE 11

Eryops sp. a. Dorsal view of the basisphenoid and posterior portion of the
parasphenoid. b. Ventral view. c. Lateral view, x 1.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 11.

tftoeW

***:■

N5H

PAG

PLATE 12

Sawin — Cranial Anatomy of Eryops megacephalus

PLATE 12

Eryops megacephalus. Endocranial cast. A. Lateral view. B. Ventral view.
C. Dorsal view, x 1/2.

BULL. MUS. COMP. ZOOL. Sawin. Cranial Anatomy of Eryops megacephalus. Plate 12.

CBL MES

PRO

NCH

END

NVN

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE

Vol. LXXXVIII, No. 6

REVISION OF THE AFRICAN TERRAPIN OF THE
FAMILY PELOMEDUSIDAE

By Arthur Loveridge

_ j. ary

Museum of Comparative Zoology
--^ Harvard University **

CAMBRIDGE, MASS., U.S.A.
PRINTED FOR THE MUSEUM
October, 1941

OCT 31 194]

No. 6. — Revision- of the African Terrapin of the Family Pelomedusidae

By Arthur Loveridge

It has long been obvious to taxonomists that, owing to inadequate
material, the key furnished by Boulenger (1889a, p. 192) for distin-
guishing members of the genus Pelusios proved so misleading as to
result in the greatest confusion for half a century, as is shown by the
voluminous literature. Indeed, the only attempt to straighten out
the situation was that of Siebenrock (1903d), who failed, on account of
the alternative characters which he selected proving no less variable
than those of Boulenger which he rejected.

Recent nomenclatorial changes necessitated my investigating the
status of certain names in order to ascertain which were applicable
to three species of the family collected on my last visit to East Africa.
I found the whole situation so involved that eventually I decided on
the present revision, which is humbly offered in the hope that its con-
clusions will form a stable basis that will stand the test of time.

The most important change is the rejection of niger from the West
African fauna, except as a synonym of subniger Lacepede, for the
'niger' of Boulenger and subsequent authors is a synonym of gabonensis
Dumeril, and nothing to do with the niger of Dumeril & Bibron who
believed, and probably correctly, that their 180 mm. type came from
Madagascar. I am deeply indebted to Mons. Angel for detailed notes
on the type of niger, without which I could not have settled the point.

I take this opportunity of thanking Mons. Angel (Paris), Dr. Oscar
de Beaux (Genoa), Dr. L. D. Brongersma (Leiden), Dr. P. R. Reveil-
liod (Geneva), Dr. L. Forcart (Basel), Mr. G. Netting (Pittsburgh), and
Mr. H. W. Parker (British Museum) for the loan of material or for
answering my queries respecting material in their care. And by no
means least, Mr. Vesey FitzGerald of the Malay States, who took the
trouble to secure a series of seychellensis, which he has presented to the
Museum of Comparative Zoology.

No attempt has been made to complete bibliographical references
prior to 1889. From that date I have attempted to list all those found
in a search through 1,500 papers on African herpetology. I hope that
omissions will not prove numerous. A synopsis of the information
culled from this literature, is given for each species, though in the
fields of anatomy and physiology only the barest reference is made.

Taxonomic changes. Many earlier synonymizings are confirmed,
while the following alterations in nomenclature are made for the first
time.

468 bulletin: museum of comparative zoology

Pelomedusa galeata orangensis Hewitt

= Pelomedusa s. subrufa (Lacepede).
Pelomedusa galeata devilliersi Hewitt

= Pelomedusa s. subrufa (Lacepede).
Pelomedusa galeata damare?isis Hewitt

= Pelomedusa s. subrufa (Lacepede).
Pelomedusa subrufa wettstcini Mertens

= Pelomedusa s. subrufa (Lacepede).
Emys olivaeea Schweigger is revived as

Pelomedusa s. olivaeea (Schweigger).
Pelomedusa gasconi Rochebrune

= Pelomedusa s. olivaeea (Schweigger).
P. g. var. disjuncta Vaill. & Grand.

= Pelomedusa s. olivaeea (Schweigger).
Sternothaerus niger Blgr. (non D. & B.)

— Pelusios gabonensis (A. Dumeril).
Sternothaerus derbianus Gray = Pelusios subniger (Lacepede).

Sternothaerus oxyrhinus Boulenger = Pelusios subniger (Lacepede).
S. nigricans seycheUensis S\ebenrock= Pelusios subniger (Lacepede).

Origins. It would appear to me that the Pelomedusidae entered
Africa from the north or northwest with Pelomedusa s. olivaeea (Sene-
gal to Eritrea), if recognisable, as the oldest living form, giving off
P. s. subrufa (Sudan to Cape) which in turn produced Pelusios adan-
sonii (Senegal to the Nile). The latter would appear to have given rise
to the handsomest member of the genus P. gabonensis (Liberia to
Congo) an inhabitant of the West African rainforest region. The
widespread P. subniger may also be assumed to have risen from
adansonii or some common ancestor, and in the east developed the
handsome sinuatus (Somaliland to Natal).

Designation of Headshields. The shields called frontals by Boul-
enger, prefrontals by some other authors, I designate supraorbitals.
The huge shield which he called a parietal, appears to have been formed
by fusion of the frontal with the parietals, as may be deduced from the
frequency with which grooves appear on its posterior half, indicative
of the lines of fusion. This shield I call the frontal, it is flanked on
either side by a large temporal which is subject to subdivision.

Description. At the risk of monotonous repetition, I have drafted
somewhat detailed descriptions based on the material at my disposal,
augmented in footnotes by further variations observed by other
workers. This has been done because some recent authors have been

loveridge: African pelomedusidae 469

apt to attach undue importance to extremely variable characters,
often those which have long been known to be sexual or subject to
alteration with age. The descriptions now drawn up show that most of
the species run the whole gamut of such variations.

Localities. Rochebrune's (1884a) localities have been omitted as
untrustworthy. Otherwise I have attempted to list, after each species,
all the localities from the literature for the past fifty years, except
such as have been subsequently shown to be based on erroneous
identifications, now transferred to their appropriate species. I should
have liked to cite the author responsible for each locality, this, how-
ever, would have added enormously to the burden of printing.

Political areas have been arranged on a definite geographical plan,
the place names alphabetically within their respective political areas.

Family PELOMEDUSIDAE
Freshwater Terrapin

Semiaquatic tortoises covered with horny shields overlying the bony
plates of a more or less depressed box-like exoskeleton ; head and neck
completely retractile within the shell by lateral flection; nostrils at end
of snout; jaws covered with a horny beak; dentary single; palatine
bones in contact; no nasals; prefrontals in contact; temporal region
not roofed over; a bony temporal arch; no parietosquamosal arch;
digits moderately elongate; hind foot webbed; five or four1 claws.

For detailed structural definition of the Pleurodira, in which super-
family the Pelomedusidae are included, see Boulenger, 1889a, Cata-
logue of Chelonians ... in the British Museum, pp. 187-190.

Range. Africa, Seychelles, Madagascar and South America.

Key to the Genera

Anterior lobe of plastron immovable; pectoral shields participating equally
with abdominals on bridge; plastral fenestration persisting till late in life;

mesoplastra small and lateral Pelomedusa

(p. 470)

Anterior lobe of plastron movable in adult; pectoral shields almost excluded

from bridge by abdominals; plastral fenestration closed very early in life;

mesoplastra extending right across the plastra Pelusios

(p. 481)

1Four on hind foot of Malagasy Erymnochelys and S. American Podocnemis.

470 bulletin: museum of comparative zoology

Genus Pelomedusa

1830. Pelomedusa Wagler, Nat. Syst. Amphib., p. 136 (type galeata =

subrufa).

1831. Hydraspis Gray (part), Syn. Rept., p. 39 (type Hydrastis longicollis) .
1835. Pentonyx Dumeril & Bibron, Erpet. Gen., 2, p. 389 (type capensis =

subrufa).

Skull without supratemporal roof; quadrato-jugal widely separated
from parietal ; upper jaw with very indistinct median ridge on alveolar
surface ; between the orbits are a pair of supraorbital shields separated
by a longitudinal suture and followed by a large frontal flanked by
temporals; plastron narrow, without hinge; mesoplastral bones small,
lateral, wedged in between the hyoplastra and hypoplastra; plastral
fenestration persisting till late in life; digits very short, mostly with 2
phalanges, feet with 5 claws.

Range. Africa south of the palaearctic zone; Madagascar.

Remarks. A single species (Pelomedusa subrufa) with a poorly de-
fined race occurring along the northern fringe of its range. Scattered
individuals possessing the character of the northern form crop up
within the range of typical subrufa, but should not be considered to
invalidate the recognition of the northern form unless they prove to
be far more numerous than our present information would indicate.

Key to the Races

Pectoral shields in contact on the median line of the plastron s. subrufa

(p. 470)
Pectoral shields separated on the median line of the plastron s. olivacea

(p. 480)

Pelomedusa subrufa subrufa (Lacepede).

1788. La Roussatre Lacepede, Hist. nat. Quadrup. ovip. Serpens, 1, p. 173,

pi. xii: "de l'lnde" as Sonnerat coll. restr. to Cape.

1789. Testudo subrufa Lacepede, Hist. nat. Quadrup. ovip. Serpens, 2,

Synopsis methodica (a table in which binomials are employed.)

1789. Bonnaterre, p. 27.

1802. Daudin, p. 132.

1792. Testudo galeata Schoepff, Hist. Testud., p. 12, pi. hi, fig. 1: "India
orientali, Carolina" restr. to Cape Flats, S. Africa.

1802. Daudin, p. 136.

1798. Testudo Badia Donndorff, Zool. Beytr. Linn. Natur., 3, p. 34: No lo-
cality. Based on Lacepede's "La Roussatre."

1814. Emys galeata Schweigger, p. 38.

1814. Emys subrufa Schweigger, p. 39.

loveridge: African pelomedusidae 471

1830. Pelomedusa galeata Wagler, p. 136, pi. ii, figs. 36-43.
1848. Peters, p. 492, pi. xvii, figs. 1-3.

1849. Smith, App., p. 1

1862. Strauch, p. 150.

1865. Strauch, p. 111.

1869b. Peters, p. 657.

1870. *Steindachner, p. 326.

1880a. Boulenger, p. 146.

1880b. Peters, p. 509.

1881b. *Boettger, p. 410.

1881c. Boettger, p. 535.

1882a. Peters, p. 6.

1884a. *Rochebrune, p. 22.

1887a. Bocage, p. 202.

1887b. Boettger, p. 140.

1888a. Boettger, p. 13.

1889. Boettger, p. 296.
1889a. *Boulenger, p. 197.
1890a. *Muller, p. 296.

1890. Strauch, p. 103.
1893a. Boettger, p. 14.
1893b. Boettger, pp. 113, 122
1894a. Boettger, p. 88.
1894. Fleck, p. 83.

1894. Gunther, p. 85.
1895a. Bocage, p. 5.
1895b. *Boulenger, p. 531.

1895. Jeude, p. 227.
1896a. Bocage, p. 97.
1896a. *Boulenger, p. 546.
1896b. Boulenger, p. 6.
1896c. *Boulenger, p. 16.
1896e. Boulenger, p. 213.
1897g. Boulenger, p. 277.

1897. Siebenrock, p. 247.

1898. Jeude, p. 9.
1898a. Vaillant, p. 135.
1898. Sclater, p. 97.

1898. Tornier, p. 282.

1899. Siebenrock, p. 566.
1900a. Mocquard, p. 94.
1900b. Tornier (part), p. 583.
1901. Gadow, p. 391.

* These should be transferred in whole or in part to olivacea should that race prove to be
recognisable.

472

bulletin: museum of comparative zoology

1901c.

1902d.

1902a.

1902b.

1902c.

1903e.

1903a.

1904.

1905.

1905c.

1906a.

1906b.

1907J.

1907.

1908.

1909.

1910.

1910b.

1910.

1910a.

1911c.

1911.

1911b.

1911d.

1912c.

1912b.

1913.

1913c.

1915.

1917.

1919.

1921d.

1922a.

1922.

1923b.

1923g.

1924b.

1924a.

1925b.

1926a.

1927.

1928.

Tornier, p. 67.
Boulenger, p. 445.
Mocquard, p. 6.
Tornier, p. 580.
Tornier, p. 665.
Boulenger, p. 217.
Siebenrock, p. 255.
Peracca, p. 1.
Neumann, p. 390.
*Tornier, p. 366.
Mocquard, p. 480.
Siebenrock, p. 40.
Boulenger, p. 483.
Lonnberg, p. 2.
Werner (1907), p. 1826.
Siebenrock, p. 561. •
Meek, p. 414.

apparently transposed with that of gabonensis)

MeeK, p. 414.
Nieden, p. 7. (text

Siebenrock, p. 718.

. 26, 53, pi. xx, figs. 1-2, pi. xxi, fig. 1.

Vaillant & Grandidier, pp

Werner, p. 305.

Boulenger, p. 162.

Lampe, p. 148.

Sternfeld, p. 411.

Sternfeld, p. 53, figs. 65-66.

Sternfeld (part), p. 201.

Werner, p. 469.

Boettger, p. 319.

Nieden, p. 64.

Ptawitz, p. 670.

Sternfeld, p. 414.

Schmidt, pp. 415, 598, 601, fig

Loveridge, p. 52.

Angel, p. 39.

Kaudern, p. 449.

Calabresi, p. 156.

Loveridge, pp. 930, 932.

Loveridge, p. 2.

Werner, p. 266.

Flower, p. 933.

Mertens, p. 152.

Calabresi, pp. 20, 38.

Cott, p. 952.

* These should be transferred in whole or in part to olioacea should that race prove to be
recognisable.

loveridge: African pelomedusidae

473

1928d.

1928b.

1928m

1929.

1929h.

1929.

1930a.

1931d.

1931.

1932b.

1933.

1933h.

1933.

1933m

1934.

1935b.

1935.

1936.

1936h.

1936 j.

1937.

1937f.

1937b.

1938.

1939b.

1940a.

1831.

1835.

1860.

1844.

1855.

1867a.

1870.

1872a.

1873b.

1888a.

1934a.

1937e.

1849.

Loveridge, p. 51.

*Scortecci, p. 336.

Witte, p. 45.

Lindholm, p. 288.

Loveridge, p. 16.

Rose, p. 185, fig. 123.

*Scorteeci, p. 215.

Angel, p. 551.

Mann, p. 366.

Monard, p. 110.

Power, p. 48.

Parker, p. 340.

Flower, p. 752.

Loveridge, p. 211.

Schmidt, p. 3.

Witte, p. 67.

Pitman, p. 307.

FitzSimons, p. 307.

Hewitt, p. 325.

Cowles, p. 6.

Loveridge, p. 19.

Loveridge, p. 225.

Buxton, p. 102.

Loveridge, pp. 489, 492, 495.

Monard, p. 146.

FitzSimons, p. 155.

FitzSimons, p. 20.

Scortecci, p. 6.
Hydraspis subrufa Gray, p. 39.
1Pentonyx Capensis Dumeril & Bibron, Erpet. Gen., 2, p. 390, pi. xix,

figs. 2-2b: Cape of Good Hope, etc. restr. by Mertens.

A. Dumeril, p. 163, pi. xiii, fig. 3.
Pelomedusa subrufa Gray, p. 38.

Gray, p. 53.

Steindaehner, p. 6.

Gray, p. 81.

Gray, p. 24.

Gray, p. 71.

Glint her, p. 50.

Mertens, p. 10.

Hewitt, p. 14, pi. iv, figs. 1-3; pi. ivA, fig. 1; pi. xxvii, fig. 3.
Pentonyx americana Cornalia, Vert. syn. Mus. Mediolan., p. 13: "New

~wr i'i l it ci _ . i l A J? • *

entonyx americana Lornaiia, vert. syn. ivius
York" probably a South African specimen

* These should be transferred in whole or in part to olivacea should that race prove to be
recognisable.

1 This specimen also served as the type of Tesludo subrufa Lacepede.

474 bulletin: museum of comparative zoology

1855. Pelomedusa Mozambica "Peters M.S.S. 1848" Gray, Cat. Shield Rept.,

p. 53: Mozambique (nomen nudum).

1856. Pelomedusa mossambicensis "Peters" Lichtenstein & V. Martens,

Nomenclator Rept., p. 2: Mozambique (nomen nudum).
1863a. Pelomedusa nigra Gray, Ann. Mag. Nat. Hist. (3), 12, p. 99: Natal.
1870. Gray, p. 81.
1873b. Gray, p. 72.

1872. Hydras-pis galeata Gray, in Sowerby & Lear, p. 7, pis. xlix-1.
1935. Pelomedusa galeata subrufa Hewitt, p. 326.
1935. Pelomedusa galeata orangensis Hewitt, Rec. Albany Mus., p. 332, pi.

xxxi, fig. 3, pi. xxxii, figs. 3-4: ? Kimberly, Cape Province.
1935. Pelomedusa galeata nigra Hewitt, p. 335, pi. xxxii, figs. 1-2.
1935. Pelomedusa galeata devilliersi Hewitt, Rec. Albany Mus., p. 337, pi.

xxxi, figs. 2, 4: Besondermeid, Steinkop, Namaqualand.
1935. Pelomedusa galeata damarensis Hewitt, Rec. Albany Mus., p. 338, pi.

xxxiii, figs. 1-4: Quickborn, nr. Okahandja, S. W. Africa.
1935. Pelomedusa galeata galeata Hewitt, p. 342, pi. xxxiv, figs. 3-4.
1937a. Pelomedusa. subrufa wettsteini Mertens, Zool. Anz., 117, p. 141, figs. 1,

4: Majunga, West Madagascar.
1937b. Pelomedusa subrufa damarensis Mertens, p. 5.
1937d. Pelomedusa subrufa subrufa Mertens, p. 3.
1937e. Pelomedusa subrufa orangensis Hewitt, p. 14.

Names. Marsh Terrapin, Helmeted Terrapin (English); njaba
(Ganda); ngurv. (Kitosh); malwala (Gogo); camba na madsi (at Rios de
Sena); nambe (at Querimba) ; fuduc (at Quilimane); isifudu (in Um-
zumbe Valley) ; ufodo (Fingo, but not specific) ; ofiufiu (at Kalukembe) ;
kitio (at Quisange).

History. Mertens (1937, p. 139) has shown that the name galeata,
so long employed for the marsh terrapin, is antedated by submfa of
Lacepede, not of Daudin.

Description. Head broad, snout short; a pair of supraorbital shields1
followed by a very large frontal, flanked by a pair of temporals; nar-
rowest interorbital width much shorter than the longitudinal suture
between the supraorbitals; upper jaw angularly rounded; chin with,
rarely without2, a pair of small barbels; carapace depressed, its height
included in its length from 2.50 to 3.50 times, slightly elongate, its
posterior margin rounded or very slightly serrate; vertebral shields 5,
more or less obtusely keeled, first vertebral largest, broader than long,
fourth also broader than long; costals 4, very rarely 53; marginals 22,

1 Transversely divided in a Gomodimo terrapin (FitzSimons, 1935b, p. 307).

2 Absent in specimens from Nkate (FitzSimons, 1935b, p. 307).

3 4 on left and 5 on right in Oda example (Tornier, 1905c, p. 366).

loveridge: African pelomedusidae 475

very rarely 24; supraeaudals 2; plastron very much smaller than the
opening of the carapace; anterior lobe truncate, broader than posterior
which is more or less angularly and deeply notched; intergular much
longer than a gular, its sides straight1; humerals 2, very rarely 42;
pectorals forming a median suture of variable length; pectoral and
abdominal shields nearly equally developed on the bridge; width of
bridge contained about 2^ times in the width of the plastron.

Anatomy. The skull has been described by Siebenrock (1897, p. 247),
and later (1899, p. 566) the glottis; the musk gland by Peters (1848,
p. 492), the alimentary canal by Vaillant & Grandidier (1910, p. 55),
and the nervous system by Rawitz (1915, p. 670).

Coloration. Above, head and limbs dark olivaceous with, or without
darker and lighter vermiculations; carapace light olive, yellow brown,
or dark brown, uniform or with the margins of the shields edged with
black; in young, which are olive green, the marginals are edged with
creamy yellow alternating with black. Below, chin and throat white to
yellowish white; plastron yellowish or horn colour with brown or black
infuscations about the edge and along the sutures of the shields, or
plastron entirely black.

In East Africa the only specimens seen which had entirely black
plastrons came from the Mabira Forest and near Mount Elgon; per-
haps, therefore, such colouring occurs in forested, or recently de-
forested, areas, while yellow plastrons are associated with more arid
regions?

South African specimens whose plastrons have a maroon or dark
red tinge, may supposedly be the result of staining from laterite soil,
as is stated to be the case with Natal and Malagasy snbniger.

Peters (1882a, p. 6) remarks that the iris has a silver gleam and is
grey towards the periphery. Mann (1931, p. 366) shows that both
the narrow circumpupillary zone as well as the narrow peripheral zone
are covered with silver pigment while the vessels of the iris have much
chocolate pigment in their walls obscuring the blood colour. Hewitt
(I937e, p. 11) remarks that the pupil is circular and the iris has a
yellow margin. Possibly based upon preserved material?

Measurements. Length of carapace of a d71 from Tanganyika
Territory, 200 mm., breadth 135 mm., height 68 mm.; length of cara-
pace of a 9 from Dodoma 161 mm.3, breadth 128 mm., height 63 mm.
Both surpassed by South African measurements furnished by Hewitt

1 Pyriform in some South African specimens (Hewitt, 1935, p. 326).

2 In an aberrant individual from Ngare na Nyuki (Lonnberg, 1907, p. 2).
:The 9 of 179 mm. recorded by me (1933h, p. 212) proves to be a d\

476 bulletin: museum of comparative zoology

(1935, p. 331) of a cf from Kingwilliamstown district (presumed)
measuring 325 mm., breadth 245 mm., and a 9 from Albany district
(presumed) measuring 241.5 mm., breadth 180 mm. A cT from Khor
Arbat with a carapace length of 200 mm. weighed 1.25 kilos (2.75 lbs.)
according to Flower (1933, p. 753).

Sexual dimorphism. Males, recognisable by their longer tails, have
narrower shells than females, while their claws are said by Werner
(1910a, p. 305) to be stronger, not longer. Usually also they exhibit
a slight depression in the posterior third of the plastron, the notch of
its posterior lobe is not a guide. Hewitt (1935, p. 327) cites other dif-
ferences which are individual rather than sexual, cf. contradictions
regarding mesial notch and pear-shaped intergular.

Breeding. The only accounts of ovipositing are those published by
Hewitt (I937e, p. 13), the most detailed being condensed as follows.
One warm November evening, near Grahamstown, Miss. N. White
surprised a female laying in a hole which had a surface diameter of 2}^
inches, and a depth of from 4 to 5 inches terminating in a chamber
slightly larger than a tennis ball. The surrounding ground was sun-
baked, hard and dry, but the site had been softened by the terrapin
discharging cloacal water, two other spots where an attempt to dig had
been abandoned, were found within a couple of yards.

There were already 5 eggs in the hole and the terrapin continued
depositing others at the rate of about one per minute until a total of
14 had been laid. Each egg had a dent in its tough membranous sur-
face, and was covered with slime which took about an hour to dry.
A noteworthy feature was the absence of any musky odour during
ovipositing, though when the site was revisited next morning a foul
smell was immediately apparent. The hole had been filled in with the
excavated muddy soil and flattened, so that once sunbaked it would
have escaped detection. An egg measured 37 x 25 mm.

Schmidt (1933, p. 3) records a hatchling, from Angola, which had
already lost its egg tooth, as measuring 25 mm. in length and 19.5 mm.
in breadth; the smallest East African (Guaso Nyiro) terrapin which
I have seen had a carapace length of 34 mm., breadth 27 mm. and
height of 11 mm.

Longevity. 16 years, 10 months, 11 days (Flower, 1925b, p. 933).
Diet. Spiders, grasshoppers — one of which was warningly coloured
in black and yellow, stink ants (Paltothyreus tarsatus) were readily
taken, the latter being snapped in half; the head and thorax crunched
first, then the abdomen (Loveridge). Earthworms and mealworms
(Boettger).

loveridge: African pelomedusidae 477

According to Rose (1929, p. 186) large prey is not killed and eaten
quickly, these terrapin having an unpleasant habit of seizing a hapless
frog by a hind foot, then, with forward thrusts of the claws, shredding
the unfortunate creature's limb to tatters. In captivity they ate
chopped meat, tadpoles, fish, crabs, snails and even weeds.

W. Cloete (in Hewitt, 1937e, p. 12) adds prickly-pear fruit to their
dietary, saying: "when the fruit dropped they visited the tree and
carried the fruit into the water to eat." They are partly nocturnal,
being a nuisance to eel fishermen by taking their bait. After prolonged
drought marsh terrapin become thin and emaciated and according to
Bowker (in Hewitt) will catch small ducklings and goslings or congre-
gate to feed upon a dead sheep should one fall into the dam where they
live.

Enemies. Marsh terrapin are eaten by natives at Quissange, Angola,
also by the Hereros. They are dug from their retreats by ratels and
preyed upon by jackals and eagles, the latter tearing out the hinder
part of the plastron of small examples according to Bradfield (in
Hewitt, 1935, p. 340). Angel (1922a, p. 39) records the recovery of a
small terrapin, under 40 mm. in length, from the stomach of a sea
eagle (Haliaetus vorifer) in the French Sudan.

Defence. When seized it defends itself effectively by squirting the
contents of its cloaca at its assailant (Boettger, 1881b, p. 410). Apart
from their offensive musky odour, emanating from the secreting glands
by four minute pores close to the carapace opposite the fourth and
eighth marginal shields, these terrapin chiefly rely on withdrawing the
head and limbs within the shell, though the protection afforded is less
effective than is the case with the other species of African terrapin and
tortoises. The long neck necessitates the head lying sidewise, from
which position the fishy eye of the reptile continues to appraise the
extent of the danger.

Aestivation. When marshes and ponds commence drying out at the
onset of the dry season, these terrapin dig themselves into the mud by
an outward and upward movement of the hind limbs. Buxton (1937,
p. 102) thinks that in the prolonged droughts of Turkanaland they
must remain buried deeply in the sandy water-courses for years at a
time. At Voi, on April 22, I encountered many in such a water-course
after a heavy downpour, one of several inaugurating the arrival of the
big rains. At Dodoma, during July, a hundred of these terrapin which
I had been keeping in a tank, either from a sense of overcrowding —
though food was abundant, or warned by some seasonal instinct —
for the rains had ended six weeks before, insisted with one accord in

478 bulletin: museum of comparative zoology

clambering out and piling themselves against the wire netting sur-
rounding their inclosure. Returning them to their tank proved futile
so for a week they preferred to stay exposed to the cold winds which
rose at night, until I removed them indoors and packed them in crates
of straw. At Mtali's, near Mkalama, on October 20, after heavy rain
storms on the two preceding days, one of these terrapin was encoun-
tered at 8.15 a.m. endeavouring to dig itself into the middle of a hard
though sandy road. At Mangasini, on December 12, an initial down-
pour of the small rains commenced at 6 p.m. and lasted till noon on the
following day. This resulted in the emergence of many aestivating
terrapin, numbers of which I captured in a dry water course that had
held a torrent only the night before.

In Somaliland marsh terrapin aestivate from January to March and
from July to September (Gadow, 1901, p. 391). In South Africa both
aestivation and hibernation takes place for, should the dams and
vleis dry up, they are said to leave them to bury, reemerging after
heavy thunderstorms.

Hibernation. During the South African winter, marsh terrapin
usually leave the water from May to August to bury themselves in soft
soil or among dead leaves beneath trees, but at Skietop, near Sidbury,
with a surface temperature of from 52° to 58° in June, they remain in
the dams throughout the year. (Hewitt, 1937e, p. 11.) Fleck (1894,
p. 83) refers to digging them from the mud in winter in South- West
Africa.

Habits. In the South African summer they like to sun at the water's
edge or, on occasion, to climb the branches of overhanging trees so as
to bask a couple of feet above the water, into which they retreat at the
slightest alarm, their sharp eyesight enabling them to detect an in-
truder at some distance (Hudson, in Hewitt, 1937e, p. 12). In the
tropics basking is rarely indulged in by these terrapin, though I have
occasionally seen one sunning in the early morning at the edge of a
water hole.

Habitat. Though quite at home and active on land, the marsh
terrapin may be seen to best advantage in water where it swims and
dives with confident agility, remaining below for long periods without
coming to the surface to breathe. Owing to their ability to exist in
swamps and small bodies of water from the coastal plain to the upland
savannas at 6,000 feet, these terrapin have a distinct advantage over
the so-called soft-shelled turtles whose distribution is restricted to the
lakes and larger rivers.

loveridge: African pelomedusidae 479

Localities. Anglo-Egyptian Sudan: El Obeid; Gamilab Hills s.
of Suakin; Gebel Moya near Sennar; Khoi Arbat near Port Sudan;
Wad Medani; White Nile between Fashoda and Renk. Ethiopia:
Abdallah; Ela Gura; Gonda; west of Juba River; Meddo Erelle; Oda*;
Sabarguma; Sereba Ghattas. British Somaliland: Buran dist.;
Erigavo dist. Italian Somaliland: Boran; Djugi; Giblene; Stagno
Saha.* Uganda: Bussu; Kabulamuliro. Kenya Colony: Bulessa;
Guaso Nyiro; Kapiti Plains; Kind's in Kitosh; Lukenya Hills; Moyale;
Nairobi; Sotik; Voi. Tanganyika Territory: Dodoma; Iringa;
Kasanga; Kikamero; Kilimatinde; Lake Eyasi to Isanssu; Lake Vic-
toria ; Luguo ; Mahaka ; Mangasini ; Masai Ny ika ; Mbulu ; Mtali's near
Mkalama ; Mtita's near Dodoma ; Mukwese ; Ngare na Nyuki ; Nzinga ;
Pumbo near Mondo; Ruaha River; Tabora; Tanga; Ukerewe Island.
Mozambique: Lumbo; Quelimane; Querimba; Tete. Nyasaland:
Livingstonia. Northern Rhodesia: (fide Pitman). Southern
Rhodesia: Sabi River at Birchenough Bridge. Bechuanaland
Protectorate: Gomodino Pan; Gomodimo to Koke; Kuruman
Lobatsi; Metsimaklaba River; Nkate Pan ;Zweizwe River. Transvaal
Junction of Comati and Crocodile Rivers; Naawpoort1 (M.C.Z.)
Tuefloop on north slope Drakensburg. Natal: Uzumbwe Valley
Winkle Spruit. Orange Free State: Emmaus; Thabanchu. Cape
Province: Besondermeid near Steinkop; Bushman's River; Cape
Peninsula; Deelfontein; Gleniffer near Kei Road; Graff Reinet; near
Grahamstown; Great Fish River; Kimberly; Malmsbury; Mortimer;
Port St. Johns (M.C.Z.) ; Queenstown; Skietkop near Sidbury; Sun-
day's River; Warrenton. South- West Africa: Aroab (M.C.Z.);
Aus; Chamis; Gochaganas S. of Windliuk; Hoffnung; Mookane;
Namutoni; Oas; Okonjati; Possession Island,1 Quickborn near Oka-
handja; Rehoboth; Rietmond; Windhuk. Angola: Capangombe,1
Catumbila; Chitau; Duque de Braganca; Gauca; Humbe; Kahuihui,
Kalukembe; Kalundunga; Kuvangu River S. of Vila da Ponte;
Maconja; Mossamedes; Mucungu; Quilengu; Quisangue. Belgian
Congo: Fucafuca, Yellala Falls, Congo River; Kikamero on Ruchuru
plains; Manda; Mahagi Port; Moirbuttu; Tembwe. . Madagascar:
(many localities) .

Range. The typical race ranges outside the rainforest from the
Cape to Natal, northwards to Somaliland and the Anglo-Egyptian
Sudan where intermediates with the northern race occur over a wide
belt from Somaliland through northern Kenya and Uganda to Senegal.

1 Individuals with pectorals separated occur here also.

4S0 bulletin: museum of comparative zoology

Without the continent it occurs in Madagascar, but the Sinai record
(Boulenger, 1889a, p. 197) is questioned by Flower (1933, p. 752) who
quotes Boulenger as saying that the old entry in the British Museum
register on which it was based "may well be erroneous."

Folklore. The odour of the marsh terrapin, according to Kalahari
natives, resembles that of a lion and will stampede cattle (FitzSimons,
1935b, p. 307).

Pelomedusa subrufa olivacea (Schweigger)

1814. Emys olivacea Schweigger, Prodromi mon. Chelon., p. 38: "In Fabulosis

Nigritae" Adanson coll. = Senegal.
1835. Pentonyx Gehafie Riippell, Neue Wirbelth. Fauna Abyss., Amph., p. 2,

pi. i: Massaua, Eritrea.

1851. Dumeril & Dumeril, p. 18.

1852. A. Dumeril, p. 245.

1860. A. Dumeril, p. 163, pi. xiii, fig. 4.

1922a. Mertens, p. 169.

1844. Pelomedusa gehafice Gray, p. 38.

1855. Gray, p. 53.

1865. Strauch, p. 113.

1862b. Peters, p. 271.

1870. Blanford, p. 444.

1870. Gray, p. 81.

1871. Sclater, p. 325, fig. (Upper Zambezi)
1873b. Gray, p. 71.

1884a. Rochebrune, p. 22.

1935. Hewitt, p. 325.

1884a. Pelomedusa Gasconi Rochebrune, Faune Senegambie, Rept., p. 25, pi.
i, figs. 1-2: Dagana, Senegal (restricted). No type preserved.1

1910. Pelomedusa galeata var. disjuncta Vaillant & Grandidier, Hist. phys.
nat. pol. Madagascar, 17, Rept., p. 56, pi. xx, fig. 3. Du Bourg de
Bozas coll. = Shore of Lake Abaya, Sidamo, Ethiopia. 1

1936e. Pelomedusa galeata gehafie Parker, p. 609.

1937a. Pelomedusa subrufa gehafie Mertens, p. 140.

In addition to the above are many citations listed under the typical
form, most of them being indicated by an asterisk before the name.

Names. Gehafie (at Massaua); nguru (Kitosh); njaba (Ganda).

History. As the type of olivacea consisted of a carapace, it is pure
assumption, based on the fact that gasconi of Senegal has the pectorals
separated, that the name is employed. The race, apparently the only

1 Data kindly supplied from Paris Museum catalogue by Mons. Angel, though the specimen
could not be located at the moment.

loveridge: African pelomedusidae 481

one found in Eritrea, occurs alongside typical subrufa at Oda, Ethiopia
and Stagno Saha, Italian Somaliland. Individuals with separated
pectorals crop up again on Possession Island, South-West Africa;
Capangombe, Angola; Zambezi River and Madagascar, being taken
with typical subrufa. Though a poor race it should be recognised,
perhaps, on the grounds that it is pure in Eritrea.

Description. Differs only from the typical form in having the
pectoral shields widely separated on the median line of the plastron.

Localities. Anglo-Egyptian Sudan: Sennar. Eritrea: Anseba
River; near Cheren; Lebka River; Massau; Sabarguma. Ethiopia:
Lake Abaya; Oda. Italian Somaliland: Stagno Saha. Kenya
Colony: Kaliokwell River; Kind's in Kitosh Uganda: Mabira
Forest; Mt. Elgon. Nigeria: Keana.

While the following are tentatively referred to this race. French
Cameroon: Bipindi (?det.). Togoland: Mangu. Gold Coast:
French West Africa. Senegal: Dagana; Rufisque (Rochebrune's
other localities omitted).

Range. As indicated above, the limits of distribution of the race
olivacea are still uncertain, it may be regarded as the drier regions of
a belt extending from Senegal to Eritrea, intergrading with the
typical form in the Anglo-Egyptian Sudan, Ethiopia, Somaliland,
northern Kenya and Uganda.

Genus Pelusios

1825. Sternothaerus Bell (part), (not Sternotherus Gray, 1825, p. 211: type

odoratus), p. 305.
1830. Pelusios Wagler, Nat. Syst. Amphib., p. 137 (type subniger).
1863. Tanoa Gray, Proc. Zool. Soc. London, p. 193 (type sinuatus).
1863. Notoa Gray, Proc. Zool. Soc. London, p. 195 (type castaneus, subniger).

Skull without supratemporal roof; quadra to-jugal widely separated
from parietal ; upper jaw with very indistinct median ridge on alveolar
surface; between the orbits are a pair of supraorbital shields separated
by a longitudinal suture and followed by a large frontal flanked by
temporals; plastron large, a hinge between hyoplastra and hypoplastra
permits closing of front lobe in adults; mesoplastral bones extending
right across plastron; digits very short, the median with 3 phalanges;
feet with 5 claws.

Range. Africa south of the palaearctic zone; Madagascar and
islands of the Indian Ocean.

482 bulletin: museum of comparative zoology

Key to the Species

1. Suture between abdominals included more than twice in length of anterior

lobe of plastron 2.

Suture between abdominals included less than twice in length of anterior
lobe of plastron 3.

2. Crown of head vermiculated; no black vertebral line extending entire length

of carapace ; suture between humerals 3 to 4 times as long as that between
the pectorals adansonii

(p. 483)
Crown of head uniform except for a black Y-shaped marking uniting orbits
posteriorly with nape ; a black vertebral line extends entire length of cara-
pace ; suture between humerals 1 }4 to 2 times as long as that between the
pectorals gabonensis

(p. 485)

3. Plastron entirely black, entirely yellow, or yellow with darker infuscations,

the latter not forming a sharply defined angular pattern round the pe-
riphery (in some Transvaal and Malagasy terrapin the pattern of sinuatus
is foreshadowed but not sharply defined); its anterior lobe always longer
than the suture between the abdominals; posterior margin of carapace
usually rounded; narrowest interorbital width equal to, or slightly longer,
or slightly shorter than the suture between the supraorbitals . . subniger.

(p. 489)
Plastron yellow (white) broadly edged with black, the latter forming a
sharply defined angular pattern round the periphery (sometimes blurred
or masked in old individuals with a carapace length of 250 mm. or more);
its anterior lobe longer than the suture between the abdominals (equal to
or slightly shorter in old specimens) ; posterior margin of carapace sharply
serrate except in very old terrapin ; narrowest interorbital width only
two-thirds the length of the suture between the supraorbitals, .sinuatus1

(p. 502)

I had fully anticipated that subniger would be separable into a
number of local forms, but though weeks were spent on this one species
I have failed entirely to find any characters of colour or proportions
which offer a reasonable prospect of separating alleged races. Appar-
ently striking differences invariably turn out to be those of sex or age,
while those of colour, constant in one locality, crop up here and there
without obvious geographical significance.

My method was to draw up a detailed description based solely on 34
terrapin from Kaimosi in Kenya, then a similar description based on

1 The two specimens from Port Natal, referred to sinualus by Boulenger, agree with that
species in having the anterior lobe shorter than the suture between the abdominals (they are
old, being over 200 mm.), but Mr. Parker informs me that the female's plastron is entirely black,
that of the male dark brown with the central area somewhat lighter.

loveridge: African pelomedusidae 483

12 specimens from the Seychelles, and so on. When these descriptions
were compared no character of consequence could be detected. Such as
appeared from the wording, were rechecked by direct comparison of
the specimens and usually found to have no significance. Yet the two
series cited were readily separable on their respective wholly black
or wholly yellow plastrons; but such a colour difference could not be
correlated with distributional areas, and so many variants of an inter-
mediate pigmentation crop up here and there that even recognition
of an average difference appeared unjustifiable.

I can only hope that enthusiasts for naming local forms will pause
to accumulate big series before embarking on the description of
further races for which single individuals may offer apparent justifica-
tion. Comparisons, except between individuals of the same sex and age,
lead only to confusion.

Pelusios adansonii (Schweigger)

1814. Emys Adansonii Schweigger, Prodromi mon. Chelon., p. 39:

"Nigritis." i.e. Cape Verde, Senegal [vide Dum. & Bib., 1835).

1831. Hydraspis Adansonii Gray, p. 40.

1835. Pentonyx Andansonii (sic) Dumeril & Bibron, p. 394.

1844. Pelomedusa? Adansonii Gray, p. 38.

1851. Sternothaerus Adansonii Dumeril & Dumeril, p. 19.

1855. Gray, p. 52.

1864b. Gray, p. 296, pi. xxiii.

1865. Strauch, p. 109.

1870. Gray, p. 80.

1873b. Gray, p. 70.

1884a. Rochebrune, p. 21.

1889a. Boulenger, p. 196.

1890. Biittikofer, p. 436.

1900. Flower, p. 967.

1903d. Siebenrock, p. 197.

1904. Andersson, p. 9.

1906a. Siebenrock, p. 826.

1908. Werner (1907), p. 1826.

1909. Siebenrock, p. 560.

1910. Miiller, p. 623.

1912b. Werner, in Brehm, p. 470.

1917. Sternfeld, p. 414.

1919. Schmidt, p. 600.

1924a. Warner, p. 268.

1925b. Flower, p. 933.

1934a. Pelusios adansoni Mertens, p. 10.

484 bulletin: museum of comparative zoology

Names. Adanson's Terrapin.

Illustrations. Gray (1864b, pi. xxiii) furnishes a drawing of
the upper aspect of this terrapin in life.

History. The type is in the Paris Museum. Boeage (1867a, p. 217)
recorded adansonii from Angola, later correcting his identification to
derbianus, now a synonym of svbniger.

Description. Head broad, snout short; a pair of supraorbital shields
followed by a very large frontal, flanked by a pair of temporals; nar-
rowest interorbital width equals the longitudinal suture between the
supraorbitals ; upper jaw neither hooked nor bicuspid ; chin with a pair
of elongate barbels; scales on anterior aspect of fore limb irregular in
size; carapace moderately depressed, its height included in its length
about 2.64 times, broadening posteriorly, its posterior margin rounded;
vertebral shields 5, the anterior 4 keeled throughout life1, broader than
long in young, nearly as long as broad in adult, first and fifth subequal
in length and breadth, as broad (at their broadest) as long; costals
4 pairs; marginals 22; supracaudals 2; plastron considerably smaller
than the opening of the carapace; anterior lobe somewhat rounded,
not or but slightly projecting beyond the carapace; posterior lobe
angularly and deeply notched; intergular 1^ to 2 times as long as a
gular, 2 times as long as broad, its sides wedge-shaped; numerals
forming a suture 3 to 4 times as long as that of the pectorals, outer
border of a humeral equal to, or shorter than, that of a pectoral;
pectorals not excluded from bridge by abdominals; width of bridge
contained 2 (young) times in the width of the plastron ; suture between
abdominals less than half the length of the anterior lobe of plastron.

This description is based on a Shari River juvenile in the Museum
of Comparative Zoology, together with descriptions in the literature.

Anatomy. An abnormal Sennar shell in which the anterior costal
bones are in contact, thus separating nuchal and first neural, has been
described by Werner (1924a, p. 268).

Coloration. Above, head yellow vermiculated with brown; carapace
yellowish to pale brown with radiating brown lines and dots. Below,
labial region, plastron, and all lower parts yellowish.

Werner (1924a, p. 268), however, reports on a 160 mm. male from
Tonga as being shining black both above and below except along the
sutures of the plastron which are white and worn.

Measurement. Length of carapace of a Sennar specimen 185 mm.,
breadth 125 mm. (Werner, 1924a, p. 268). •

1 fide Boulenger.

loveridge: African pelomedusidae 485

Breeding. In March, at Tonga, White Nile, 7 eggs measuring 29.5
x 18 and 33 x 19 mm., were found in a 9 by Werner (1924a, p. 268).

Longevity. 9 years, 8 months, and still alive at Giza Zoological
Gardens (Flower, 1925b, p. 933).

Habitat. Rivers outside the rainforest. Siebenrock (1906a, p. 826)
remarks on the fact that this species ranges to the edge of the palae-
arctic region and is thus the most northerly member of its genus. His
suggestion that the deserts of Darfur and Kordofan have interrupted
its distribution is unlikely, the answer is more probably to be found
in the very little collecting which has been done in this area.

Localities. Anglo-Egyptian Sudan : Bahr el Gebel near Mongalla ;
Bahr el Ghazal; Bahr el Zeraf; Gondokoro; Khor Attar; Sennar;
Tonga; White Nile south of Abu Zeit. French Cameroon: Guffei on
Shari River; Tara on Isade River. Senegal: Cape Verde. Liberia.

Range. Senegal east to the White Nile (from Abu Zeit south to
Gondokoro), Anglo-Egyptian Sudan. Reported from Liberia.

Pelusios gabonensis (A. Dumeril)

1856. Pentonyx Gabonensis A. Dumeril, Rev. Mag. Zool., 8, p. 373: Gabon.

1860. A. Dumeril, p. 164, pi. xiii, figs. 2, 2a.

1864a. Gray, p. 168.

1874. Reichenow, p. 298.

1862. Pelomednsa gabonensis Strauch, p. 45.

1865. Strauch, pp. 107, 113.

1864. Pelomedusa gabonica Peters, p. 644.

1873a. Sternothaerus sp. Gray, p. 393, fig.

1873b. Sternothaerus derbianus Gray (part), p. 69.

1876a. Peters, p. 117.

1875a. Sternothaerus niger Peters (not of Dumeril & Bibron), p. 198.

1889a. Boulenger, p. 194, fig. 46.

1890. Biittikofer, p. 478.

1893c. Matschie, p. 208.

1897. Sjostedt, p. 33.

1898. Werner, p. 204.
1902c. Tornier, p. 665.
1903d. Siebenrock, p. 191.
1905a. Siebenrock, p. 461.
1906. Johnston, pp. 820, 833.
1906a. Mocquard, p. 480.

1909. Siebenrock, p. 555.

1910. Miiller, p. 622.

1910b. Nieden, p. 7 (text apparently transposed with that of galeata)

486 bulletin: museum of comparative zoology

1911. Lampe, p. 148.

1925b. Flower, p. 932.
1889a. Sternothaerus gabonensis Boulenger, p. 197.

1897. Sjostedt, p. 33.

1898. Werner, p. 204.
1900b. Boulenger, p. 447.
1901. Siebenrock, p. 7.
1902c. Tornier, p. 665.
1903d. Siebenrock, p. 197.
1905a. Siebenrock, p. 461.
1906a. Mocquard, p. 480.
1907. Siebenrock, p. 6.
1909. Siebenrock, p. 560.

1916. Siebenrock, p. 10, pi. i, fig. 1, pi. ii, fig. 4.

1919g. Boulenger, p. 12.

1933m. Witte, p. 67.

1901. Sternothaerus Steindachneri Siebenrock, Zool. Anz., 25, p. 6:

"Madagascar" Bought from a dealer.
1919. Pelusios gabonensis Schmidt, pp. 413, 598, fig. 1.
1934a. Mertens, p. 10.
1937c. Loveridge, p. 269.

1919. Pelusios niger Schmidt (not of Dumeril & Bibron), pp. 598, 600.
1934a. Mertens, p. 10.
1934. Muller, p. 166.
1937a. Flower, p. 14.
1938b. Mertens, p. 33.

1924. Sternothaerus heinrothi Kanberg, Zool. Anz., 60, p. 195, fig.: Cameroon.
1926. Kanberg, p. 225.

Erroneous records of gabonensis (Bocage, 1866a, lS66b, and Roche-
brune, 1884a) will be found under subniger.

Names. Gaboon Terrapin.

Illustrations. Siebenrock (1916, pis. i-ii) furnishes excellent black
and white figures of an adult from above and below; Dumeril's (1860,
pi. xiii) figures of the juvenile type do not show its distinguishing
characteristics; Schmidt (1919, fig. 1) an outline drawing of a juvenile
from below, nobody has figured the handsome young or middle-aged
stage with its black vertebral line on a yellowish brown carapace.

History. The tangled synonymy of this species may be said to have
commenced when Gray (1873a, p. 393) wrote "we have fortunately
discovered a very large skull, evidently belonging to the genus Sterno-
thaerus, which M. du Chaillu had used (as he did the new species of

loveridge: African pelomedusidae 487

Bush-buck which I described . . . ) to stuff out the skin of a large
African mammal." Later in the year Gray referred this skull to
derbianus (which I regard as a synonym of svhniger). Because it agreed
with Dumeril & Bibron's description of nigcr (1835), however, Boul-
enger (18S9a, p. 194) identified and figured it as that species.

The pertinent portion of the description "museau allonge; machoire
superieure se recourbant en bee croehu;" is rather one of age in mem-
bers of this family. It is particularly pronounced in Du Chaillu's
specimen, obviously from an aged animal, as Gray states that it
measured 2 }/q inches from cheek to cheek, 2^ inches from nose to
condyle. The largest gabonensis in the Museum of Comparative
Zoology, an alcoholic with a carapace length of 245 mm., has a head
measuring 2 inches from cheek to cheek, its skull length cannot be
measured.

Since Boulenger, there has been a tendency to refer all old black
Cameroon terrapin to nigcr, all half-grown or young with characteristic
vertebral stripe to gabonensis though the evolution of colouring of the
carapace is clearly seen in a good series of all ages. Siebenrock even
invokes the age character of colour as being the only reliable means of
separating gabonensis from what he calls niger, i. e. the black throat
and underparts of the young which become dirty yellow in older
reptiles.

I reject niger as being applicable to the West African gabonensis, not
only because Dumeril & Bibron thought that it came from Madagas-
car, which may have been the case, but because they state that the top
of the head in niger is marbled with brown on a fawn ground, and the
jaws are horn coloured with vertical rays of a maroon tint. This is the
coloring of sub?iiger, not of gabonensis, which has a black Y-shaped
marking on the occiput and the jaws uniformly coloured.

Siebenrock (1903d, p. 197) himself referred steindachneri to the
synonymy of gabonensis after examining the type of the latter in the
Paris Museum.

Miiller (1934, p. 166) produced overwhelming proof for placing
heinrothi in the synonymy of "niger," by which he meant gabonensis
as here understood.

Description. Head broad, snout only moderately short, acuminate;
a pair of supraorbital shields followed by a very large frontal, flanked
by a pair of temporals; narrowest interorbital width equals or is shorter
than the longitudinal suture between the supraorbitals; upper jaw
angularly rounded in young, slightly notched and bicuspid in adults,
hooked in very old specimens; chin with a pair of barbels which are

488 bulletin: museum of comparative zoology

long in young and short in adults; scales on anterior aspect of fore limb
irregular in size; carapace depressed, its height included in its length
2 34 to 3.76 times, elongate, its posterior margin rounded, not or but
slightly serrate even in young; vertebral shields 5, all keeled in young,
the rather nodose keels of the posterior 3 persisting except in very old
individuals, all 5 broader than long in young and adults though oc-
casionally the fourth and fifth may be longer than broad in adults;
costals 4 pairs; marginals 22; supracaudals 2; plastron much smaller
than the opening of the carapace; anterior lobe rounded or intergular
slightly projecting, usually not, though occasionally slightly projecting
beyond the carapace; posterior lobe more or less angularly and deeply
notched; intergular normally 2, rarely 3, times as long as a gular, 1}/%
to \}/2 times as long as broad, its sides straight, wedge-shaped, or
pyriform; humerals forming a suture 13^ to 2 times as long as that of
the pectorals outer border of a humeral longer than, rarely equal to,
that of a pectoral ; pectorals not quite excluded from bridge by abdom-
inals; width of bridge contained \x/i (adults) to 2 (young) times in the
width of the plastron ; suture between abdominals included 2^4 to V/2
times in the length of the anterior lobe of plastron.

Anatomy. The skull of an old terrapin has been figured and des-
cribed at length by Gray (1873a, p. 393).

Coloration. Above, head yellow brown, a broad black Y-shaped
marking connects the eyes and extends backwards on the nape,
another stripe between eye and tympanum is sometimes present;
limbs black in young, drab buff or greyish yellow in adults; carapace
yellow brown in young darkening with age through the appearance of
dark radial lines on the discoidal shields till almost black, in old indi-
viduals, a black vertebral streak, broadening anteriorly on the suture
between the anterior marginals, extends to the supracaudal suture,
this characteristic marking is less conspicuous in very old terrapin
but can be found if sought. Below, jaws and throat uniform horn
colour except in young when the throat is black; plastron uniform
black or with the sutures between the shields narrowly edged with
yellow; occasionally plastron yellowish brown, each shield so heavily
overlaid with radiating black markings as to appear mostly black,
carapace light yellow, each marginal blotched with black, particularly
in the region of the bridge.

Measurements. Length of carapace of a c? from Edea, 270 mm.,
breadth 167 mm., height 88 mm. (Miiller, 1910). Length of carapace
of a 9 from Isongo, 258 mm. (Mertens, 1938b).

Sexual dimorphism. Males, recognizable by their longer tails, ex-

loveridge: African pelomedusidae 489

hibit a slight depression in the posterior third of the plastron, which is
flat in females.

Longevity. 7 years, 4 months, 20 days (Flower, 1937a, p. 14) is sur-
passed by one stated to be 10 years and still alive by Midler (1934,
p. 166).

Diet. In captivity, water insects, ant pupae, worms, snails, fish and
raw beef should be furnished in variety, particularly for young, at tem-
perature of 18° to 20° C.

Parasites. Leeches were about the hind limbs of a terrapin taken
from the Lepoko River (Lang, in Schmidt, 1919, p. 414).

Enemies. Eaten by the Congolese (Lang, in Schmidt, 1919, p. 414).

Habitat. Young and half grown are to be found in swampy places
and smaller streams, but large examples must be sought for in the
rivers of the rainforest belt, where they are frequently drowned in fish
traps (Lang, in Schmidt, 1919, p. 414).

Localities. Cabinda. Belgian Congo: Akenge; Avakubi; Buta;
D jamba; Gamangui; Ituri River west of Ruwenzori Mountains;
Medje; Nepoko River; Niapu; Nyonga; Pawa; Poko; Saidi's Village
on Avakubi-Irumu road; Stanleyville. French Congo: Gaboon.
French Equatorial Africa: Nola. Spanish Guinea. French
Cameroon: Ebolowa; Isongo; Jaunde; Kribi; Metet; Sakbayeme;
Sanaga River near Edea; Sangmelina. Nigeria: Benin. Gold Coast:
Akusi. Liberia (thrice reported, but without locality.)

Range. West Africa from Liberia south to Cabinda, eastwards
through the Belgian Congo to the Ituri River.

Pelusios subniger (Lacepede)

1788. La Noiratre Lacepede, Hist. nat. Quadrup. ovip. Serpens, 1, p. 175.

pi. xiii: No locality.

1789. Testudo subnigra Lacepede, Hist. nat. Quadrup. ovip. Serpens, 2,

Synopsis niethodica (a table in which binomials are employed).
1789. Bonnaterre, p. 30, fig. 6.
1802. Daudin, p. 197.
1798. Testudo Nigricans Donndorff, Zool. Beytr. Linn. Natur., 3, p. 34:

No locality.
1814. Emys castanea Schweigger, Prodromi mon. Chelon., p. 45: No locality.
1814. Emys subnigra Schweigger, p. 46.
1820. Terrapene nigricans Merrem, p. 28.
1825. Kinosternon nigricans Bell, p. 305.

1825. Sternothaerus Leachianus Bell, Zool. Journ., 2, p. 306: No locality.
1831. Sternotherus subniger Gray, p. 38.

490

bulletin: museum of comparative zoology

1844.

1855.

1863b.

1864b.

1870.

1873b.

1831.

1835.

1865.

1866b.

1869a.

1878a.

1880b.

1881c.

1884a.

1891.

1910.

1931d.

1835.

1848.

1862.

1865.

1867a,

1877b.

1882a.

1884a.

1889.

1889a.

1891.

1893a.

1893b.

1895.

1896a.

1896.

1897.

1898.

1898a.

1900b.

1903a.

1903d.

1907a.

Gray, p. 37.

Gray, p. 51.

Gray, p. 195, fig.

Gray, p. 168, fig.

Gray, p. 79.

Gray, p. 70.
lStemotherus castaneus Gray, p. 38.

Dumeril & Bibron, p. 401.

Strauch, p. 108.

Peters, p. 887.

Peters, p. 12.

Peters, p. 202.

Peters, p. 509.

Vaillant, p. 797.

Boettger, p. 535.

Rochebrune, p. 19.

Vaillant, p. 94.

Vaillant & Grandidier, pp. 26, 58, pis. xviii-xix.

Angel, p. 550.
Sternotherus Niger Dumeril & Bibron, p. 397.
Stemotherus Nigricans Dumeril & Bibron, p. 399.

Peters, p. 494, pi. xvii, fig. 6.

Strauch, p. 148.

Strauch, p. 108.

Steindachner, p. 6.

Peters, p. 455.

Peters, p. 8.

Rochebrune, p. 19.

Boettger, p. 297.

Boulenger, p. 195.

Vaillant, p. 94.

Boettger, p. 13.

Stejneger, p. 713.

Rathgen, p. 200.

Bocage, p. 97.

Tornier, p. 4.

Tornier, p. 63.

Tornier, p. 282, fig. 2.

Vaillant, p. 133.

Tornier (part), p. 582.

Siebenrock, p. 254.

Siebenrock, p. 195.

Boulenger, p. 6.

i The spelling Siernothaerus has not been separated for separate headings.

loveridge: African pelomedusidae 491

1907J. Boulenger, p. 482.

1908. Chubb, p. 220.

1908. Rembold, p. 743, figs. 2-3.

1909h. Boulenger, p. 295.

1909a. Chubb, p. 592.

1909b. Chubb, p. 34.

191Qa. Andersson, p. 11.

1911b. Masi, p. 132.

1911. Sternfeld & Nieden, p. 385.

1912c. Sternfeld, p. 201.

1912b. Werner, p. 469.

1913c. Nieden, p. 59.

1916. Siebenrock, p. 6, pi. i, fig. 2, pi. ii, fig. 5.

1923g. Loveridge (part), pp. 930, 932.

1924b. Loveridge, p. 2.

1925b. Flower, p. 933.

1928. Cott, p. 952.

1931. Monard, p. 109.

1933m. Witte, p. 67.

1937b. Monard, pp. 146, 148.

1939a. Rendahl, pp. 304, 322, figs. 11-12.

1844. Sternotherus Derbianus Gray, Cat. Tortoises Brit. Mus., p. 37:

Gambia (restricted).

1855. Gray, p. 52, pi. xxii.

1863b. Gray, p. 194.

1864a. Gray, p. 167.

1865. Strauch, p. 109.

1866a. Bocage, p. 41.

1866b. Bocage, p. 57.

1867a. Bocage, p. 218.

1870. Gray (part), p. 79.

1873b. Gray (part), p. 69.

1877c. Peters, p. 611.

1881b. Boettger, p. 409.

1884. Greeff, p. 48.

1884a. Rochebrune, p. 20.

1886a. Bocage, p. 66.

1888a. Boettger, p. 15.

1889a. Boulenger, p. 195.

1889. Hesse, p. 262.

1890. Buttikofer, pp. 436, 478.
1890a. Muller, p. 296.

1890. Strauch, p. 102.

1892c. Bocage, p. 230.

1893a. Boettger, p. 13.

492 bulletin: museum of comparative zoology

1893c. Matschie, p. 208.

1895a. Bocage, p. 3.

1896a. Bocage, p. 74.

1897. Sjostedt, p. 7.

1898. Jeude, p. 9.
1898. Werner, p. 204.
1900b. Boulenger, p. 447.
1901. Gadow, p. 391.
1901c. Tornier, p. 67.
1902c. Tornier, p. 665.
1903d. Siebenrock, p. 196.

1905. Bocage, p. 90.
19061. Boulenger, p. 197.

1906. Johnston, pp. 820, 833.
1906a. Mocquard, p. 480.

1907. Johnson, pp. 14, 69, photo.

1908. Rembold, p. 742, fig. 1.

1909. Siebenrock, p. 559.
1911c. Boulenger, p. 162.
1911. Lampe, p. 148.
1912b. Werner, p. 470.
1919g. Boulenger, p. 12.
1919. Schmidt, pp. 598, 600.
1921a. Chabanaud, p. 461.
1921b. Chabanaud, p. 522.
1923g. Loveridge, p. 932.
1924b. Loveridge, p. 2.
1925b. Flower, p. 933.

1928. Cott, p. 952.

1930b. Witte, p. 84.

1933. Schmidt, p. 3.
1933m. Witte, p. 67.

1934a. Mertens, p. 10.

1936. Frade, p. 67, pis. vii-viii.

1937. Andersson, p. 2.

1851? Sternotherus nigricans ? var. Bianconi, p. 58, nigrescens, pi. vii. (a mis-
print as Sthernotherus nigrescens Dum. Bib. also given).

1866a. Sternothaerus gabonensis Bocage (non Dumeril), p. 40.

1866b. Bocage, p. 57.

1884a. Rochebrune, p. 23.

1867a. Sternothaerus Adansoni Bocage (non Schweigger), p. 217.
1881c. Sternothaerus subniger Boettger, p. 535.

1910. Vaillant & Grandidier, p. 26, pis. xviii-xix.

1884a. Sternothaerus niger Rochebrune (non Dumeril & Bibron), p. 19.
1884a. Sternothaerus sinuatus Rochebrune (non Smith), p. 20.

loveridge: African pelomedusidae 493

1889. Boettger, pp. 296-7.

1889a. Boulenger (part), p. 194.

1893a. Boettger, p. 13.

1895a. Bocage, p. 4.

1896. Tornier (part), p. 4.

1902d. Boulenger, p. 445.

1909b. Boulenger, p. 302.

1909h. Boulenger, p. 295.

1910. Roux, p. 100. (photographs of specimen seen).

1911. Lampe, p. 148.

1912c. Sternfeld (part), p. 200.

1937b. Monarch pp. 146, 148.

1885d. Sternothaerus sp. Miiller, p. 716.

1909. Gendre, 1909, p. cvi.

1897f. Sternothaerus oxyrhinus Boulenger, Proc. Zooh Soc. London, p. 919, pi.

liii: No locality. (Based on live specimen in zoo.)
1906b. Sternothaerus nigricans castaneus Siebenrock, p. 35, pi. v, fig. 18.
1909. Siebenrock, p. 557.
1913. Boettger, pp. 318, 352.
1915. Rawitz, p. 658, pi. xlviii, figs. 62-67.

1918. Barbour, p. 489.

1922. Kaudern, p. 449, fig. D.

1906b. Sternothaerus nigricans nigricans Siebenrock, pp. 36, 40, pi. v, fig. 19.

1909. Siebenrock, p. 558.

1913. Boettger, p. 319.

1915. Rawitz, p. 663.

1922. Kaudern, p. 449.

1939b. Rendahl, p. 3.

1906b. Sternothaerus nigricans seychellensis Siebenrock in Voeltzkow, Reise in
Ostafrika in den Jahren 1903-1905, 2, p. 38: Gloriosa Island, Sey-
chelles.

1909. Siebenrock, p. 558.

1912c. Pelotnedusa galeata Sternfeld (part), p. 201.

1919. Pelusios nigricans Schmidt, pp. 411, 460, fig. 1, pi. xi, fig. 3.
1928c. Barbour & Loveridge, p. 104.

1934. Pitman, p. 307.

1927a. Pelusios nigricans nigricans Hewitt, p. 375.

1933h. Loveridge, p. 209.

1936J. Loveridge, p. 223.

1937c. Loveridge, p. 269.

1937f. Loveridge, pp. 489, 492, 495.

1927a. Pelusios nigricans castaneus Hewitt, p. 375.

1928d. Loveridge, p. 51.

1935. Hewitt, p. 344.

494 bulletin: museum of comparative zoology

1927a. Pelusios nigricans rhodesianus Hewitt, Rec. Albany Mus., p. 375,

figs., la, lc, pi. xxvi, figs. 2-3: Mpika district, N. Rhodesia.
1933h. Loveridge, p. 210.

1934. Pitman, p. 307.

1929. Pelusios subniger Lindholm, p. 288.

1931. Pelusios nigricans castanoides Hewitt, Ann. Natal Mus., 6, p. 463, pi.

xxxvi, figs. 1-2: Richard's Bay, Zululand.

1932. Pelusios bechuanicus FitzSimons, Ann. Transvaal Mus., 15, p. 37:

Thamalakane River at Maun, Ngamiland, Bechuanaland Pro-
tectorate.

1934a. Mertens, p. 10.

1935b. FitzSimons, p. 306, pi. xi.

1933b. Pelusios subniger subniger Mertens, p. 263.

1934a. Mertens, p. 10.

1933b. Pelusios subniger castaneus Mertens, p. 263.

1934a. Mertens, p. 10.

1933. Pelusios sinuatus sinuatus Schmidt (not Smith), p. 3.
1934a. Pelusios subniger castanoides Mertens, p. 10.

1934a. Pelusios subniger rhodesianus Mertens, p. 10.

1934a. Pelusios subniger seyehellensis Mertens, p. 10.

1935. Pelusios rhodesianus Hewitt, p. 345.
1937b. Pelusios derbianus Angel, p. 1696.
1937a. Flower, pp. 14, 36.

1937f. Loveridge, pp. 489, 503.

1938e. Mertens, p. 430.

1939a. Sternothaerus castaneus seyehellensis Rendahl, pp. 308, 322, figs. 13-14.

1939b. Rendahl, p. 3.

Erroneous records of castaneus (Gray, 1831, 1844) and nigricans
(Loveridge, part, 1923g, 1928d) will be found under sinuatus, and
derbianus (Gray, 1873b; Peters, 1876a) under gabonensis.

Names. Black Terrapin (English); lihodu (Tereki); likudu (Ragoli);
kikui (Kami) ; malfudi (Gogo) ; kajamba (Nyakusa). In Angola kantuva
(at Osi); ombeo (at Bimbi); otyiti (at Elende); tyitunda (at Kuvangu).
Malagasy names are not included.

Illustrations. Good photographs of living terrapin are furnished by
Johnson (1907, p. 14) and Schmidt (1919, pi. xi), while most of the
more recent figures cited in the synonymy illustrate some form of this
variable reptile.

History. Vaillant (1891, p. 94) claims to be able to distinguish castaneus
in West Madagascar by a fine silver line encircling the pupil, while sub-
niger (or nigricans) of East Madagascar is said to have a uniform
brown iris, the difference is supported by other characters of minor

loveridge: African pelomedusidae 495

importance. This alleged distinction appeared to me to be a sexual
one, an idea which later received support from the plates supplied by
Vaillant and Grandidier (1910) which show a cf shell as castaneus, a
9 as svbviger. The point requires checking in the field, but the more
rounded shell of the female appears to have misled several workers into
supposing that they were dealing with two forms. Rawitz's (1915,
p. 582) statement that he noticed differences in spinal structure be-
tween Eastern and Western Malagasy specimens should also be investi-
gated.

Mertens (1933b, p. 263) repeats the other characteristics as typifying
a West Madagascar race, and they are certainly present in our solitary
West Madagascar specimen but also in a Zanzibar and occasional
mainland individuals. As the row of large polygonal shields anterior
to the barbels may be present or absent in mainland terrapin, I do not
attribute specific importance to them.

Siebenrock (1903d, p. 196) failed to find any constant characters by
which dcrbianus might be separated, though he proposed retaining the
name in the hope that some average difference might be found, re-
marking that apparently most herpetologists used the name on geo-
graphic, rather than taxonomic, grounds. Both black and yellow-brown
terrapin are represented in a series from Portuguese Guinea and simi-
lar colour differences have been reported from other West African ma-
terial. Our topotypic Gambia terrapin can be matched by individuals
of the same age and sex from the East so I cannot see any possibility
of regarding derbianus as separable even in a subspecific sense.

The type of oxyrhinus arrived at the zoo with a shipment containing
adansonii. In describing it, Boulenger stated that derbianus was its
closest relative but it might be distinguished on its more acuminate
snout. On this account Siebenrock (1903d, p. 192) synonymized it with
"niger", by which he meant gabonensis as here understood. This action
was incorrect as the vermiculate head markings are those of the sub-
niger group and cannot be confounded with the distinctive black
Y-shaped marking on the otherwise uniform crown of gabonensis.

S. n. seychellensis was long ago synonymized by Boulenger (1909h,
p. 295) and in detail by Nieden (1913c, p. 60). As I entertained expec-
tations that it would prove a valid race, I induced Mr. Vesey Fitz-
Gerald to send me a dozen specimens from the Seychelles. After draw-
ing up a detailed description based solely on this material, I compared
it with a similar description founded only on the thirty-four Kaimosi,
Kenya Colony, terrapin. The only significant difference, apart from
an average colour, was the better bicuspid development of the upper

496 bulletin: museum of comparative zoology

jaw, which appears to be constant in the Seychelles, highly variable on
the mainland. More recently, Rendahl (1939a, p. 308),' with half-a-
dozen Seychelle specimens, has revived seychellensis but as a race of
castaneus! He claims that both nigricans and c. seychellensis occur on
La Digue Id., and gives (p. 313) fourteen characters which allegedly
separate the two. I have carefully tested these with our Seychelle
material and find that Nos. 1, 3, 7 and 8 are individual variations, 11
is an age character, the rest are sexual, usually, though not invariably
constant. What he has done is to separate the sexes.

In describing P. n. rhodesianus from Mpika district, Hewitt (1927a,
p. 375) remarked that he had also a series of fairly typical nigricans
from the same district as well as from Kenya Colony. Later, finding
rhodesianus at Entebbe where it occurs alongside typical subniger (inc.
nigricayis) he thought it better to raise rhodesianus to specific rank.
Actually it is founded on an inconstant character — the shape of the
intergular — which does not remain constant as Hewitt had hoped.

P. n. castanoides of Zululand, based on a very old 325 mm. terrapin,
was considered to differ in the absence of barbels. The state of develop-
ment of these barbels varies considerably and though I have found no
specimens in which they could not be detected, I suggest that in cas-
tanoides they are probably absent as an individual variation, possibly
some injury to the derm in youth? FitzSimons (1935b, p. 307) found
them present and absent in specimens of Pelomedusa subrufa in the
Kalahari.

P. bechuanicus, founded on a single young 107.5 mm. terrapin, re-
sembles subniger in its uniform dark brown plastron and posteriorly
rounded carapace, but sinuatus in the strongly developed protuber-
ances on the vertebral shields. There are, however, individuals of
approximately the same size in our Kaimosi series which match this
unusual development, though the majority in the series are typically
subniger in this respect. Being so young, one would have expected
the vertebral shields to be broader than long, as indeed they seem
to be in the photographs; in description and diagnosis, however, they
are said to be longer than broad.

Description. Head broad, snout short; a pair of supraorbital shields
followed by a very large frontal, flanked by a pair of temporals (rarely
subdivided), often with a wedge-shaped group of small shields inserted
posteriorly between frontal and temporal, occasionally even separating
them; narrowest interorbital width equal to, or slightly longer, or
slightly shorter1 than the longitudinal suture between the supraorbital

i Ethiopia, Tanganyika, N. Rhodesia, Transvaal, Madagascar, Mauritius, Zanzibar and
Seychelles, but with exceptions.

loveridge: African pelomedusidae 497

shields; upper jaw angularly rounded, not definitely bicuspid1 but
sometimes indicated with a more or less prominent vertical groove
anteriorly ; chin with a pair of small barbels 2 ; scales on anterior aspect
of fore limb irregular in size; carapace moderately depressed, its height
included in its length 2.17 to 3.30 times, elongate in males, roundish in
females, its posterior margin rounded, not or but slightly serrate in
young; vertebral shields 5, the middle 33 obtusely (rarely quite strongly
with protuberances) keeled, usually 4 much broader than long in young
and as long as, or longer than, broad in adults, first and fifth very
variable but at their widest usually as broad as, or broader than long;
costals 4, rarely 55, pairs; marginals 22, rarely 20; supracaudals 2;
plastron smaller than the opening of the carapace in males, consider-
ably smaller in females; anterior lobe rounded or intergular slightly
projecting, not (males) or very distinctly (females) projecting beyond
the carapace; posterior lobe angularly and deeply notched (with a
tendency to be more acutely in males, more widely in females), the
points upturned, recurved, or horizontal; intergular X^/i to 4 times as
long as a gular, 1 }/% to 3 times as long as broad, its sides straight, wedge-
shaped, converging anteriorly or pyriform ; numerals forming a suture
1 to 5 times (averaging 2 for fifty terrapin) as long as that of the pec-
torals; outer border of a humeral normally much longer, occasionally
equal to, or much shorter than that of a pectoral; pectorals excluded
from bridge by abdominals; width of bridge contained \}/2 to 2 times in
the width of plastron, suture between abdominals shorter than (i.e. in-
cluded \x/i to 1^4 times) the length of the anterior lobe of the plastron.

The above description is based on all the subniger material in the
Museum of Comparative Zoology, while variations in the literature
are for the most part included as footnotes.

A deformed individual from Dakar is described in detail by Strauch
(1890, p. 102).

Anatomy. Tornier (1S96, p. 9) presents a diagram to illustrate the
changes in the outline of a carapace which accompany growth, it should
be accepted with reserve owing to his confusing sinuatus with subniger.
Vaillant (1880b, p. 797) comments on the vertebral column. Frade
(1936, p. 67) on the epithelial tissue, etc. Rawitz (1915, pp. 658, 663)
deals with the nervous system, and Peters (1848, p. 494) with the musk
gland.

1 See Tornier (1900b, p. 582) on variability in single terrapin.

2 Absent in type of castanoides fide Hewitt. See comments above.

3 These 3 broken up into 5 in a Malagasy terrapin (Anderson, 1910a, p. 11).

4 Exception a Chitau, Angola, juvenile and types of bechuanicus.
6 Mus. Comp. Zool. Nos. 7869, 40031.

498 bulletin: museum of comparative zoology

Coloration. Above, head yellowish or greenish gray vermiculated
with brown or black, the latter sometimes predominating to produce a
uniformly dark head ; limbs drab or yellowish brown ; tail (according to
Boettger, 1888a) yellow above with a blackish median line in males;
carapace in young, black, or yellowish with brown areolae which spread
till carapace becomes dark brown, often leaving light brown coloured
areas along the periphery, or black in adults. Below, jaws horn colour
with vertical striae of brown or black; plastron horn colour or yellow-
ish with brown or black infuscations which in Transvaal and Mala-
gasy terrapin simulate the more definite angular markings of sinuatus,
but usually are irregular, penetrating along the sutures between the
shields, in some areas the plastron is entirely black, or black with the
sutures between the shields white. In Natal {fide Strauch) and West
Madagascar (Siebenrock and M.C.Z. specimen) the entire reptile may
be stained reddish, due, it is said, to laterite soil.

Boulenger (1889a, p. 195), with only Madagascar material, writes of
nigricans "upper surface of head without spots or vermiculations."
Malagasy specimens in the Museum of Comparative Zoology exhibit
vermiculations though these are greatly reduced in one example.

Perhaps it might be helpful to list our material on the basis of
plastral coloration, allocating them into three groups though in some
cases with difficulty.

Plastron wholly black, or the sutures white or horn colour. Ex.
Kaimosi; Yala River; Entebbe and Ukerewe Id., Lake Victoria; Nyam-
kole, Lake Tanganyika ; Mwaya, Lake Nyasa ; Dodoma (though those
from this last locality and from Ukerewe Id. appear to have lost most
of the black through friction and polishing by sand).

Plastron yellow in centre, infuscated with brown on the margins.
Ex. Gambia River; or the brown forming a rather definite pattern as
figured by Dumeril & Bibron for eastaneus (pi. xx). Ex. Ethiopia;
Transvaal ; Madagascar.

Plastron wholly yellow, or almost so. Ex. Butiaba, Lake Albert;
Pemba Id. ; Zanzibar Id. ; Seychelle Ids.

Measurements. Length of carapace of a d71 from Faradje, 290 mm.,
breadth 185 mm., height 92 mm. Length of carapace of a 9 from
Faradje, 250 mm., breadth 177 mm., height 106 mm. (Schmidt, 1919,
p. 412).

More than once I have been inclined to postulate a small race, as
for example at Dodoma where the largest of fifty terrapin brought in
by natives, measured only 175 mm. It is wiser to bear in mind, how-
ever, the relative rarity and greater difficulty in capturing the largest

loveridge: African pelomedusidae 499

terrapin. Looking over the maximum carapace lengths recorded in the
literature I am inclined to think that the species attains its greatest
dimensions in the region of the Great Lakes and is slightly smaller
as one proceeds East and West. A trans-African series of records
follow.

San Thome, 170 mm. (Greef); Gambia, 176 mm. (M.C.Z.); Angola,
250 mm. (Monard) ; Belgian Congo, 290 mm. (Schmidt) and 280 mm.
in s.e. Lake Albert (Sternfeld); Uganda, 241 mm. (Hewitt); Kenya
Colony, 257 mm. (Loveridge) ; Tanganyika Territory, 235 mm. (Lov-
eridge); Northern Rhodesia, 218 mm. (Hewitt); Zululand, 225 mm.
(Hewitt); Madagascar, 220 mm. (Siebenrock) ; Seychelles, 168 mm.
(M.C.Z.)

If Lang (in Schmidt, 1919, p. 412) is correct in saying that this
species attains a length of 380 mm. (15 inches), which I am inclined
to doubt, it is very little smaller than sinuatus, long considered the
largest representative of the genus.

Sexual dimorphism. The measurements of male and female, fur-
nished above, give a very fair idea of the difference in proportion of
the sexes. The carapace tends to be parallel-sided in males, much
more rounded as well as arched (to accommodate the eggs) in females,
consequently the plastron appears broader posteriorly in males in its
relation to the carapace, than is the case in females, it may be slightly
concave in males though I am by no means certain that this is always
the case as some flat-plastroned terrapin appear to be males; the
notch in the posterior lobe may be more acute in males, more obtuse
in females (to aid oviposition?) but this appears to be questionable,
if an average difference. The tail of a male is longer, but is so short
that without comparative material of the opposite sex it would not be
recognised.

Boettger's (1888a, p. 15) alleged colour distinctions do not appear
to be valid, though it may be anticipated that sexual difference in the
colouration of the soft parts in life will be demonstrable. In this con-
nection see the remarks on page 494 under the heading History, where
it is suggested that the fine silver line encircling the pupil indicates a
male, its absence a female; a point worth investigating.

Breeding. The only account of the oviposition of this species which
I have come across, is that of Kaudern (1922, p. 449, fig. D), who
surprised a female at the edge of a desiccating pond at St. Marie de
Marovoay, Madagascar. He alleges that she excavated a flask-shaped
hole with her front legs and head1, deposited a dozen eggs in it in the

1 Almost certainly a mistake.

500 bulletin: museum of comparative zoology

course of an hour, then filled in the hole. Kaudern secured both the
terrapin and her eggs. It would be interesting to know her measure-
ments.

On March 1, at Kaimosi, Kenya Colony, a young terrapin with
abdominal shields still unhealed in the umbilical region, and carapace
length of 30 mm., was presumed to have hatched very recently.

Growth. In captivity, a young "nigricans" from Madagascar grew
from 88 to 100 mm. in four years, during the same period a young
"derbianus" from West Africa grew from 78 to 92 mm.: they shared
the same aquarium (Rembold, 1908, p. 743).

Longevity. For "derbianus" 40 years, 8 months, 13 days (Flower,
1937a, p. 14); for "nigricans" 9 years, 2 months, 11 days (Flower,
1925b, p. 933).

Diet. Grass and claws of crabs (Potamon sp.) in faeces at Kaimosi
(Loveridge). In captivity earthworms, mealworms, snails, fish, frogs,
and raw meat (Rembold, 1908, p. 743).

Enemies. Frequently taken in fish traps at Faradje, Dungu River
(Lang). Two terrapin, one having had a piece taken out of its side,
appeared to have been bitten when young by hyenas or other carni-
vore (Loveridge). According to natives, African Sea Eagles (Cuncuma
v. vocifer) were responsible for deviscerated shells found on the shores
of Lake Victoria (Loveridge).

Leeches commonly occurred on these terrapin at Kaimosi.

Defence. When first alarmed, the Black Terrapin retreats within
its shell, but if persistently annoyed it hisses and makes a crunching
sound, presumably with its jaws, finally emerging to snap at its tor-
mentor. If held, it scratches and discharges fluid; though the latter
action may be due to fear, it may act as a deterrent to predators.

Aestivation. On February 13, at Kaimosi, the first rain for months
fell between 4.30 and 5.30 p.m. The following morning a terrapin,
its back caked in mud, was found wandering: the deduction that it
had just emerged from aestivation appeared warranted by the mud on
its carapace.

Disease. Rembold (1908, p. 742) furnishes a photograph and ac-
count of a West African "derbianus" which developed a sac-shaped
swelling on the neck. After two years this swelling attained the size
of a beechmast and was accompanied by signs of nervousness, in
marked contrast to the creature's previous behaviour, though its
appetite remained good. In walking, the swelling was dragged along
the bottom of the aquarium by the terrapin, and considerable effort
was required for the reptile to raise its head for air. It frequently

loveridge: African pelomedusidae 501

rested its head upon some floating cork-bark and evinced an increasing
desire to leave the water.

Habits. Somewhat sluggish and secretive, at least the adults which
may be found in shallow water at the edge of large ponds at night.
Being nocturnal, they feed best in the evening, alternately gorging
and fasting according to Rembold, who claims evidence that they have
a well developed sense of taste. During the favourable conditions
prevailing in the rainy season they are prone to wander. Lang (in
Schmidt, 1919, p. 413) states that they rest upon submerged debris
or aquatic vegetation with the head and shell partly out of water.
Writing of "bechuanicus" , FitzSimons (1935b, p. 306) states that
they could be seen basking on exposed rocks, but slipped into the river
before one could get near, being extremely shy. His type was captured
with rod and line when he was fishing for barbel. I have never ob-
served this basking habit in the equatorial belt where the waters
are presumably warmer.

Habitat. Papyrus swamps, stagnant pools, ricefields, lakes and
rivers outside the rainforest in coastal plain and upland savanna.

Localities. Uganda : Bunjako ; Butiaba (Rutiala) ; Bussu near Jinja ;
Bwamba (Wawamba) ; Entebbe; Sesse Islands (Ussi Id.). Kenya
Colony: Athi River near Malemboa; Kaimosi; Yala River (M.C.Z.).
Tanganyika Territory: Bukoba; Dar es Salaam; Kaombwe's village,
Nkila, Ukimba; Karagwe (Karawe) ; Kasanga; Kilosa; Kisaki; Lake
Mkwera; Lake Nyasa; Lake Rukwa; Manda (as Wiedhafen) ; Mtita's
village near Dodoma; Mukwese; Mwaya; Sassi, Momba; Tanga
(? sinuatus); Ukerewe Id, Lake Victoria; Uluguru Mtns. at Nyange;
Wiedhafen (see Manda). Zanzibar: Pemba Id. : Mbuyuni; Zanzibar.1
Seychelle Islands: Diego Garcia; Gloriosa Id.; La Digue Id.; Mahe
Id. Mauritius. Madagascar — many localities. Mozambique:
Beira; Caia; Charre; Lorenzo Marques ; Mesuril; Ziweziwe. Nyasaland :
Shire River. Northern Rhodesia: Chambeshi River at Bwela Flats;
Lulimalala River at Chiwali's village; Mpika district; Msofu River;
Munyamadzi River; Nyamkolo. Southern Rhodesia: Gwamayaya
River at Gwelo; Mazoe; Mashonaland (as sinuatus). Bechuanaland
Protectorate: Thamalakane River at Maun. Transvaal: Aapies
River near Pretoria. (M.C.Z.). Zululand: Richard's Bay. Natal:
Durban Bay (Hewitt questions this locality on grounds collector lived
in Rhodesia); Angola: Ambriz; Bimbi; Chimporo; Chitau; Cubal;
Dondo; Duque de Bragan^a; Elende; Kuvangu; Loanda; Osi; Rio
Cuce; Rio Quilo. Cabinda: Chinchoxo. Belgian Congo: Banana;

1 Specimen so labelled in Mus. Comp. Zoology.

502 bulletin: museum of comparative zoology

Dika; Dungu River at Faradje; Eala; Kando; Kikondja; Kwamouth;
Lake Edward; Lukafu; Mahagi; Manda; Nyonga; Stanleyville.
Belgian Ruanda- Urundi: Usumbura. French Congo: ? Sao
Thome Id. and Rolas. French Cameroon: also ? (only Sjostedt's
1897 record of an entirely black 70 mm. terrapin) Togoland: Bis-
marckburg; Kete Kratje; Mangu. Liberia: Grand Cape Mtn. ; Junk
River. Sierra Leone. French Guinea: Kerouane; Labe, Fouta
Djalon; Tumbo Id. Portuguese Guinea: Bissau; Bolama. Gambia:
Gambia River; MacCarthy Id. Senegal: Cape Verde; Rufisque;
(Rochebrune's records are omitted). Cape Verde Ids.: S. Tiago
(Jago) Id. at Praia (Praja) Bay.

Range. West Africa from the Cape Verde Islands and Senegal south
to Angola, east to Zululand (and possibly Natal), north to Kenya
Colony, also islands of the Indian Ocean, viz. Pemba, Zanzibar,
Seychelles, Madagascar and Mauritius.

Pelusios sinuatus (Smith)

1831. Sternotherus castaneus Gray (not Schweigger), p. 38.

1844. Gray, p. 37.

1855. Gray, p. 52.

1838. Sternotherus sinuatus A. Smith, 111. Zool. S. Africa, Rept., pi. i: In

rivers to the north of 25° S., South Africa.
1851. Dumeril & Dumeril, p. 19.
1863b. !Gray, p. 193, fig.
1864a. Gray, p. 166, fig.
1865. Strauch, p. 109.
1866b. Peters, p. 887.
1869a. Peters, p. 11.
1870. Gray, p. 78, fig.
1873b. Gray, p. 69.
1882a. Peters, p. 8.
1889a. Boulenger (part), p. 194.
1894a. Gtinther (1893), p. 618.
1894. Gunther, p. 85.
1896c. Boulenger, p. 15.

1896. Tornier (part), p. 4.
1897g. Boulenger, p. 277.

1897. Tornier, p. 63.

1898. Jeude, p. 9.
1898. Johnston, p. 361.

1 The spelling Sternothaerus was adopted after this date, various minor misspellings of the
generic and specific name are ignored.

loveridge: African pelomedusidae 503

1898. Sclater, p. 97.

1898. Tornier, p. 282.

1899. Mocquard, p. 219.
1902b. Boulenger, p. 15.
1902b. Scherer, p. 265, fig.
1903d. Siebenrock, p. 193.
1905h. Boulenger, p. 251.
1907a. Boulenger, p. 6.
1908b. Mocquard, p. 557.
1909. Siebenrock, p. 556.
1911. Lonnberg, p. 7.
1911b. Masi, p. 132.
1912b. Boulenger, p. 329.
1912c. Sternfeld (part), p. 200.
1912b. Werner, p. 470.
1913c. Nieden, p. 55.

1915. Rawitz, p. 665, pi. xlix, figs. 68-75.

1916. Calabresi, p. 42.

1916. Siebenrock, pi. i, fig. 3; pi. ii, fig. 6.

1921d. Loveridge, p. 52.

1923b. Calabresi, p. 150.

1923g. Loveridge, p. 932.

1924b. Loveridge, p. 2.

1925b. Flower, p. 932.

1927. Calabresi, pp. 20, 37.

1930d. Witte, p. 85.

1939b. Rendahl, p. 2, figs. 1-5.

1848. Sternotherus dentatus Peters, Arch. Anat. Phys., p. 494: No locality.

1895L Stemothaerus bottegi Boulenger, Ann. Mus. Civ. Stor. Nat. Genova (2),

15, p. 9, pis. i-ii: Bardera, Italian Somaliland.

1897g. Boulenger, p. 277.

1900b. Pelomedusa galeata Tornier (part), p. 583.

1912c. Sternfeld (part), p. 201.

1923g. Stemothaerus nigricans Loveridge (part), p. 930.

1928d. Loveridge (part), p. 51.
1927a. Pelusios sinuatus Hewitt, p. 360. .,

1927c. Power, p. 411.

1929h. Loveridge, p. 5.

1931. Hewitt, p. 462, pi. xxxvi, fig. 3.

1933h. Loveridge, p. 208.

1934. Pitman, p. 307.

1935. Hewitt, p. 345.
1936h. Loveridge, p. 19.
1936j. Loveridge, p. 222.
1936e. Parker, p. 607.

504 bulletin: museum of comparative zoology

1937a. FitzSimons, p. 261, pi. x.

1937f. Loveridge, pp. 489, 492, 495.

1939b. FitzSimons, p. 19.

1927a. Pelusios sinuatus zuluensis Hewitt, Rec. Albany Mus., 3, p. 371, fig.

Id, pi. xx, figs. 1-3: Near Umsinene River, Zululand.
1934a. Mertens, p. 10.
1937b. Mertens, p. 5.

1931. Pelusios sinuatus sinuatus Hewitt, p. 462, pi. xxxvi, fig. 3.
1934a. Mertens, p. 10.

1933. Pelusios sinuatus leptus Hewitt, Oec. Papers Rhodesian Mus., p. 45,

pi. ix, figs. 1-2: Isoka, Northern Rhodesia.

1934. Pitman, p. 307.

Omitted from the preceding bibliography, having been transferred
to subniger, are records of sinuatus from Senegambia (Rochebrune,
1884a); Angola (Bocage, 1895a; Schmidt, 1933; Monard, 1937b);
Belgian Congo (Sternfeld, 1912c); Uganda (Tornier, 1896, Boulenger,
1902d, 1909b; Nieden, 1913c); Seychelles (Boulenger, 1889a; Schmidt,
1933; Monard, 1937b); Madagascar (Boettger, 1889, lS93a). All of
these regions being outside of its range.

Names. Serrated Terrapin (English) \fulwc (Jiji); ngongo (Konde).

Illustrations. Smith's plate of the type is a good representation of
an old individual in which many of the distinctive characteristics of
the species are blurred. Boulenger's (1895i) figures of bottegi from
above and below, give a much better idea of this distinctive terrapin.
Gray (1863b) has figured the head, and Siebenrock (1916) the young
for comparison with that of nigricans, i.e. subniger. Rendahl (1939b)
furnishes excellent figures of the plastral pattern, particularly that of
an aberrant individual (fig. 3).

History. Smith's type of sinuatus was recently located in the Royal
Scottish Museum (No. 1859.13.1864) by FitzSimons (1937a, p. 261,
pi. x), whose redescription, measurements, and figures are invaluable
to all workers on the involved tangle centering round its synonymy.
I have recently examined the type of bottegi in the Genoa Museum
(C. E. 2319) and confirm Siebenrock and Calabresi in their decision
to refer it to the synonymy of sinuatus. Cogent reasons for regarding
zuluensis and leptus as synonyms were advanced by me (1936J, p. 222)
some time ago, an opinion which I consider more than ever justified
since seeing additional material. Rendahl (1939b) with very limited
material, furnishes many measurements of variable characters in
contrast with those of subniger (which he calls nigricans and castaneus
seychellensis).

loveridge: African pelomedusidae 505

Description. Head broad, snout short; a pair of supraorbital shields
followed by a very large frontal, flanked by a pair of temporals (rarely
subdivided); narrowest interorbital width much less1 (about two-
thirds) than the longitudinal suture between the supraorbitals; upper
jaw angularly rounded in young, sometimes notched and bicuspid in
adults; chin with a pair of barbels; scales on anterior aspect of fore
limb irregular in size; carapace moderately depressed, its height in-
cluded in its length from 2.39 to 3.68 times, its posterior margin
strongly serrate in young, more or less serrated or sinuated except in
very old individuals; vertebral shields 5, the anterior 4 keeled, the
median 3 more or less protuberant posteriorly, occasionally smooth
in old individuals (of 221 mm. or over), as long as, or much longer
than, broad in adults, much broader than long in young, first and fifth
subequal in length and greatest breadth or broader than long; costals
4 pairs; marginals 22, rarely 24; supracaudals 2; plastron slightly
smaller than the opening of the carapace; anterior lobe rounded, not
or but slightly projecting beyond the carapace; posterior lobe angu-
larly and deeply notched; intergular shield 1 to 3 times as long as a.
gular, 134 to 2}/2 times as long as broad, its sides straight or wedge-
shaped; numerals forming a suture 1 to 2 times as long as that of the
pectorals, outer border of a humeral shorter, rarely slightly longer,
than that of a pectoral ; pectorals excluded from bridge by abdominals ;
width of bridge contained V/2 (adults) to nearly 2 times (young) in
the width of the plastron; suture between abdominals longer (adults)
or shorter (young) than the length of the anterior lobe of plastron.

The above description is based solely on a score of terrapin from
Kenya Colony and Tanganyika Territory, yet will be found, I think,
to embrace all variations recorded in the literature or in descriptions
of subspecies with the exception of those included in the footnote.

Anatomy. The skull (of zulucnsis) has been described by Hewitt
(1927a, p. 373).

Coloration. Above, head yellowish or pale olive finely speckled,
striated, or vermiculated with dark., brown; limbs drab or yellowish
brown flecked with darker; carapace in young, olive green to gray
brown turning to dark brown or black in adults. Below, plastron in
very young terrapin, brick red edged with black, the sutures between
the shields broadly edged with white; in young and half-grown, as
well as in most adults, the plastron is rich yellow, its periphery edged
with black in a very characteristic angular pattern whose sharp angles
and regular outline become blurred only in very old individuals.

1 Said to be equal to in botiegi, to be less or greater in zuluensis.

506 bulletin: museum of comparative zoology

Measurements. Length of carapace of a Lake Jipe specimen1, 380
mm. (Peters, 1882a, p. 8) exceeded only by a 9 from Amani2 of 383
mm. (Mertens, 1937b, p. 5). My largest, from Ruaha River, had a
carapace length of 360 mm., breadth 244 mm., height 138 mm. The
largest, specimens are said to attain a weight of 20 lbs.

Breeding. Native fishermen at Ujiji informed me that these terra-
pin come ashore to oviposit in July; this is hardly confirmed by the
presence there of two 51 mm. young on March 10, and two 51 mm.
young at Mbanja on April 27, unless the period of development is very
long.

Longevity. 8 years, 2 months, 27 days (Flower, 1925b, p. 932).

Diet. Scherer (1903b, p. 336) says that a score of young, about 50
mm. long, lived on a diet of raw fish and meat for three months, then
lost appetite, refused to eat, and died. He points out that under
natural conditions the insects, snails and fish upon which they feed,
probably supply calcium and vitamins essential to growth. Larger
terrapin of 80 mm. or more in length, throve on worms, fish and raw
meat. Frogs are apparently taken, see Loveridge (M. S.S.).

Enemies. At Ujiji, some natives admitted eating these terrapin,
others scornfully denied doing so.

Defence. A 173 mm. terrapin, when picked up and turned over,
ejected a fine jet of fluid from its right axilla or shoulder to a distance
of one foot, a second jet followed from the region of the left fore leg,
than a third from the right hind leg. Usually on being disturbed, a
terrapin will withdraw and enclose its head and fore limbs within the
shell, in my experience never attempting to use its strong jaws for
defensive purposes. Scherer (I.e.) states that they make low moaning
noises — which he thought originated in deep breathing — when within
the shell.

Habitat. Lakes and the larger rivers in the coastal belt and upland
savanna, to 5,000 feet.

Localities. Italian Somaliland: Bardera; Bulo Burti; Dolo; Imi;
Juba River; Lugh; Webi Mana. Kenya Colony: Archer's Post; Athi
River near Malemboa; Bulessa; Galass waterhole near Lake Rudolf;
Guaso Xyiro; Juja Farm; Tsavo River; Ukamba. Tanganyika
Territory: Amani2; Dar es Salaam; Kaombwe's village, Nkela,
Ukimba; Kasanga; Kilimanjaro; Lake Jipe; Lake Rukwa; Lake
Tanganyika; Little Ruaha River; Mombas River near Sassi;

i Not 385 mm. as stated by Rendahl (1939b, p. 2, footnote) .

2 I would suggest that this specimen more probably came from the Sigi River below Amani
and was taken up the mountain to sell by a native.

loveridge: African pelomedusidae 507

Morogoro; Muhesa; Pangani River; Ruaha River; Ruvu River;
Ruvuma River; Tanga; Ujiji; Usambara. Belgian Ruanda-Urundi:
Usumbura. Mozambique: Boror; Mesuril; Quelimane; near Tette.
Nyasaland: Shire Highlands. Northern Rhodesia: Isoka; Munya-
madzi River; Petauke. Southern Rhodesia: Mt.Chirinda; Sabi River
at Birchenough Bridge; Salisbury district. Bechuanaland Protector-
ate : Lobatsi (seen) ; Notuani River mouth in Limpopo Valley.
Transvaal: Gravelotte; Koedoeopoort near Pretoria; Letaba River
near Rubber Vale; Malta near Leydsdorp; Mawobya Creeks, Great
Letaba Rivers; Naboomspruit ; Vaalwater, Waterberg (M.C.Z.).
Zululand: Black Umfolosi River at 6 k.m. from Majimba Hill; L^m-
folosi Station; Umsinene River. Natal: Port Natal.1

Records from Senegambia, Angola, Belgian Congo, Uganda,
Seychelles and Madagascar, as listed on p. 504, are rejected, having
been based, so far as I have been able to ascertain, on examples of
subniger.

Range. East Africa from Italian Somaliland south to Natal, west to
Lake Tanganyika (but not Lake Victoria for Tornier (1896) and
Boulenger's (1909b) records from Sesse (Ussi) Islands, and Tornier
(1896) and Roux (1910) records from Bukoba were based on sub-
niger). Absent also from the Seychelles and Madagascar.

1 See footnote to key.

508 bulletin: museum of comparative zoology

BIBLIOGRAPHY

of works mentioning African pelomedusids from 1788-1939

Andersson, L. G.

1904. "List of Reptiles and Batrachians collected by the Swedish Zoo-
logical Expedition to Egypt, the Sudan and the Sinaitic Penin-
sula." in Jagerskiold, "Results Swed. Zool. Exp. to Egypt and the
White Nile 1901," 1, No. 4, pp. 1-12.

1910a. "Reptiles and Batrachians from the north-west of Madagascar
collected by V. Kaudern 1906-1907." Ark. Zool., 7, No. 7, pp. 1-15.

1937. "Reptiles and Batrachians collected in the Gambia by Gustav
Svensson and Birger Rudebeck (Swedish Expedition 1931)." Ark.
Zool., 29A, No. 16, pp. 1-28, figs. 1-8.

Angel, Fernand

1922a. "Sur une Collection de Reptiles et de Batraciens, recueillis au

Soudan francais par.le Mission du Dr. Millet Horsin." Bull. Mus.

Paris, 28, pp. 39-41.
1931d. "Reptilia et Batracia." in "Contribution a l'etude de la Faune

de Madagascar. Faune des Colonies Frangaises." 4, pp. 495-558,

figs. 1-15, pis. viii-ix.
1937b. "Sur la Faune herpetologique de l'Archipel du Cap- Vert." Comptes

Rendus XII. Cong. Internat. Zool. Lisbonne 1935, pp. 1693-1700.

Barbour, Thomas

1918. "Amphibia and Reptilia." in "Vertebrata from Madagascar."
Bull. Mus. Comp. Zool, 61, pp. 480-489.

Barbour, T. and Loveridge, A.

1928c. "A Comparative Study of the Herpetological Fauna of the Ulu-
guru and LTsambara Mountains, Tanganyika Territory, with De-
scriptions of new Species." Mem. Mus. Comp. Zool., 50, pp. 87-
265, pis. i-iv.

Bell, Thomas

1825. "A Monograph of the Tortoises having a movable Sternum, with
Remarks on their Arrangement and Affinities." Zool. Journ., 2,
pp. 299-310.

Bianconi, J. J.

1847-1859. "Specimina Zoologica Mosambicana." pp. 1-282, pis. i-xvii.

Blanford, W. T.

1870. "Observations on the Geology and Zoology of Abyssinia, made
during the Progress of the British Expedition to that Country in
1867-68." pp. i-xii + 1-487, figs., pis. i-viii, maps i-iv.

loveridge: African pelomedusidae 509

bocage, j. v. b. du

1866a. "Lista dos reptis das possessoes portuguezas d'Africa occidental
que existem no Museu de Lisboa." Jorn. Sci. Lisboa, 1, pp. 37-56.

1866b. "Reptiles nouveaux ou peu connus recueillis dans les possessions
portugaises de l'Afrique occidentale qui se trouvent au Museum
de Lisbonne." Jorn. Sci. Lisboa, 1, pp. 57-78, pi. i.

1867a. "Segunda lista dos reptis das possesoes portuguezas d'Africa occi-
dental que existem no Museu de Lisboa." Jorn. Sci. Lisboa, 1,
pp. 217-228.

1886a. "Reptis e amphibios de S. Thome." Jorn. Sci. Lisboa, 11, pp. 65-70.

1887a. "Melanges herpetologiques." Jorn. Sci. Lisboa, 11, pp. 177-211.

1892c. "Notice sur les Amphibiens et Reptiles recueillis par M. A. F.
Moller aux lies de la Guinee par le Dr. J. Bedriaga." Jorn. Sci.
Lisboa (2), 2, pp. 229-232.

1895a. "Herpetologie d'Angola et du Congo." pp. i-xx + 1-203, pis. i-ix.

1896a. "Reptis de algumas possessoes portuguezas d'Africa que existem
no Museu de Lisboa." Jorn. Sci. Lisboa (2), 4, pp. 65-104, pis. i-ii.

1905. "Contribution a la faune des quatre ties du Golfe de Guinee."
Jorn. Sci. Lisboa (2), 7, pp. 65-96.

BOETTGER, OSKAR

1881b. "Aufzahlung der von Frhrn. H. und Frhr. A. von Maltzan im
Winter 1880/81 am Cap Verde in Senegambien gesammelten
Kriechthiere." Abhand. Senckenberg. Naturf. Ges., 12, pp. 393-
418, pi. i.

1881c. "Die Reptilien und Amphibien von Madagascar. Dritter Naeh-
trag." Abhand. Senckenberg. Naturf. Ges., 12, pp. 435-558, pis. i-v.

1887b. "Zweiter Beitrag zur Herpetologie Siidwest- und Sud-Afrikas."
Berichte Senckenberg. Ges., pp. 135-173, pi. v.

1888a. "Materialien zur Fauna des unteren Congo II." Berichte Sencken-
berg. Ges., pp. 3-108, pis. i-ii.

1889. "Herpetologische Miscellen." Berichte Senckenberg. Ges., pp.
267-316.

1893a. "Katalog der Reptilien-Sammlung im Museum der Senckenberg-
ischen Naturforschenden Gesellschaft in Frankfurt-am-Main."
Frankfurt, pp. i-ix +1-140.^

1893b. "Ubersicht der von Prof. C. Keller anlasslich der Ruspoli'schen
Expedition nach den Somalilandern gesammelten Reptilien und
Batrachier." Zool. Anz., 16, pp. 113-119 and 129-132.

1894a. "Aufzahlung der Arten." Berichte Senckenberg. Ges., pp. 88-93-

1913. "Reptilien und Amphibien von Madagascar, den Inseln und dem
Festland Ostafrikas. (Sammlung Voeltzkow 1889-1895 u. 1903-
1905.)" in Voeltzkow, 1908-1917, "Reise in Ostafrika." 3, pp. 269-
376, pis. xxiii-xxx.

510

bulletin: museum of comparative zoology

BONNATERRE, M. L'ABBE

1789. "Tableau encyclopedique et methodique des trois Regnes de la
Nature: Erpetologie." pp. i-xxviii + 1-70, pis. A+i-xlii.

BOULENGER, G. A.

1880a. "Sur ['existence d'une seule espece du genre Pelomedusa Wagler."

Bull. Soc. Zool. France, 4, pp. 146-151, figs. a-g.
1889a. "Catalogue of the Chelonians, Rhynchocephalians, and Crocodiles

in the British Museum (Natural History)." pp. i-ix + 1-311, figs.

1-73, pis. i-vi.
1895b. "An Account of the Reptiles and Batrachians collected by Dr. A.

Donaldson Smith in Western Somali-land and the Galla Country."

Proc. Zool. Soc. London, pp. 530-540, pis. xxix-xxx.
1895i. Esplorazione del Guiba e dei suoi affluenti compeuta dal Cap. V.

Bottego durante gli anni 1892-93. II. Rettilie Batraci. Ann. Mus.

Civ. Stor. Nat. Genova (2), 15, pp. 9-18, pis. i-iv.
1896a. "A List of the Reptiles and Batrachians collected by Dr. Ragazzi

in Shoa and Eritrea." Ann. Mus. Civ. Stor. Nat. Genova (2), 16,

pp. 545-554.

BOULENGER, G. A.

1896b. "A List of the Reptiles and Batrachians collected by the late
Prince Eugenie Ruspoli in Somaliland and Gallaland in 1893."
Ann. Mus. Civ. Stor. Nat. Genova (2), 17, pp. 5-14.

1896c. "Report on Capt. Bottego's second Collection of Reptiles and
Batrachians from Somaliland." Ann. Mus. Civ. Stor. Nat. Genova
(2), 17, pp. 15-23, Pl. i.

1896e. "Second Report on the Reptiles and Batrachians collected by Dr.
A. Donaldson Smith during his Expedition to Lake Rudolf."
Proc. Zool. Soc. London, pp. 212-217, pis. vii-viii.

1897f. "Description of a new Tortoise of the Genus Sternothaerus." Proc.
Zool. Soc. London, p. 919, pl. liii.

1897g. "A List of the Reptiles and Batrachians of Somaliland and Galla-
land." Ann. Mus. Civ. Stor. Nat. Genova (2), 17, pp. 275-280.

1900b. "A List of the Batrachians and Reptiles of the Gaboon (French
Congo), with Descriptions of new Genera and Species." Proc.
Zool. Soc. London, pp. 433-456, figs. 1-2, pis. xxvii-xxxii.

1902b. "List of the Cold-blooded Vertebrates hitherto recorded from the
Uganda Protectorate, in Johnston, "The Uganda Protectorate."
1, pp. 445-447.

1902d. A List of the Fishes, Batrachians, and Reptiles collected by Mr.
J. Ffolliott Darling in Mashonaland, with Descriptions of new
Species." Proc. Zool. Soc. London, 2, pp. 13-18, pis. ii-iv.

1903e. "On a Collection of Batrachians and Reptiles from the Interior of
Cape Colony." Ann. Mag. Nat. Hist. (7), 12, pp. 215-217, pis.
xvi-xvii.

loveridge: African pelomedusidae 511

1905h. "On a Collection of Batrachians and. Reptiles made in South
Africa by Mr. C. H. B. Grant, and presented to the British
Museum by Mr. C. D. Rudd." Proc. Zool. Soc. London, pp. 24S-
255.

1906i. "On a Collection of Fishes, Batrachians, and Reptiles, made by
Mr. S. A. Neave in Rhodesia, North of the Zambesi, with Field
Notes by the Collector." Mem. Proc. Lit. Phil. Soc. Manchester,
51, pp. 1-12.

1907a. "Report on the Reptiles collected by the late L. Fea in West
Africa." Ann. Mus. Civ. Stor. Nat. Genova (3), 2, pp. 196-216,
figs. 1-9.

1907J. "Second Report on the Reptiles and Batrachians collected in South
Africa by Mr. C. H. B. Grant, and presented to the British Mu-
seum by Mr. C. D. Rudd." Proc. Zool. Soc. London, pp. 478-487,
figs. 140-141, pis. xxi-xxii.

1909b. "On a second Collection of Reptiles, Batrachians, and Fishes made
by Dr. E. Bayon in Uganda." Ann. Mus. Civ. Stor. Nat. Genova
(3), 4, pp. 302-307.

1909h. "A List of the Freshwater Fishes, Batrachians, and Reptiles ob-
tained by Mr. J. Stanley Gardiner's Expedition to the Indian
Ocean." Trans. Linn. Soc. London (2), 12, pp. 291-301, pi. xl.

1911c. "On a third Collection of Reptiles and Batrachians made by
Dr. E. Bayon in Uganda." Ann. Mus. Civ. Stor. Nat. Genova (3),
6, pp. 161-169.

1912b. "Missione per la Frontiere Italo-Etiopica sotto il Comando del
Capitano Carlo Citerni. Resultati zoologici. List of the Reptiles
and Batrachians." Ann. Mus. Civ. Stor.. Nat. Genova (3), 5, pp.
329-332.

1919g. "Batraciens et Reptiles recueillis par le Dr. C. Christy au Congo
beige dans les districts de Stanleyville, Haut-Uele et Ituri en
1912-1914." Rev. Zool. Afr. Bruxelles, 7, pp. 1-29.

BtJTTIKOFER, JOHANNES

1890. "Reisebilder aus Liberia, 2, Die Bewohner Liberia's — Thierwelt."
pp. 1-510, figs., pis. xix-xxxii.

Buxton, R. D.

1937. "A Natural History of the" Turkana Fauna." Journ. E. A. &
Uganda Nat. Hist. Soc, 13, pp. 85-104, pis. A-H and i-vi.

Calabresi, Enrica

1916. "Batraci e Rettili raccolti nella Somalia meridionale dai Dott.

Stefani e Paoli." Mon. Zool. Ital. Firenze, 27, pp. 33-45, pi. ii.
1923b. "Anfibi e Rettili dell'Africa orientale raccolti durante le Spedi-

zioni Franchetti e Zammarano." Atti Soc. Ital. Sci. Nat. Milano,

62, pp. 145-163, pi. v.

512 bulletin: museum of comparative zoology

1927. "Anfibi e Rettili raccolti nella Somalia dai Proff. G. Stefanini e
N. Puccioni." Atti Soc. Ital. Sci. Nat. Milano, 66, pp. 14-60,
pi. i.

Chabanaud, Paul

1921a. "Contribution a l'etude de la Faune herpetologique de l'Afrique

Occidentale. Deuxieme Note." Bull. Comite d'etudes hist. sci.

l'Afrique Occidentale Francaise, pp. 445-472, pis. i-iv, map.
1921b. "Mission Paul Chabanaud en Afrique Occidentale (1919-1920).

Liste des Batraciens et des Reptiles." Bull. Mus. Paris, 27, pp. 519-

525.

Chubb, E. C.

1908. "List of Batrachia and Reptilia collected in Northern Matabele-

land." Ann. Mag. Nat. Hist. (8), 2, pp. 218-221.
1909a. "The Batrachians and Reptiles of Matabeleland." Proc. Zool. Soc.

London, pp. 590-597.
1909b. "List of Rhodesian Batrachians and Reptiles in the Rhodesia

Museum Collection." Rhodesia Mus. Bulawayo 8th Ann. Rep.,

pp. 34-36.

Cott, H. B.

1928. "Report on the Zoological Society's Expedition on the Zambesi
1927." Proc. Zool. Soc. London, pp. 923-961, pis. i-iv.

Cowles, R. B.

1936. "Casual Notes on the Poikilothermous Vertebrates of the Um-
zumbe Valley, Natal, South Africa." Copeia, pp. 4-8.

Daudin, F. M.

1802. "Histoire naturelle, generate et particuliere des Reptiles." 2, pp.
1-432, pis. xvi-xxviii.

DONNDORFF, J. A.

1798. "Zoologische Beytrage zur xiii Ausgabe des Linneischen Natur-
systems." 3, "Amphibien und Fische." pp. 1-980.

DuMERIL, A. M. C. AND DuMERIL, A.

1851. "Catalogue methodique de la Collection des Reptiles du Museum
d'Histoire naturelle de Paris." pp. i-ix + 1-224.

DUMERIL, A. M. C. AND BlBRON, G.

1835. "Erpetologie Generale ou Histoire naturelle complete des Rep-
tiles." 2, pp. 1-680, pis. xi-xxiv.

DUMERIL, AUGUSTE

1852. "Description des Reptiles nouveaux ou imparfaitement connus
de la Collection du Museum d'Histoire naturelle et Remarques
sur la Classification et les Caracteres des Reptiles." Arch. Mus.
Paris, 6, pp. 209-264, pis. xiv-xxii.

loveridge: African pelomedusidae 513

1856. "Note sur les Reptiles du Gabon." Rev. Mag. Zool. (2), 8, pp.

369-377, pi. xx.
1860. "Reptiles et Poissons de TAfrique occidentale." Arch. Mus. Paris,

10, pp. 137-240, pis. xiii-xix.

FitzSimons, Vivian

1932. "Preliminary Descriptions of new Forms of South African Reptilia
and Amphibia, from the Vernay-Lang Kalahari Expedition, 1930."
Ann. Transvaal Mus., 15, pp. 35-40.

1935b. "Scientific Results of the Vernay-Lang Kalahari Expedition,
March to September, 1930. Reptilia and Amphibia." Ann.
Transvaal Mus., 16, pp. 295-397, figs. 1-30, pis. x-xi.

1937a. "Notes on the Reptiles and Amphibians collected and described
from South Africa by Andrew Smith." Ann. Transvaal Mus., 17,
pp. 259-274, pi. x.

1938. "Transvaal Museum Expedition to South-West Africa and Little
Namaqualand, May to August 1937. Reptiles and Amphibians."
Ann. Transvaal Mus., 19, pp. 153-209, pis. ii-iv, map.

1939b. "An Account of the Reptiles and Amphibians collected on an
Expedition to South-Eastern Rhodesia during December 1937 and
January 1938." Ann. Transvaal Mus., 20, pp. 1-46.

Fleck, Eduard

1894. "Vorkommen und Lebensweise der Reptilien und Batrachier."
Berichte Senckenberg. Ges., pp. 83-87.

Flower, S. S.

1900. "Notes on the Fauna of the White Nile and its Tributaries."
Proc. Zool. Soc. London, pp. 967-968.

1925b. "Contributions to our Knowledge of the Duration of Life in
Vertebrate Animals. III. Reptiles." Proc. Zool. Soc. London,
pp. 911-981.

1933. "Notes on the recent Reptiles and Amphibians of Egypt, with a
List of the Species recorded from that Kingdom." Proc. Zool.
Soc. London, pp. 735-851, map.

1937a. "Further Notes on the Duration of Life in Animals. III. Rep-
tiles." Proc. Zool. Soc. London, pp. 1-39.

Frade, Fernando

1936. "Contribution a l'Etude des Formations epitheliales de la Voute
pharyngienne et specialement de l'Hypochorde, chez l'Embryon
de Sternothaerus derbianus Gray." Arch. Portugaises Sci. Biol., 6,
pp. 67-75, pis. vii-viii.

Gadow, Hans

1901. "Amphibia and Reptiles" in "The Cambridge Natural History."
8, pp. 1-668, figs. 1-181, map. (reprinted in 1920).

514 bulletin: museum of comparative zoology

Gendre, E.

1909. "Liste de quelques Especes de Reptiles du Fouta Djalon." Extr.
Comptes Rendus, in Actes Soc. Linn. Bordeaux, 63, pp. cv-cvi.

Gray, J. E.

1831. "Synopsis Reptilium or Short Descriptions of the Species of
Reptiles." pp. i-viii+1-85, pis. i-xi.

1844. "Catalogue of the Tortoises, Crocodiles, and Amphisbaenians, in
the Collection of the British Museum." pp. 1-80.

1855. "Catalogue of Shield Reptiles in the Collection of the British
Museum. Part I. Testudinata (Tortoises)." pp. 1-82, pis. i-xlii.

1863a. "Notice of a new Species of Pelomedusa from Natal." Ann. Mag.
Nat. Hist. (3), 12, pp. 99-100.

1863b. "On the Species of the Genus Sternothaerus with some Observa-
tions on Kinixys." Proc. Zool. Soc. London, pp. 192-197, figs.

1864a. "On the Species of the Genus Sternothoeerus with some Observa-
tions on Kinixys." Ann. Mag. Nat. Hist. (3), 13, pp. 165-170, figs.

1864b. "Note on Sternothaerus adansonii from West Africa." Proc. Zool.
Soc. London, pp. 296-297, pi. xxiii.

1870. "Supplement to the Catalogue of Shield Reptiles in the Collec-
tion of the British Museum. Part I. Testudinata (Tortoises)."
pp. i-ix+ 1-120, figs. 1-40.

1872a. "Catalogue of Shield Reptiles in the Collection of the British Mu-
seum. Part II. Emydosaurians, Rhynchocephalians, and Amphis-
baenians." pp. 1-41, figs. 1-25.

1872b. in Sowerby and Lear, "Tortoises, Terrapins, and Turtles." pp.
1-16, pis. i-lx.

1873a. "On the Skull of Sternothaerus." Proc. Zool. Soc. London, pp. 392-
394, figs. 1-5.

1873b. "Hand-List of the Specimens of Shield Reptiles in the British
Museum." pp. 1-124.

Greeff, S. R.

1885. "tleber die Fauna der Guinea-Inseln S. Thome und Rolas." Sitz.
Ges. Beford. gesam. Naturwiss. zu Marburg, pp. 41-80.

GfjNTHER, Albert

1888a. "Report on a Collection of Reptiles and Batrachians sent by Emin
Pasha from Monbuttu, Upper Congo." Proc. Zool. Soc. London,
pp. 50-51.

1894a. "Second Report on the Reptiles, Batrachians, and Fishes trans-
mitted by Mr. H. H. Johnston, C.B., from British Central Africa."
Proc. Zool. Soc. London, pp. 616-628, pis. liii-lvii.

1894. "Report on the Collection of Reptiles and Fishes made by Dr.
J. W. Gregory during his Expedition to Mount Kenia." Proc. Zool.
Soc. London, pp. 84-88, pi. viii.

loveridge: African pelomedusidae 515

Hesse, Paul

1889. "Ueber einige Reptilien des untenen Congogebiets." Zool.
Garten, 30, pp. 257-267.

Hewitt, John

1927a. "Further Descriptions of Reptiles and Batrachians from South

Africa." Rec. Albany Mus., 3, pp. 371-415, figs. 1-2, pis. xx-xxiv.
1931. "Descriptions of some African Tortoises." Ann. Natal Mus., 6,

pp. 461-506, figs. 1-5, pis. xxxvi-xxxviii.
1935. "Some new Forms of Batrachians and Reptiles from South Africa."

Rec. Albany Mus., 4, pp. 283-357, pis. xxvii-xxxvi.
1937e. "A Guide to the Vertebrate Fauna of the Eastern Cape Province,

South Africa. Part II. Reptiles, Amphibians and Freshwater

Fishes." pp. 1-141, pis. i-xxxiv.

Jeude, T. W. van L. de

1895. "On a Collection of Reptiles from the Transvaal." Notes Leyden

Mus., 16, pp. 227-230.
1898. "Catalogue osteologique des Poissons, Reptiles et Amphibiens."
Mus. Hist. nat. Pays-Bas, 10, pp. 1-54, 1-52, and 1-11.

Johnson, W. B. and S. C.

1907. "Reptile Life." pp. 14, 69, photo. Glasgow.

Johnston, H. H.

1898. "British Central Africa." pp. 355-361a (list of Reptiles and Ba-
trachians), pi. xxiv.
1906. "Liberia." 2, pp. i-xvi + 521-1183, 454 illustr., 22 maps.

Kanberg, Hans

1924. "Uber eine neue Schildkrote aus Kamerun." Zool. Anz., 60, pp.

195-197, fig.
1926. "Erganzende Bemerkungen liber Sternothaerus heinrothi Kbg."
Zool. Anz., 67, pp. 225-227.

Kaudern, Walter

1922. "Sauropsidien aus Madagascar." Zool. Jahrb. Syst., 45, pp. 416-
458, figs. A-F, pis. xii-xiv.

Lacepede, M. le Comte de

1788. "Histoire naturelle des Quadrupedes ovipares et des Serpens." 1,
pp. 1-651, pis. i-xli.

1789. "Histoire naturelle des Quadrupedes ovipares et des Serpens."
2, pp. 1-527, pis. i-xxii, and table "Synopsis methodica . . ."

Lampe, Eduard

1911. "Erster Nachtrag zum Katalog der Reptilien — und Amphibien —
Sammlung des Naturhistorischen Museums der Stadt Wiesbaden."
Jahrb. Nassau Ver. Naturk. Wiesbaden, 64, pp. 137-236.

516 bulletin: museum of comparative zoology

LlCHTEN STEIN, M. H. C. AND MARTENS, E. VOn

1856. "Nomenclator Reptilium et Amphibiorum Musei zoologici beroli-
nensis." pp. i-iv + 1-48.

Lindholm, W. A.

1929. "Revidiertes Verzeichnis der Gattungen der rezenten Schildkroten
nebst Notizen zur Nomenklatur einiger Arten." Zool. Anz., 81,
pp. 275-295.

LoNNBERG, ElNAR

1907. "Reptilia and Batrachia" in Sjostedt, "Wissenschaftliche Ergeb-
nisse der Schwedischen Zoologischen Expedition nach dem Kili-
mandjaro, dem Meru und den umgebenden Massaisteppen 1905-
1906." pp. 1-28, pi. i.

1911. ' 'Reptiles, Batrachians and Fishes collected by the Swedish Zoolog-
ical Expedition to British East Africa 1911." Svenska. Vetensk. —
Akad. Handl., 47, No. 6, pp. 1-42, pis. i-ii.

Loveridge, Arthur

1921d. "Notes on Tortoises collected in East Africa. 1915-1919." Journ.

E. Africa and Uganda Nat. Hist. Soc, No. 16, pp. 50-52.
1923g. "Notes on East African Tortoises collected 1921-1923, with the

Description of a new Species of Soft Land Tortoise." Proc. Zool.

Soc. London, pp. 923-933, pis. i-ii.
1924b. "Check List of the Reptilia recorded from the British Territories

in East Africa." Spec. Suppl. No. 3 to Journ. E. Africa and Uganda

Nat. Hist. Soc, pp. 1-16.
1928d. "Field Notes on Vertebrates collected by the Smithsonian-Chrysler

East African Expedition." Proc. U. S. Nat. Mus., 73, Art. 17,

pp. 1-69, pis. i-iv.
1929h. "East African Reptiles and Amphibians in the United States Na-
tional Museum." Bull. U. S. Nat. Mus., No. 151, pp. 1-135, pi. i.
1933h. "Reports on the Scientific Results of an Expedition to the South-
western Highlands of Tanganyika Territory. VII. Herpetology.

Bull. Mus. Comp. Zool., 74, pp. 197-416, pis. i-iii.
1936h. "African Reptiles and Amphibians in Field Museum of Natural

History." Field Mus. Nat. Hist. Zool. Series 22, pp. 1-111.
1936J. "Scientific Results of an Expedition to Rain Forest Regions in

Eastern Africa. V. Reptiles." Bull. Mus. Comp. Zool., 79, pp. 209-

337, pis. i-ix.
1937c. "Zoological Results of the George Vanderbilt African Expedition

of 1934. VII. Reptiles and Amphibians." Proc. Acad. Nat. Sci.

Philadelphia, 89, pp. 265-296.
1937f. "Scientific Results of an Expedition to Rain Forest Regions in

Eastern Africa. IX. Zoogeography and Itinerary." Bull. Mus.

Comp. Zool., 79, pp. 481-541, pis. i-iv.

loveridge: African pelomedusidae 517

Mann, Ida

1931. "Iris Pattern in the Vertebrates." Trans. Zool. Soc. London, 21,
pp. 355^412, figs. 34-35, pis. xx-xxiv.

Masi, Luigi

1911b. "Osservazioni sopra due Esemplari de Sternothaerus." Boll. Soc.
Zool. Italiana Roma (2), 12, pp. 131-139.

Matschie, Paul

1893c. "Die Reptilien und Amphibien des Togogebietes." Mitt. Fors. Gel.
Deutsch Schutzgeb., 6, pp. 207-215.

Meek, S. E.

1910. "Batrachians and Reptiles from British East Africa." Field Mus.
Nat. Hist. Zool. Series, 7, pp. 403-414.

Merrem, Blasitjs

1820. "Versuch eines Systems der Amphibien." pp. 1-191, pi. i.

Mertens, Robert

1922a. "Verzeichnis der Typen in der herpetologischen Sammlung des

Senckenbergischen Museums." Senckenbergiana, 4, pp. 162-183.
1926a. "Herpetologische Mitteilungen VIII-XV. XIV. Zur Kenntnis Her-

petofauna von Angola." Senckenbergiana, 8, pp. 137-155.
1933b. "Die Reptilien der Madagaskar-Expedition Prof. Dr. H. Bluntsch-

li's." Senckenbergiana, 15, pp. 260-274.
1934a. "Systematische Liste der lebenden Schildkroten." Blatter Aquar.

Terrar., 45, pp. 1-12.
1937a. "Bemerkungen iiber die Rassen von Pelomedusa subrufa (La

Cepede)." Zool. Anz., 117, pp. 139-142, figs. 1-4.
1937b. "Reptilien und Amphibien aus dem sudlichen Inner- Afrika."

Abhand. Senckenberg. Naturf. Ges., 435, pp. 1-23.
1937d. "liber eine herpetologische Sammlung aus dem Gebiete des

Njarasa-Grabens Ost-Afrika." Veroft. Deutschen Kolon. Uber-

see-Mus. Bremen, 2, pp. 1-9.
1938e. "Amphibien und Reptilien aus Angola." Senckenbergiana, 20,

pp. 425-443, figs. 1-6.

Mocquard, Francois

1899. "Reptiles rapportes de l'Afrique australe et centrale par M.
Edouard Foa." Bull. Mus. Paris, 5, pp. 218-219.

1900a. "Nouvelle Contribution a la Faune Herpetologique de Madagas-
car." Bull. Soc. Philom. Paris (9), 2, pp. 93-11 1, pi. ii.

1902a. "Sur une Collection de Reptiles et de Batraciens recueillis par
M. Alluaud dans le sud de Madagascar." Bull. Soc. Philom.
Paris (9), 4, pp. 5-25, pis. i-ii.

1903. "Notes herpetologiques." Bull. Mus. Paris, 9, pp. 209-220.

518 bulletin: museum of comparative zoology

1906a. "La Faune herpetologique du Congo Francais." Revue Coloniale,

pp. 477-485, 554-564, 603-614, 668-674.
1908b. "Reptiles du Zambeze et des Grandes Lacs." in Foa, "Resultats

scientifiques des Voyages en Afrique d'Edouard Foa." pp. 557-

558.

Monard, Albert

1931. "Mission scientifique suisse dans l'Angola. Resultats scientifiques.

Reptiles." Bull. Soc. Neuchatel. Sci. nat., 55, pp. 89-111, figs. 1-5.
1937b. "Contribution a l'Herpetologie d'Angola." Arqu. Museu Bocage,

Lisboa, 8, pp. 1-154.

MtJLLER, FRIEDRICH

1885d. "Vierter Nachtrag zum Katalog der herpetologischen Sammlung

des Basler Museums." Verh. Naturf. Ges. Basel, 7, pp. 668-717,

pis. ix-xi.
1890a. "Fiinfter Nachtrag zum Katalog der herpetologischen Sammlung

des Basler Museums." Verh. Naturf. Ges. Basel, 8, pp. 249-296,

pis. i-iii.

MtJLLER, LORENZ

1910. "Beitrage zur Herpetologie Kameruns." Abhand. Akad. Wiss.

math. phys. Kl. Miinchen, 24, pp. 543-626, figs. 1-5, pi. —
1934. "tlber Pelusios niger (Dumeril & Bibron) und Pelusios heinrothi

(Kanberg)." Blatter Aquar. Terrar., 45, pp. 166-172.

Neumann, Oscar

1905. "Uber nordost-afrikanische und arabische Kriechtiere." Zool.
Jahrb. Syst., 22, pp. 389-404.

Nieden, Fritz

1910b. "Die Reptilien (ausser den Schlangen) und Amphibien Kameruns."
in "Die Fauna der deutschen Kolonien." 1, part 2, pp. i-iv + 1-74,
figs., map.
1913c. "Neues Verzeichnis der Kriechtiere (ausser den Schlangen) von
Deutsch-Ostafrika. I. Reptilia." Mitt. Zool. Mus. Berlin, 7,
pp. 51-100.

Parker, H. W.

1932b. "Two Collections of Reptiles and Amphibians from British So-

maliland." Proc. Zool. Soc. London, pp. 335-367, figs. 1-3.
1936e. "Reptiles and Amphibians collected by the Lake Rudolf Rift

Valley Expedition." Ann. Mag. Nat. Hist. (10), 18, pp. 594-609

figs. 1-13.

Peracca, M. G.

1904. "Rettili ed Anfibii dell'Eritrea raccolti Dott. Achille Tellini nel
1903." Boll. Mus. Zool. Torino, 19, No. 467, pp. 1-6.

loveridge: African pelomedusidae 519

Peters, W. C. H.

1848. "Ueber eigenthi'imliche Moschusdriisen bei Schildkroten." Arch.
Anat. Phys. Wiss. Med., pp. 492-496, pi. xvii.

1862b. "tjber die von dem so friih in Afrika verstorbenen Freiherrn von
Barnim und Dr. Hartmann auf ihrer Eeise durch Aegypten,
Nubien und dem Sennar gesammelten Amphibien." Monatsb.
Akad. Wiss. Berlin, pp. 271-279, pi. —

1864. "Eine neue Art der Baumvipern, Atheris polylepis, aus Liberia."
Monatsb. Akad. Wiss. Berlin, pp. 642-645.

1866b. "Ubersicht der aus dem Nachlafs des Baron Carl von der Decken
stammenden und aus seiner ostafrikanischen Reise gesammelten
Saugethiere und Amphibien." Montasb. Akad. Wiss. Berlin,
pp. 884-892.

1869a. "Amphibien." in Von der Decken, "Reisen in Ost-Afrika." 3,
Abt. 1, pp. 11-18, pis. i-ii.

1869b. "Forteckning pa de af J. Wahlberg i. Damaralandet insamlade
Reptilierna." Ofver. Kongl. Vetens.-Akad. Forh., pp. 657-662.

1875a. "tlber die von Hrn. Professor Dr. R. Buchholz in Westafrika
gesammelten Amphibien." Monatsb. Akad. Wiss. Berlin, pp. 196-
212, pis. i-iii.

1876a. "Eine zweite Mittheilung iiber die von Hrn. Professor Dr. R.
Buchholz in Westafrika gesammelten Amphibien." Monatsb.
Akad. Wiss. Berlin, pp. 117-123, pi. i.

1877b. "Uber die von Hrn. Prof. Dr. K. Mobius 1874 auf den Maskarenen
und Seychellen, sowie iiber die von Hrn. Dr. Sachs im vorigen Jahr
in Venezuela gesammelten Amphibien." Monatsb. Akad. Wiss.
Berlin, pp. 455-460, pi.

1877c. "Ubersicht der Amphibien aus Chinchoxo (Westafrika), welche
von der Afrikanischen Gesellschaft dem Berliner zoologischen
Museum tibergeben sind." Monatsb. Akad. Wiss. Berlin, pp. 611-
621, pi.—

1878a. "Uber die von Hrn. J. M. Hildebrandt wahrend seiner letzten ost-
afrikanischen Reise gesammelten Saugethiere und Amphibien."
Monatsb. Akad. Wiss. Berlin, pp. 194-209, pis. i-ii.

1880b. "Uber die von Hrn. J. M. Hildebrandt auf Nossi-Be und Mada-
gascar gesammelten Saugethiere und Amphibien." Monatsb.
Akad. Wiss. Berlin, pp. 508-511.

1882a. "Naturwissenschaftliche Eeise nach Mossambique auf Befehl
seiner Majestat des Konigs Friedrich Wilhelm IV. in den Jahren
1842 bis 1848 ausgefiihrt. Zoologie, 3, Amphibien." pp. i-xv + 1-
191, pis. i-xxx.

Pitman, C. R. S.

1934. "A Check List of Reptilia and Amphibia occurring and believed to
occur in Northern Rhodesia." in Pitman, "Report on a Faunal
Survey of Northern Rhodesia." pp. 1-500, maps.

520 bulletin: museum of comparative zoology

Power, J. H.

1927c. "On the Herpetological Fauna of the Lobatsi-Linokana Area.

Part I." Trans. Roy. Soc. S. Africa, 14, pp. 405-422, fig. 1, pis.

xviii-xxii, map.
1931. "On the Herpetological Fauna of the Lobatsi-Linokana Area.

Part II." Trans. Roy. Soc. S. Africa, 20, pp. 39-49, figs. 1-2.

Rathgen, W.

1895. "Eine ostafrikanische Schildkrote." Blatter Aquar. Terrar., 6,
pp. 200-203.

Rawitz, Bernhard

1915. "Das zentrale Nervensystem einiger Madagassischer Reptilien."
in Voeltzkow, 1906-1915. "Reise in Ostafrika in den Jahren 1903-
1905." 4, pp. 581-682, figs, a-11, pis. xl-xlix.

Reichenow, Anton

1874. "Eine Sammlung Lurche und Kriechthiere von Westafrika."
Arch. Naturg., 40, 1, pp. 287-298, pi. ix.

Rembold, Robert

1908. "Einige Beobachtungen hinsichtlich Sternothaerus derbianus Gray
und Sternothaerus nigricans Donndorff." Blatter Aquar. Terrar.
19, pp. 742-745, figs. 1-3.

Rendahl, Hialmer

1939a. "Zur Herpetologie der Seychellen. I. Reptilien." Zool. Jahrb.

Syst., 72, pp. 255-328, figs. 1-16.
1939b. "Einige Bemerkungen liber Sternothaerus sinuatus Smith." Ark.
Zool., 31A, No. 2, pp. 1-15, figs. 1-5.

ROCHEBRUNE, A. T. DE

1884a. "Faune de la Senegambie. Reptiles." pp. 1-221, pis. i-xx.

Rose, Walter

1929. "Veld and Vlei." pp. i-xxiii + 1-240, figs. — , photos 1-125.

Roux, Jean

1910. "Reise von Dr. J. Carl im nordlichen Zentral-Afrikanischen
Seengebiet. Reptilien und Amphibien." Rev. Suisse. Zool., 18,
pp. 95-103.

Ruppell, Eduard

1835. "Neue Wirbelthiere zu der Fauna von Abyssinien gehorig." 3,
Amphibien." pp. 1-18, pis. i-vi.

Scherer, Joseph

1902b. "Herpetologische Reiseskizzen aus Zentral-Ost-Afrika." Blatter
Aquar. Terrar., 13, pp. 263-267, figs.

loveridge: African pelomedusidae 521

1903b. "Sterrwthaerus sinuatus B." Blatter Aquar. Terrar., 14, pp. 335-
337.

Schmidt, K. P.

1919. "Contributions to the Herpetology of the Belgian Congo based
on the Collection of the American Museum Congo Expedition
1909-1915. Parti. Turtles, Crocodiles, Lizards and Chameleons."
Bull. Amer. Mus. Nat. Hist., 39, pp. 385-624, figs. 1-27, pis.
vii-xxxii.
1933. "The Reptiles of the Pulitzer Angola Expedition." Ann. Carnegie
Mus., 22, pp. 1-15, figs. 1-2, pis. i-ii.

SCHOEPFF, J. D.

1792. "Historia Testudinum." pp. i-xii + 1-136, pis. i-xxxii.

SCHWEIGGER, A. F.

1814. "Prodromi mon. Cheloniorum." pp. i-vi + 1-58.

Sclater, W. L.

1871. "Shell of a River-tortoise of the Genus Pelomedusa from the Upper
Zambesi." Proc. Zool. Soc. London, pp. 325-326, fig.

1898. "List of the Reptiles and Batrachians of South Africa, with De-
scriptions of new Species." Ann. S. African Mus., 1, pp. 95-108,
pi. v.

Scortecci, Giuseppe

1928b. "Contributo alia conoscenza degli anfibi dell 'Eritrea." Atti. Soc.

Ital. Sci. Nat. Milano, 68, pp. 176-192, pis. viii-ix.
1930a. "Rettili ed anfibi raccolti dal prof. E. Zavattari in Eritrea."

Atti. Soc. Ital. Sci. Nat. Milano, 69, pp. 193-217, figs.
1940a. "Reptilia," in "Missione Biologica nel paese dei Borana. Raccolte

Zoologiche." Reale Accad. d'ltalia, 2, pp. 5-30.

SlEBENROCK, FrIEDRICH

1897. "Das Kopfskelet der Schildkroten." Sitz. Akad. Wiss. Wien, 106,
1, pp. 245-328, pis. i-vi.

1899. "Uber den Kehlkopf und die Luftrohre der Schildkroten." Sitz.
Akad. Wiss. Wien, 108, 1, pp. 563-595, pis. i-iii.

1901. "Eine neue Schildkrote aus -Madagascar (nach Gerrard)." Zool.

Anz., 25, pp. 7-8.
1903a. "Schildkroten von Madagascar und Aldabra." Abhand. Sencken-

berg. Naturf. Ges., 27, pp. 241-259 pis. xxxiii-xxxv.
1903d. "Zur Systematik der Gattung Sternothaerus Bell." Zool. Anz.,

26, pp. 191-199.
1905a. "Chelonogische Notizen." Zool. Anz., 28, pp. 460-468.
1906a. "Ergebnisse einer zoologischen Forschungsreise von Dr. Franz

Werner nach Agypten und im agyptischen Sudan. IV. Krokodile

522 bulletin: museum of comparative zoology

und Schildkroten." Sitz. Akad. Wiss. Wien, 115, 1, pp. 817-839,

figs. 1-3.
1906b. "Schildkroten von Ostafrika und Madagascar." in Voeltzkow,

1906-1910, "Reise in Ostafrika in den Jahren 1903-1905, 2, pp.

1-40, pis. i-v.
1907. "Uber zwei Schildkroten aus Kamerun." Ann. Naturhist. Hofmus.

Wien, 22, pp. 1-8, fig. — , pi. i.

1909. "Synopsis der rezenten Schildkrote mit Benicksichtigung der
in historischer Zeit ausgestorbenen Arten." Zool. Jahrb. Syst.
Supp., 10, pp. 427-618.

1910. "Schildkroten aus Sud- und Siidwestafrika, gesammelt von Dr. R.
Poch und J. Brunnthaler." Sitz. Akad. Wiss. Wien, 119, 1, pp.
693-720, pis. i-iii.

1916. "Schildkroten aus dem nordlichen Seengebiet und von Belgisch-
Kongo." Ann. Naturhist. Hofmus. Wien, 30, pp. 1-12, figs. 1-2,
pis. i-ii.

Sjostedt, Yngve

1897. "Reptilien aus Kamerun West-Afrika." Bihang Kongl. Svenska
Vet.- Akad. Hand. 23, Part 4, No. 2, pp. 1-36, pis. i-iii.
Smith, Andrew

1838 "Illustrations of the Zoology of South Africa . . ." pp. 1-28,
to pis. i-lxxviii.
1849.

Sowerby, J. de C. & Lear, E. see Gray, J. E. 1872b.

Steindachner, Franz

1867a. "Reise der Osterreichischen Fregatte Novara um die Erde in den

Jahren 1857, 1858, 1859, unter den Befehlen des Commodore B.

von Wullerstorf-Urbair. Zool., 1, Reptilien." pp. 1-98, pis. i-iii.
1870. "Herpetologische Notizen II. I. Reptilien gesammelt wahrend

einer Reise in Senegambien (October bis December 1868). II.

Uber einige neue oder seltene Reptilien des Wiener Museums."

Sitz. Akad. Wiss. Wien, 62, pp. 326-350, pis. i-viii.

Stejneger, Leonhard

1893b. "On some Collections of Reptiles and Batrachians. from E. Africa
and the adjacent Islands recently received from Dr^ W. L. Abbott
and Mr. William Astor Chanler, with Descriptions of new Spe-
cies." Proc. U. S. Nat. Mus., 16, pp. 711-741.

Sternfeld, Robert

1911b. "Zur Herpetologie Sudwestafrikas." Mitt. Zool. Mus. Berlin, 5,
pp. 393-411.

191 Id. "Die Reptilien (ausser den Schlangen) und Amphibien von
Deutsch-Siidwestafrikas." in "Die Fauna der deutschen Kolo-
nien." 4, part 2, pp. i-iv + 1-65, figs. 1-76, map.

loveridge: African pelomedusidae 523

1912c. "Reptilia." in "Wiss. Ergeb. der Deutschen Zentral-Afrika-
Expedition 1907-1908." 4, 1913, pp. 197-279, figs. 1-4, pis. vi-ix.

1917. "Reptilia und Amphibia." in "Wiss. Ergeb. der Zweiter Deutschen
Zentral-Afrika-Expedition 1910-1911." 1, pp. 407-510, pis. xxii-
xxiv.

Sternfeld, R. & NlEDEN, F.

1911. "Zoologische Ergebnisse der Expedition des Herrn Hauptmann a.
D. Fromm 1908-1909 nach Deutsch-Ostafrika. III. Reptilia,
Amphibia." Mitt. Zool. Mus. Berlin, 5, pp. 383-385.

Strauch, Alexandre

1862a. "Chelonologische Studien, mit besonderer Beziehung auf die

Schildkrotensammlung der Kaiserlichen Akademie der Wissen-

schaften zu St. Petersburg." Mem. Acad. Imp. Sci. St. Peters-

bourg (7), 6, No. 7, pp. 1-196, pi. i.
1865. "Die Vertheilung der Schildkroten liber den Erdball. Ein zooge-

ographischer Versuch." Mem. Acad. Imp. Sci. St. Petersbourg

(7), 8, No. 13, pp. 1-207.
1890. "Bemerkungen iiber die Schildkrotensammlung im Zoologischen

Museum der Kaiserlichen Akademie der Wissenschaften zu St.

Petersburg." Mem. Acad. Imp. Sci. St. Petersbourg (7), 38,

No. 2, pp. 1-127, pis. i-iv.

TORNIER, GUSTAV

1896. "Die Kriechthiere Deutsch-Ost-Afrikas. Beitrage zur Systematik
und Descendenzlehre." pp. i-xiii + 1-164, figs. 1-11, pis. i-v.

1897. "Zur Faunistik Deutsch-Ost-Afrikas. III. Reptilien und Amphi-
bien." Arch. Naturg., 63, part 1, pp. 63-66.

1898. "Reptilien und Amphibien." in Werther, "Die mittleren Hoch-
liinder des nordlichen Deutsch-Ost-Afrika." pp. 281-304, figs.
1-15.

1900b. "Neue Liste der Crocodile, Schildkroten und Eidechsen Deutsch

Ost-Afrikas." Zool. Jahrb. Syst., 13, pp. 579-618, figs. A-H.
1901c. "Die Crocodile, Schildkroten und Eidechsen in Togo." Arch.

Naturg., 67, pp. 65-88.
1902b. "Herpetologisch Neues aus Ost-Afrika." Zool. Jahrb. Syst., 15,

pp. 578-590, figs. — .
1902c. "Die Crocodile, Schildkroten und Eidechsen in Kamerun." Zool.

Jahrb. Syst., 15, pp. 663-677, pi. xxxv.
1905c. "Schildkroten und Eidechsen aus Nordost-Afrika und Arabien."

Zool. Jahrb. Syst., 22, pp. 365-368.

Vaillant, Leon

1880b. "Sur la disposition des vertebres cervicales chez les Cheloniens."
Comptes Rendus Acad. Sci. Paris, 91, pp. 795-798.

524 bulletin: museum of comparative zoology

1891. "Remarques sur les Caracteres qui peuvent permettre de distin-
guer le Sternothaerus nigricans Lacepede du Sternothaerus castaneus
Schweigger." Bull. Soc. Philom. Paris (8), 3, pp. 94-96.

1898a. "Note sur le dessin inedit de Jossignyi (1770)." Bull. Mus. Paris,
4, pp. 133-138, fig. — .

Vaillant, L. & Grandidier, G.

1910. "Histoire Naturelle des Reptiles. Premiere Partie: Crocodiles
et Tortues." in "Histoire physique, naturelle et politique de
Madagascar." 17, pp. 1-86, pis. i-xxvii.

Wagler, Johannes

1830. "Natlirliches System der Amphibien, mit Vorangehender Classi-
fication der Saugthiere und Vogel." pp. 1-354, figs. — , pis. i-ix.

Werner, Franz

1898. "tiber Reptilien und Batrachier aus Togoland, Kamerun und
Tunis aus dem kgl. Museum fur Naturkunde in Berlin. II."
Verh. Zool.-Bot. Ges. Wien, 48, pp. 191-213, figs., pi. ii.

1908 "Ergebnisse der mit Subvention aus der Erbschaft Treitl unter-

(1907). nommenen zoologischen Forschungsreise Dr. Franz Werner's
nach dem agyptischen Sudan und Nord-Uganda. XII. Die Rep-
tilien und Amphibien." Sitz. Akad. Wiss. Wien, 116, 1, pp.
1823-1926, pis. i-iv.

1910a. "Reptilia et Amphibia." in Schultze, "Zoologische und Anthro-
pologische Ergebnisse einer Forschungsreise in Westlichen und
Zentralen SudAfrika. IV." Denks. Med.-Nat. Ges. Jena, 16,
pp. 279-370, figs. 1-15, pis. vi-xi.

1912b. "Die Lurche und Kriechtiere." in Brehm, "Thierleben." 4, ed. 4,
pp. i- + 1- , figs, maps, pis.

1924a. "Wissenschaftliche Ergebnisse der mit Unterstiitzung der Akade-
mie der Wissenschaften in Wien aus der Erbschaft Treitl von F.
Werner Unternommenen Zoologischen Expedition nach dem
Anglo- Agyptischen Sudan (Kordofan) 1914. XVIII. Schild-
kroten." Denks. Akad. Wiss. Wien, 99, pp. 263-273, figs.

WlTTE, G. F. DE

1928m. "Une Tortue de la region de Tabora." Revue Zool. Bot. Afr.,

16, Supp., p. 45.

1930b. "Reptiles rapportes par Madame Tinant." Revue Zool. Bot. Afr.,

17, Supp., p. 84.

1930d. "Une Tortue du Tanganyka." Revue Zool. Bot. Afr., 17, Supp.,

p. 85.
1933m. "Reptiles recoltes au Congo Beige par le Dr. H. Schouteden et

par M. G.-F. de Witte." Ann. Mus. Congo Beige. Zool. (1), 3,

pp. 153-188, pis. v-xi.

APMP

BOOKBIKO-'NG CO., IHC.

NOV 2 9 1983

100 CAM3MDGE STREET

CHARLESTOWfM. MASS.

Harvard MCZ Library

3 2044 066 303 611

cca^aftffE D"E

DEMCO, INC. 38-2931

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Bulletin of the Museum of Comparative Zoology at Harvard College"

See other formats