^0

HARVARD UNIVERSITY

Library of the

Museum of

Comparative Zoology

Bulletin' of the Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LX. No. 1.

NEW AND OLD SILURIAN TRILOBITES FROM SOUTH-
EASTERN WISCONSIN, WITH NOTES ON THE
GENERA OF THE ILLAENIDAE.

By Percy E. Raymond.

With Four Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

January, 1916.

No. 1. — New and old Silurian Trilobites from Southeastern Wisconsin,
with Notes on the Genera of the Illaenidae.

By Percy E. Raymond.

In the F. H. Day collection, received in January 1881, as a gift of
Mr. Alexander Agassiz, the Museum of Comparative Zoology secured
one of the three great collections of the Silurian fossils of southeastern
Wisconsin. It is particularly rich in Illaenidae, but contains also
such rare forms as Harpes telleri, Trochurv^ nasutus, and Dicrano-
peltis telleri, to be found elsewhere only in Mr. Teller's magnificent
collection.

The trilobite fauna of the quarries around Milwaukee and Racine
differs considerably from that found in the vicinity of Chicago, so
that some of the species described by Weller from the latter area are
either absent from the M. C. Z. collection or represented by speci-
mens from other sources than the Day collection. The Phacopidae
have been omitted from the present study, and only such species are
mentioned as are represented by specimens which add something to
what has already been published.

ILLAENIDAE Hawle and Corda.

The Illaenidae form a remarkably homogeneous group, and in spite
of the great number of species which have been described only three
generic or subgeneric names (Illaenus, Bumastus, and Thaleops) are
in common use. Holm recognized only Illaenus and Bumastus and
other writers have been even more conservative, referring all the
species to Illaenus. The only serious attempt to subdivide the genus
is that made by Salter, who recognized eight subgenera (including
Illaenus), but none of Salter's names has ever come into general use,
although some of them could be adopted advantageously. In all,
seventeen subgenera or genera have been proposed for inclusion in
this family, but only seven of these seem to be valid. The names, in
chronological order, are: — Cryptonymus Eichwald 1825, Illaenus
Dalman 1826, Deucalion Shtsheglov 1827, Bumastus Murchison
1839, Archegonus and Dysplanus Burmeister 1843, Thaleops Conrad
1843, Alceste Hawle and Corda 1847, Rhodope Angelin 1854, Actino-

4 bulletin: museum of comparative zoology.

lobus Eichwald 1860, Panderia Volborth 1863, lUaenurus Hall 1863,
Illaenopsis Salter 1S66, Ectillaenus, Hydrolaenus, Octillaenus Salter,
1867, and Illaenoides Weller 1907. Certain other genera, such as
Symphysurus and Nileus have often been placed with lUaenus, but
they have more generally been recognized as belonging to the Asa-
phidae. The fundamental difference between the lUaenidae and the
Asaphidae is, of course, the presence in the former family of an
epistoma, and its absence in the latter. The absence of the grooves
on the pleura of the thorax in the Illaenidae and their very general
presence in the Asaphidae affords quite a safe criterion for judging
of the dorsal surface.

The genera enumerated above may be taken up in order.

Cryptonymus Eichwald, 1825.

Observations geognostico-zoologicae per Ingriam marisque Baltici
Provincias nee non de Trilobites. Casani, 1825, p. 44.

In this paper, Eichwald, evidently not understanding Brongniart's
genus Asaphus, describes the new genus Cryptonymus, and under it,
eight species. The first four are species of Asaphus and the last four
are species of Illaenidae. Asaphus had been described in 1822, but
Illaenus was not published till 1826, so that, if the four asaphids were
eliminated, it would really leave an illaenid as the type of Cryptony-
mus. The first of the illaenids described, Cryptonymus roscnhergi, is a
Bumastus, the second, Cr. wahlenbcrgi, third, Cr. rudolphii, and the
last, Cr. parkinsonii, are all species of Illaenus. To choose among
these a type for Cryptonymus would be to upset one of two well-
established names without any possible gain. Eichwald himself ac-
cepted Illaenus as the name of these species, and in 1840 transferred
his name Cryptonymus to Trilobites pimcfatus Wahlenberg, a pro-
ceeding which he afterward stoutly defended (1855). If a new generic
name is at any time necessary for any of the eight species described
by Eichwald, Cryptonymus is still available, but till such a con-
tingency arises, it seems best not to revive the name.

Illaenus Dalman, 1826.

Type, Asaphus crassicauda Wahlenberg.

Om Palaeaderna eller de sa kallade Trilobiterna. Kgl. Vet. akad.
Handl., 1826, p. 248.

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 5

Dalman divided Asaphus into four sections, the third of which he
designated as Illaenus, using the name in a subgeneric sense. lUaenus
he divided into two divisions. Divisio 1, Cornigeri, contained the
single species Asaphus {Illaenus) centrotus, which Burmeister made the
type of Dysplanus in 1843. Divisio 2, Mutici, contained Asaphus
{Illaenus) crassicauda and A. {Illaenus) laticauda. The first of these
has by general consent been made the type of Illaenus. Holm has,
in several papers, redescribed the typical species, so that its charac-
teristics are well known. Beside the characters of the family, the
species shows a short and wide, strongl}^ convex and curved cephalon
and pygidium, both without concave borders, large prominent eyes
which are situated far back, deep but short glabellar furrows, short
but rather wide free cheeks without spines on the genal angles. The
thorax has a narrow axial lobe and ten segments. The pygidium has a
short but prominent axial lobe, and is wider than long. As Clarke
has pointed out, Illaenus americanus Billings is an American species
which is very similar to /. crassicauda, and it seems that only such
species as conform to the kind of structure exhibited by the type
should be admitted to the restricted genus Illaenus.

Deucalion Shtsheglov, 1827.

Sur les Trilobites en general et en particulier sur ceux de Zarskoe-
Selo. Journ. fur neue Endeckungen in der Phys. Chem. Natur. und
Technologic St. Petersburg, 1827, no. 1, 2, p. 234, pi. 7, f. 9 a-c.

I have not seen this paper, but judge from what Holm says that
Deucalion is a svnonvm of Illaenus. The genus was founded on a new
species, D. hrongniarti, which Holm was unable to recognize.

BuMASTUs Murchison, 1839.

Type, BuMASTUS barriensis Murchison (Partim).
Silurian system, 1839, p. 656 (non figs.).

The particular features of this genus upon which Murchison himself
laid most stress were the absence of dorsal furrows, and the presence
of ten segments in the thorax. The general usage, however, has been
that of referring all illaenids having the axial lobe of the thorax very

6 bulletin: museum of comparative zoology.

broad, its width equaling or exceeding one half the total width of the
body, to the genus Bumastus. This practice will probably prevail,
for we now know that other illaenids beside Bumastus have ten seg-
ments in the thorax, and there is no illaenid known from which dorsal
furrows are absolutely absent. In view of the somewhat numerous
subdivisions of the illaenids, it may be well to reexamine the type-
species, Bumastus barriensis Murchison, in a little detail. Salter has
explained that the specimens figured by Murchison really do not
belong to this species, so that Ave are obliged to use Salter's figures of
the "Barr Trilobite." Fortunately the M. C. Z. collection contains
two plaster casts of the original specimen figured by Jukes in 1829
and later by Salter. The casts are rather carelessly made, but are in
general in fair agreement with Salter's figures. From these sources
may be derived the statement that the typical species of Bumastus
is a large Silui'ian illaenid with smooth, subequal cephalon and pygi-
dium, rounded, spineless genal angles, large eyes, situated near the
posterior margin of the cephalon, a very wide axial lobe, shallow dorsal
furrows, ten segments in the thorax, and no trace of an axial lobe on
the pygidium. The dorsal furrows on the cephalon are short, extend-
ing but little ahead of the eyes. The cephalon does not appear to
have any rim or concave depression, but the pygidium shows a slight
concavity, so that the profile of that member does not present a
smooth convex curve, but the curvature is reversed near the posterior
end of the pygidium. Both cephalon and pygidium are wider than
long.

Archegonus Burmeister, 1843.

Die organisation der trilobiten, 1843, p. 120, 121, pi. 5, f. 3.

The type of this genus is Calyviene ? aequalis H. von Meyer, as this
was the only species cited by Burmeister in the first edition of his
Organisation der trilobiten. As the type is evidently not an illaenid,
but one of the Proetidae, the genus automatically disappears from the
family.

Dysplanus Burmeister, 1843.

Die organisation der trilobiten, 1843, p. 120.

Type, Illaenus centrotus Dalman.

The type of this genus is an illaenid with rather long, parabolic
head and abdomen-shields, spines at the genal angles, small eyes far

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. /

back, narrow axial lobe, and nine segments in the thorax. The
profiles of both shields are rather flat but uniformly convex curves.
(See especially Holm, Bihang Kogl. Vet. akad. Handl., 1883, 7, pi. 4,
f. l-l2). Holm was not able to see any value in this genus, pointing
out that the only real characteristic brought forward by Burmeister
and by Angelin was the presence of genal spines, and the species with
genal spines are so highly variable among themselves as to suggest
that this character in itself does not denote any real relationship. It
seems, however, that a certain group of illaenids can properly be de-
noted by this term, and the genus will be referred to later.

Thaleops Conrad, 1843.

Proc. Acad. nat. sci. Phil, 1843, 1, p. 331.

Type, Thaleops ovata Conrad.

The presence in this species of very high eyes on long peduncles,
long narrow genal spines, deep dorsal furrows sharply delineating a
prominent glabella, a narrow axial lobe, ten thoracic segments, and a
small short pygidium, mark an unusually well-defined genus which
seems to be confined to North America, and probably to the Ordovi-
cian, though one Silurian species has been referred to the genus.

Alceste Hawle and Corda, 1847.
Prodr. monog. Bohm. tril., 1847, p. 66, pi. 4, f. 31..

Type, Alceste latissima Hawle and Corda (which is the same as
lUaenus hisingeri Barrande, according to Barrande).

This genus is not valid, for it was based upon an immature speci-
men showing only four segments, and was very incorrectly described
and figured. Without Barrande's explanation, no one would be able
to identify the Alceste laiissima with any Bohemian trilobite.

Rhodope Angelin, 1854.

Pal. Scandinavia, 1854, pt. 1, Trilobita, p. 38, pi. 22, f. 17.

Type, Rhodope lineata Angelin.

The name Rhodope was used for a gastropod by von Siebold in
1848 (Anatomic, p. 296), and Volborth replaced the name by Panderia.

8 bulletin: museum of comparative zoology.

AcTiNOLOBUS Eichwald, 1860.

Type, Illaenus atavus Eichwald, 1857.

Lethaea Rossica, 1860, 1, p. 1488.

The type is an illaenid which seems sufficiently peculiar to deserve
a distinct generic name. The cephalon is short and the pygidium long,
and both cephalon and pygidium have a-concave border; the cephalon
a narrow lip, and the pygidium as wide a border as the average Isotelus.
The eyes are rather large, far back and far apart, free cheeks small,
genal angles rounded. The dorsal furrows of the cephalon are short,
the axial lobe of the thorax is narrow; ten segments are present; and
the axial lobe of the pygidium is short and triangular. Actinolobus
ataviis is a Russian Ordovician species (Cla), and another species with
a wide border on the pygidium is the one from the Silurian described
by Schmidt as Illaenus masckei (From F, Estland).

Panderia Volborth, 1863.

Type, Panderia triquetra Volborth.

Mem. Acad. imp. sci. St. Petersburg, 1863, 6, no. 2, p. 31.

Although proposing this name primarily to replace the preoccupied
Rhodope of Angelin, Volborth made his own new species the type,
and the genus must rest upon it. Holm does not actually use
Panderia, but he seems to have considered it a fit receptacle for the
group of small trilobites with only eight thoracic segments, and gives
(1883, p. 161) a new definition according to his interpretation of the
genus. The presence of only eight segments in the thorax does not
appeal very strongly to the present WTiter as a generic characteristic.
Panderia triquetra does, however, present some rather unusual charac-
teristics in its very short, strongly convex cephalon with extremely
large eyes, the high, well-defined glabella, and the short pygidium
with long, prominent axial lobe. Species of this type are not at all
common, and may be referred to Illaenus without doing violence to
the definition of that genus.

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. \f

Illaenurus Hall, 1863.

Type, Illaenurus quadratus Hall.

16th Rept. N. Y. state cab. nat. hist., 1863, p. 176, pi. 7.

Although Hall believed this species to be closely allied to Illaenus,
as indicated by the name, it seems more probable that it belongs to
the Asaphidae and is allied to Symphysurus.

Illaenopsis Salter, 1866.
Type, Illaenopsis thomsoi^i Salter.
Mem. Geol. surv. Gt. Britain, 1866, 3, p. 256.

As has been repeatedly pointed out, the grooved pleura of the
thorax of this trilobite exclude it from the Illaenidae, and place it near
Symphysurus in the Asaphidae.

OcTiLLAENUS Salter, 1867.
Type, Illaenus hisingeri Barrande.
Monog. Brit. Silurian trilobites, 1867, pt. 4, p. 182.

This genus was erected by Salter to contain the type, a species in
which the pleura of the first thoracic segment are produced into spines.
There are eight segments in the thorax, the axial lobe is narrow, the
glabella well defined, eyes of medium size and far back, free cheeks
with sharp genal spines Pygidium about as long as wide, without
defined axial lobe. This species could probably be placed with
Dysplanus, but I would follow Salter in the recognition of the remark-
able development of spines on the first thoracic segment, it being a
unique example of such a characteristic among the smooth trilobites.

Ectillaenus Salter, 1867.
Type, Illaenus perovalis Murchison.
Monog. Brit. Silurian trilobites, 1867, pt. 4, p. 182.

Holm has pointed out that in proposing this genus Salter confused
the true /. perovalis of Murchison and a new species afterward de-

10 bulletin: museum of comparative zoology.

scribed by Hicks as /. hughesi, and not understanding clearly the
characteristics of either species, produced a name of no particular
value. The name should be dropped, unless it can be shown to be of
more value than now appears to be probable.

Hydrolaenus Salter, 1867.

Type, Illaenus conifrons Billings.
Monog. Brit. Silurian trilobites, 1867, pt. 4, p. 182.

The type-species is a Thaleops, and Hydrolaenus is therefore a
synonym of that genus.

Illaenoides Weller, 1907.

Type, Illaenoides trilobus Weller.

Bull. Chicago acad. sci., 1907, no. 4, pt. 2, p. 226.

The type of this genus is remarkable chiefly for its small eyes which
are situated halfway to the front of the head. The glabellar furrows
are narrow and shallow, the facial suture cuts the cheeks very close
to the genal angles, making the free cheeks of unusual shape. The
genal angles are rounded. The axial lobe of the thorax is wider than
in typical Illaenus, but less wide than in Bumastus. The pygidium
is long with a narrow concave border, but no trace of an axial lobe.
Type and only known species from the Niagaran at Bridgeport (Chi-
cago), Illinois.

Su7nmary. — It appears that of the genera proposed, Illaenus,
Bumastus, Thaleops, Actinolobus, and Illaenoides have unquestion-
able value. Dysplanus and Octillaenus are more or less valuable but
need further study and redefinition. Cryptonymus, Deucalion,
Panderia, and Ectillaenixs are names which cannot be used at present,
but might possibly be revived. Archegonus, Illaenurus, and Illae-
nopsis belong to other families. Alceste was never properly defined.
Rhodope was preoccupied, and Hydrolaenus is a synonym. It is in-
teresting to note the lapse of forty years between Salter's new names
and the next generic name applied to a member of this group.

It is evident that the type-genus Illaenus contains the great majority

raymomd: new and old Silurian trilobites. 11

of the species in this family, the other genera having, as a usual thing,
only from one to two or three species each. Bumastus comes next to
lUaenus in the number of species, and has its greatest development in
America where there are at least six species in the Middle Ordovician
and about fifteen in the Middle Silurian, as contrasted with three or
four species in the Silurian of Great Britain, about the same number in
Scandinavia and Russia, and two in Bohemia.

Considering the great abundance of the illaenids, we have sur-
prisingly little information as to their ancestry or relationships.
Following the usual theory, which seems to be borne out by the facts
in most cases, one would expect these smooth forms to be the descend-
ants of more normal trilobites with glabellar furrows and with ribs
on the pygidium. But among all the illaenids there does not seem to
be one which shows any trace of ribs on the pygidium, while only a
few show indications of glabellar furrows. And such indications of
furrows as exist are merely spots or slight depressions on the smooth
glabella. The nearest relatives of the Illaenidae are undoubtedly the
Goldiidae (Bronteidae) not the Asaphidae, with which family they
have usually been classed. The presence in both the Illaenidae and
the Goldiidae of an epistoma, similar hypostomas, forward expanding
glabella, large eyes which are placed far back, unfurrowed pleura in
the thorax, and short axial lobe on the pygidium, indicate a very close
relationship, some of these characteristics being apparently too funda-
mental to admit of explanation on the ground of parallelism. The
Goldiidae, in spite of their specialization, are more like the typical
trilobite than the Illaenidae, and it would be natural to place them in
the ancestral position. The geological range at once negatives this
attempt, for the Goldiidae did not appear until the Middle Ordovi-
cian, are very rare in the Ordovician and reached their greatest de-
velopment in the Silurian and Devonian. The illaenids, on the other
hand, appeared in the basal Ordovician, possibly even in the Cambrian,
reached their greatest development in the Middle Silurian and did
not survive that period. That the Goldiidae should have been de-
rived from the Illaenidae, however, seems highly improbable, for the
phylogeny of the former family pursues a normal course, the oldest
members of the family being most highly segmented, and the usual
''smoothing out" process producing such (relatively) lUaenus-like
species as Goldius dormitzeri, G. campanifer, and G. hrongniarti in the
Devonian of Bohemia. If these species, without glabellar furrows
and with highly convex almost ribless pygidia occurred in the Ordo-
vician, and forms like Goldius hmatvs (Billings) in the Devonian, we

12 bulletin: museum of comparative zoology.

should probably conclude at once that the Goldiidae were derived
from the Illaenidae.

The oldest illaenids which are well known are those which Barrande
described from the base of the Ordovician in Bohemia. Of these,
Illaenus advena is a quite typical Illaenus, while the other species are
peculiar. Illaenus bohemicus, the type of which is in the M. (". Z.,
was founded on a badlj^ preserved single specimen, which does not
seem to be an Illaenus. The pygidium sKows a long, distinct, and
ringed axial lobe and if this member were found alone, it would at
once be assigned to the Asaphidae. The nine segments of the thorax
are, however, without pleural furrows. The cephalon is too poorly
preserved to indicate any characters of value. Long wide genal
spines are present, the glabella appears to have nearly parallel sides
imtil the vicinity of the anterior end is reached, when it expands
abruptly. Rather faint glabellar furrows seem to be present. With
only this single specimen, it does not seem possible to assign the
species to any genus or even family, and it certainly throws no light
on the origin of the Illaenidae. Another species from Di, described
by Barrande, is the common Illaenus katzeri. This species differs
from other illaenids in its eyes. Barrande supposed it to be blind,
but Holub has recently shown (Bull, international Acad. sci. Boheme,
1908. German abstract, p. 7, pi. 7) that it has small eyes, situated
forward. Both shields are rather flattened, not so strongly incurved
as in the typical genus Illaenus, and the pygidium is long and parabolic
in outline, thus suggesting Dysplanus. There is nothing, however,
to indicate that /. katzeri is primitive, but it seems rather, like the
Silurian Illaenoides trilobus, to be a degenerate form. If the genus is
divided at all, this species cannot be considered as congeneric with
Illaenus crassicauda, and I would suggest that the name Wossekia
be applied to it, since Wossek, Bohemia, is the locality from which
practically all the specimens have been obtained. Illaenus ])uer
seems to be, as Brogger was the first to point out, a Symphysurus.
Illaenus calvus has a peculiar marginal rim. The species is known
from a single fragment. Illaenus aratus has a narrow glabella, and
the eyes are very far apart or absent. It is primitive for an illaenid,
but gives no suggestion as to the origin of the group.

A imthologic specimen of Illaenus. — Since this paper was written,
I have been enabled, through a grant from the Shaler Memorial fund
at Harvard. College, to visit northern Europe, and some time was
spent in studying the collections in London, Berlin, St. Petersburg,
Stockholm, and Christiania. One of the most interesting trilobites
seen was the specimen of Illaenus revelensis Holm, which was figured

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 13

Ly Holm in his description of the Russian Ordovician Illaenidae
(Mem. Acad. imp. sci. St. Petersburg, 1886, ser. 7, 33, pi. 2, f. 5a).
This is the only example I have seen among trilobites of a malforma-
tion due to moulting. As partially shown in Holm's figure, there is
an impressed line between the facial suture and the dorsal furrow on
the left side, which follows exactly the course of a facial suture, even
extending across the front of the cranidium, a point not shown in
Holm's figure. Across the posterior part of the cephalon, close to the
edge, is a furrow marking the posterior edge of the shell at the time
of the previous moult. The eye on the left side is smaller than its
opposite, and the palpebral lobe is malformed. The cephalon is
decidedly unsymmetrical, the left free cheek being drawn backward.
All these pathological features seem to be due to the partial retention
of the shell at the next previous moult.

Classification of the Illaenidae.

From the first, all classifications of the Illaenidae have rested mainly
upon the number of segments in the thorax, and secondarily upon the
width of the axial lobe of the thorax. Thus, Dalman, the describer
of Illaenus, separated the species with nine segments from those with
ten. Holm, the principal writer upon the genuS; while recognizing
only the genus Illaenus and the subgenus Bumastus, divided Illaenus
into three groups, those with ten, nine, and eight segments in the
thorax. A study of the American illaenids does not favor a classifica-
tion of this sort, for it has been repeatedly shown that the nvunber of
thoracic segments in species of both Illaenus and Bumastus is variable,
even within the limits of a single species. Likewise, the presence or
absence of genal spines is not a characteristic justifying, in itself, the .
erection of a genus of Illaenidae, for, as has been several times pointed
out, species are found with all sorts and conditions of spines, and if all
species having this characteristic were to be referred to a single genus,
there would be hardly another characteristic common to the assem-
blage. The length and convexity of the cephalon and pygidium, the
size and position of the eyes, the width of the axial lobe of the thorax,
and the shape of the glabella seem of the most importance, but I would
also take into consideration the sort of genal spines which may be
present. It is still too early to make any natural classification, and
the genera here recognized are based primarily upon the more con-
spicuous peculiarities of the type-species of each.

14 bulletin: museum of comparative zoology.

ILLAENIDAE Hawle and Corda.

Opisthoparia with large, convex, nearly smooth, cephalic and
abdominal shields. Epistoma large, hypostoma convex, ovoid.
Thorax of eight to ten segments, with unfurrowed pleura. Pygidium
without ribs, axial lobe short or absent. Ordovician and Silurian.

ILLAENINAE, subf. nov.

Illaenidae with narrow axial lobe, cephalon and pygidium without
concave border.

Illaenus Dalman. Cephalon and pygidium very convex, wider
than long, abruptly deflected; eyes large and far back, axial lobe of
pygidium high. Genal spines, when present, rounded in section.
Type, Illaenus crassicauda (Wahlenburg). Ordovician and Silurian.
Europe, North and South America, India, and Australia.

Thaleops Conrad. Cephalon and pygidium similar to Illaenus, but
with eyes on long stalks and elongate genal spines always present.
Type, Thaleops ovata Conrad. Ordovician and possibly Silurian.
North America.

Dysplanus Burmeister. Cephalon and pygidium long and flattened,
parabolic in outline, genal spines present, usually flattened. Type,
Illaenus centrotus Dalman. Ordovician. Northern Europe.

Wossekia, gen. nov. Cephalon and pygidium as in Dysplanus,
genal spines absent, eyes small and far forward. Type, Illaenus
katzeri Barrande. Basal Ordovician. Bohemia.

Octillaenus Salter. Similar to Dysplanus but with rounded genal
spines, and pleura of first segment elongated into spines. Type,
Illaenus hisingeri Barrande. Ordovician. Bohemia.

BUMASTINAE, subf. nov.

Illaenidae with (usually) concave border on one or both shields ,
axial lobe generally wide, though sometimes narrow.
• Bumastus Murchison. Axial lobe of thorax equal to more than
half the total width; eight to ten segments. Typically, a concave
border is present on the pygidium, frequently on both cephalon and
pygidium, rarely (in small Middle Ordovician species) without con-

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 15

cave border on either shield. Eyes very large. Type, Bumastus
harriensis Murchison. Ordovician and Silurian. North America.
Silurian. Europe.

Actinolobus Eichwald. Axial lobe narrow, cephalon short, pygi-
diura long, with very wide concave border. Eyes large. Type,
Illaenus atavus Eichwald. Ordovician, Russia. Silurian. Russia
and United States.

lUaenoides Weller Axial lobe with a width between that of Actino-
lobus and Bumastus, eyes very small and far forward, narrow con-
cave border on pygidium. Type, lUaenoides trilobus Weller. Silurian.
United States.

The above classification is designed to separate the species with
long more or less flattened shields from the more typical illaenids with
short and abruptly deflected cephalon and pygidium. The first two
genera are modifications of the central Illaenus type, the other three
of the more flattened Dysplanus group. In defining the subfamily
Bumastinae as I have, all the forms with a more or less Isotelus-like
pygidium are removed from the typical Illaenus group. Among the
small species constituting the earliest of the Bumastinae one finds
species like B. glohosus Billings, B. bellevillensis Raymond and Narra-
way, and a few others, which lack a concave border. On the other
hand, so large a Middle Ordovician Bumastus, as B. indeterviinatvs
(Walcott), the type of which is figured (Plate 2) for the first time, has a
distinctly concave border. It is very possibly true that the small
species mentioned above should be given a distinct name and placed
in the lUaeninae, but it still seems somewhat early to take so radical
a step.

Description of species.

We owe to Professor Weller a complete and careful description of
the Illaenidae of the Chicago area, and, as he had access to collections
made at Racine and near Milwaukee, his description in large measure
covers the Wisconsin area also. In the large collections which I have
been able to examine, I have, however, found a few specimens more
perfect than those previously described, and also a few new species.
When first studying the excellent figures given by Weller, one is struck
by the apparent triviality of the specific characteristics employed in
the discrimination of the species, but with a large collection, it is
found that the characteristics are remarkably constant. The study
of these illaenids is unusually interesting, in fact, for it seems to be

16 bulletin: museum of comparative zoology.

one of the few cases where characteristics arbitrarily chosen may serve
to define natural groups. The bumastids particularly, which are
simple trilobites with few variable characters, seem to be susceptible
to this sort of treatment. The chief variable characters are amount of
convexity, ratio of length to breadth, length of dorsal furrows, pres-
ence or absence of "lip" on cephalon and concave border on pygidium,
and size of eyes. Almost every possible combination of these few
characteristics seems to be present among the species, and each com-
bination is usually exhibited by a large number of specimens. That
these variations are not the attributes of one very plastic species but
of a number of distinct species is shown by the geographical distribu-
tion as well as by the numbers in each group. -For example, Bumastus
insignis seems to be confined to a small area in the immediate vicinit;\'
of Chicago, and B. ioxus to Joliet, Illinois and Racine, Wisconsin,
while B. cunimdus is found in vast numbers near Milwaukee, but is
rare in the Chicago area. Others of the species are equally local in
distribution.

Bumastus cuniculus (Hall).

Illaenus cuniculus Hall, 20th Rept. N. Y. state cab. nat. hist., 1868, p. 377,
pi. 22, f. 12; 1870, rev. ed., p. 421, pi. 22, f. 12. Weller, Bull. Chicago
acad. nat. hist., 1907, no. 4, pt. 2, p. 219, pi. 19, f. 1-6.

This is by far the most common species at Wauwatosa, Wise.
There are several nearly complete specimens in the M. C. Z., most of
them enrolled. They show the axial lobe of the thorax to be extremely
wide, and the dorsal furrows very shallow. There are ten segments.
The diagnostic specific characteristics are: — elongate, moderately con-
vex cephalon with very narrow rim which is prominent at the front,
but disappears before reaching the genal angles, and eyes of medium
size. Dorsal furrows faint, hardly visible at all in front of the scar-
like spots just inside the eyes. Pygidium elongate, moderately convex,
with practically no depressed border. A flattening of the convexity
near the border can be seen if the pygidium is viewed in profile.

Measurements: — A cephalon of average size is 45 mm. long, 55 mm.
broad; the eye 9 mm. long, or one fifth the total length. A large
cephalon is 56 mm. long. The pygidium of an enrolled specimen
whose cephalon is 45 mm. long is 50 mm. long and 52 mm. wide. The
axial lobe of this specimen is 42 mm. wide and the total width of the
thorax is 53 mm.

Formation and locality: — Very common in the Niagaran at Wauwa-
tosa, near Milwaukee, Wisconsin.

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 17

BuMASTUS NiAGARENSis (Whitfield).
Plate 1, fig. 3.

Illaenus niagarensis Whitfield, Ann. rept. Geol. surv. Wisconsin for 1879, 1880,

p. 68. Weller, Bull. Chicago acad. sci., 1907, no. 4, pt. 2, p. 219, pi. 19,

f. 7-11.
Illaenus madisonianus Whitfield. Geol. Wise, 1882, 4, p. 307, pi. 20, figs. 8-9.

Foerste, Bull. sci. lab. Den univ., 1885, 1, p. 106 pi. 14, f. la-b, 2a-b;

1887, 2, p. 93, pi. 8, f. 8, 9, 10, 10a; Geol. surv. Ohio, 1893, 7, p. 526, pi.

26, f. 1, 2, varieties, pi. 27, f. 7-10. Van Ingen, School of mines quar-
/ terly, Columbia univ., 1901, 23, p. 35.

This seems to be one of the few illaenids of the Chicago-Wisconsin
Niagaran area which has hitherto been known from entire specimens.
Whitfield figured an entire one, Weller had a nearly complete speci-
men, and there are five in the M. C. Z. (Day collection). One of
these is figured as it shows some characters not shown in either Whit-
field's or Weller's figure. Whitfield's figure, if it really represents this
species, is inaccurate in respect to the glabellar furrows, which are
really much longer than is indicated by his figure or description.
From the general proportions of the body, and the position of the eyes,
it would seem that his figure really does represent this species. The
specimen here figured is a little longer and narrower than those
previously figured, and has a longer and more pointed pygidium.

It will be noted that the thoi'ax is exceedingly short, though ten
segments are present. The dorsal furrows do not show in the figured
specimen, but they do on another specimen in the M. C. Z. collection,
and are also indicated in Whitfield's figure.

The specific characteristics are : — cephalon rather convex, with
long dorsal furrows, no lip or concave border on the cephalon. Eyes
of medium size, situated nearly their own length from the posterior
margin, thorax short, pygidium long, rather pointed behind, with
narrow concave border.

Measurements: — The specimen (Plate 1, fig. 3) is 58 mm. long;
cephalon 22 mm. long, 31 mm. wide; thorax 13.5 mm. long; pygidium
29 mm. long, 29 mm. wide; the eyes are damaged. On a very good
enrolled specimen preserving the test, the cephalon is 24 mm. long,
the eye 5 mm. long, and 5 mm. from the posterior margin. On a
larger specimen the eye is 1 mm. more than its own length from the
posterior margin.

18 bulletin: museum of comparative zoology.

Formation and locality: — This species is quite common in the Niag-
aran at Wauwatosa, though not nearly so common as B. cunicuhis.

BUMASTUS DAYI, Sp. nOV.

Plate 1, fig. 8-10.

The Day collection contained two fine specimens labeled sp. nov.,
and as they have proven to be such I have named them in honor of
the collector. The specific characteristics are: — cephalon and pygi-
dium short and convex, eyes large and far back, no lip or concavity
at the front of the glabella, though there is one at the sides of the free
cheeks, dorsal furrows of the cephalon long, reaching the pits in front
of the eyes.

Thorax of ten segments, axial lobe very wide, furrows shallow.
Pygidium short and evenly convex, with only a trace of a concave
margin.

It will be at once noted that this species is much like B. niagarensis,
but has large eyes far back, and the pygidium is shorter and with less
depressed margin. It differs from B. cuniculus in having long in-
stead of short dorsal furrows on the cephalon, and in lacking the rim
at the front. Of the species found near Chicago, B. chicagoensis
(Weller) is very similar to this but that species has a much shorter
and more abruptly deflected cephalon.

Measurements: — The type is 46 mm. long; cephalon 19 mm. long,
26 mm. wide; eye 7 mm. long and 2 mm. from posterior margin;
thorax 14 mm. in length; pygidium 23 mm. long, 25 mm. wide. A
large cephalon is 34 mm. long, 46 mm. wide; eye 11 mm. long, 4 mm.
from posterior margin.

Formation and locality: — Quite common in the Niagaran at Wauwa-
tosa, -Wise.

BuMASTUS DECIPIENS, Sp. nOV.

Plate 1, fig. 1, 2.

Exceedingly like B. dayi is a species of which the collection contains
only five or six specimens, two of which are approximately entire.
The specimens are of about the same size as the smaller ones of B.
dayi, and the principal difference between the two is that B. decipiens

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 19

has short glabellar furrows and no lip on the free cheek, while B. dayi
has long furrows and a lip on the free cheek. The eye of B. decijnens
seems also to be a little longer and further back than in B. dayi. The
pits in front of the eyes are exceedingly faint.

Buviastus transversalis (Weller) likewise has a short head, large
eyes, and short dorsal furrows, but is much wider than B. decipiens.
B. armatus (Hall) is much like B. decipiens but has spines at the genal
angles.

Measurements: — The holotype (Plate 1, fig. 2) an imperfect
cranidium, is 19 mm. long; the eye is 7.5 mm. long, and 2 mm. from
the posterior margin. The figured paratype (Plate 1, fig. 1) is about
51 mm. long; cephalon 21 mm. long; thorax 15 mm. long. A second
small, unfigured paratype has the cephalon 12 mm. long, 16 mm.
wide; eye 5 mm. long; 1 mm. from posterior margin; thorax 8 mm.
long; pygidium 12 mm. long, and about 16 mm. wide.

Formation and locality: — All the specimens are from the Niagaran
at Wauwatosa, Wise.

BUMASTUS TENUIS, Sp. nOV.

Plate 1, fig. 6, 7 and ? 11.

This species is like B. decipiens, rare, and the collection contains
five specimens showing the cranidium only. It is characterized by
its slight convexity and the very wide concave lip at the front. The
dorsal furrows are long, reaching to the pit in front of the eyes. These
pits are themselves very far forward and close to the margin. The eyes
are large and close to the posterior margin.

This species of course suggests B. cuniculus, but the lip is much
wider, the dorsal furrows longer, and the pits in which they end very
close to the margin. It has a flatter cephalon and wider lip than any
known Bumastus.

A pygidium (Plate 1, fig. 11) which does not seem to belong with any
other species found at Wauwatosa is tentatively referred to B. tenuis.
It is too short and not convex enough for B. niagarensis, and has too
wide a concave border for B. cuniculus, B. dayi, or B. decipiens.

Measurements: — The type cranidium is 33 mm. long and 34 mm.
wide at the palpebral lobes. The eye was approximately 8 mm. long
and 2 mm. from the posterior margin.

Formation and locality: — All the specimens are from the Niagaran
at Wauwatosa, Wise.

20 bulletin: museum of comparative zoology.

BUMASTUS INSIGNIS (Hall).

Illaenus insignis Hall, 18th Rept. N. Y. state cab. nat. hist., 1865, p. 27, f. 5,
6 adv. sheets; 20th Rept. N. Y. state cab. nat. hist., 1868, p. 331, f. 5, 6,
pi. 22, f. 13, 14; 1870, rev. ed., p. 419; f. 10. 11, pi. 22, f. 13, 14. WeUer,
Bull. Chicago acad. nat. sci., 1907, no. 4, pt. 2, p. 215, pi. 17, f. 1-5. Not
of Salter, Whitfield, Foerste, and Kindle. (See WeUer, op. cit.).

It is interesting to note that though Hall cites Waukesha and
Milwaukee, Wise, first in his list of localities for this species, the
species is not known at all from Wisconsin, and the specimens which
Hall figured were undoubtedly from near Chicago. Whitfield's
Illaenus insignis was B. cuniculus (Hall). The English specimens
figured by Salter as belonging to this species, show the same long
furrows and cephalon with a lip, but the front of the head is not so
pointed. The eyes of the English form are likewise less- elongate and
further back than in specimens from near Chicago. It seems probable
that the English form deserves a distinct name, and B. pomiata, the
name used by Salter on the plate of his publication, is still available.

BuMASTUs loxus Hall.

Illaenus (Bumastus) barriensis ? Hall, Geol. surv. Wise, 1862, 1, p. 433, (no
description); 18th Rept. N. Y. state cab. nat. hist., 1865, p. 28 adv.
sheets; 20th Rept. N. Y. state cab. nat. hist., 1868, p. 332.

Illaenus ioxus Hall^ 20th Rept. N. Y. state cab. nat. hist., 1868, p. 387, fig.,
pi. 22, f. 4-11, pi. 23, f. 1. Whitfield, Geol. Wise, 1882, 4, p. 304, pi. 21,
f. 11, 12. Foerste, Proc. Boston soc. nat. hist., 1890, 24, p. 268, pi. 5,
f. 20. Van Ingen, School of mines quarterly, Columbia univ., 23, 1901,
p. 35 (no description). Kindle, 28th Ann. rept. Dept. geol. and' nat. res.
Indiana, 1904, p. 480.. pi. 22, f. 7, pi. 23, f. 3. Weller, BuU. Chicago acad.
sci., 1907, no. 4, pt. 2, p 222, pi. 18, f. 1-3.

Illaenus (Bumastus) ioxus Hall, 20th Rept. N. Y. state cab. nat. hist., 1870,
rev. ed., p. 420, f. 12, pi. 22, f. 4-10. ? 11th Rept. Dept. geol. and nat.
hist. Indiana, 1882, p. 335, pi. 38, f. 14, non 13; Trans. Albany inst.,
1883, 10, p. 76.

The status of the name of this species is somewhat like that of
Cheirurus niagarensis. In the Pal. N. Y., 2, Hall identified the rather
common large Bumastus of the Rochester shale as B. barriensis
Murchison, and in his earlier references to the Wisconsin specimen he

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 21

used the same name. When he came to figure a Wisconsin specimen,
however, he proposed a new specific name for it, and the name was so
obviously suggested by the Wisconsin specimens that I agree with
Weller that those specimens should be considered the types of B. ioxus.

Professor Weller has suggested that the specimens from New York
may belong to another species. I have investigated the point as fully
as the material at my command would permit, and have not so far
been able to find any really good characteristics on which to base a
separation. The best specimens from New York are usually small,
and considerably flattened. The study of better material will prob-
ably reveal characteristics not now evident, and I have therefore
omitted from the synonymy the references to the New York specimens.

Hall figured a pygidium which he assigned to the species, undoubt-
edly correctly, but up to the present the thorax has been unknown.
The M. C. Z. (Day collection), contains a large specimen, whose label
states that it is the "only perfect specimen found at Racine." It is
not exactly a perfect specimen, though it retains cephalon, parts of ten
thoracic segments and the pygidium. The axial lobe of the thorax
,is somewhat less wide than one would have expected from the large
size of the animal, but, being 62% of the total width, is about the
general average among the bumastids. The pygidium does not show
an actual concave border, but there is a very decided flattening of the
curve of the profile at the back.

As the specimen is preserved, the pygidium is somewhat unnaturally
drawn in, so that the actual length is not shown. On the other hand,
the last thoracic segment is displaced from the others and there is a
considerable space between the thorax and cephalon at the anterior
end, and between the thorax and pygidium behind. The length of
this specimen, therefore, gives only a rough approximation of the
correct length. Incidentally it should be noted that the cephalon of
this species has a large median tubercle near the posterior margin.
It is shown in Hall's figure, but omitted from Weller's.

Measurements: — Length, about 180 mm.; cephalon 75 mm. long,
110 mm. wide; thorax about 75 mm. long, about 100 mm. wide at
middle, axial lobe 62 mm. wide; pygidium 65 mm. long, 102 mm. wide.
A well-preserved pygidium is 68 mm. long and 100 mm. wide.

Formntion and locality: — Hall mentions Waukesha and Wauwa-
tosa as localities for this species, but in very extensive collections from
these places no specimens of this species are present, while we have a
number of specimens from the Racine dolomite at Racine, Wise.

22 bulletin: museum of comparative zoology.

BuMASTUS graftonensis Meek and Worthen.

Illaenus (Bumastus) graftonensis Meek and Worthen, Proc. Acad. nat. sci.

PhU., 1870, p. 54; Geol. surv. Illinois, 1875, 6, p. 508, pi. 25, f. 4.
Buniastus sp. ind. Meek and Worthen, Geol. surv. Illinois, 1875, 6, pi. 24. f. 3.
Illaenus graftonensis Weller, Bull. Chicago acad. sci., 1907, no. 4, pt. 2, p. 223,

pi. 16, f. 4-6.

Only the cephalon of this species has previously been known, but
the Day collection contains two complete, though somewhat flattened
specimens from Waukesha, Wisconsin, where this species seems to be
fairly common.

The cephalon is too well known to need further description, except
to note that as in B. ioxus, there is a prominent median tubercle be-
tween the eyes and near the posterior margin.

The thorax has ten segments, a broad axial lobe, rather well-defined
dorsal furrows. The pleura of the thoracic segments are more promi-
nent and ridged, and not so flat as in most species of Bumastus. The
pygidium is short and moderately convex, with a flattening around
the margin, but not a real concave border. The pygidium resembles
that of B. ioxus, but is shorter and wider, the average ratio of length
to width in B. ioxus being .64 and in the two pygidia of B. graftonensis
which we have, .54.

Formation and locality: — Nine specimens (M. C. Z. coll.) are from
the Niagaran at Waukesha, Wise, a locality from which this species
has not previously been reported.

Bumastus indeterminatus (Walcott).
Plate 2.

Illaenus indeterminatus Walcott, 31st Ann. rept. N. Y. state mus. nat. hist.,
1877, p. 19 adv. sheets; 31st Ann. rept. N. Y. state mus. nat. hist., 1879,
p. 70.

Illaenus cf. /. indeterminatus Clarke, Pal. Minn., 1897, 3, pt. 2, p. 716, f. 24.

Bumastus indeterminatus Raymond and Narraway, Ann. Carnegie mus., 1908,
4, p 253, pi. 62, f. 8, 9.

The M. C. Z. contains the type of this species, and it is figured
(Plate 2) for the first time. The specimen consists of a good cephalon,
part of one free cheek, a very much dislocated thorax, of which only
se\en segments can be definitely made out, and a very fine pygidium.

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 23

The pygidium is considerably longer and more convex than the cepha-
lon, and the axial lobe is evidently very wide and without very strong
dorsal furrows.

The dorsal furrows of the cephalon fade out just before reaching the
pits in front of the eyes, but this is probably an individual variation.
These pits are not directly in front of the eyes as in most species of
Bumastus, but considerably inside the projection of a line drawn
through the length of the eye.

The pygidium is very convex, with concave slopes to the lateral
and posterior margins.

Formation and locality: — The type, M. C. Z. no. 650, is from Russia
(Newport), Herkimer Co., New York. The horizon is the Leray-
Black River (Ordovician).

ACTINOLOBUS AMERICANUS, Sp. UOV.

Plate 1, fig. 4, 5.

The M. C. Z. contains a single pygidium which, on account of its
great length and wide flat border, is referred to Actinolobus, a genus
previously unrepresented in this country. The specimen is damaged
on the right side and has been restored in plaster.

There is also a break on the left side not shown in the figure, but
the outline is believed to be accurate. The specimen is 38 mm. long
and 20 mm. wide at the front. At the widest part, the border has a
width of 10 mm. and is somewhat concave. The central part of the
pygidium is strongly convex. The anterior margin indicates that the
axial lobe of the thorax was wide, though not so wide as in the species
of Bumastus found with it.

Formation and locality: — From the Racine dolomite at Racine,
Wisconsin.

LICHADIDAE Hawle and Corda.

Trochurus nasutus (Weller).

Plate 3, fig. 1, 2.

Dicranopeltis nasuta Weller, Bull. Chicago acad. sci., 1907, no. 4, pt. 2, p. 240,
pi. 22, f. 5-7.

A figure of the specimen of this species in the Day collection is
introduced in order to show the wav in which the frontal lobe of the

24 bulletin: museum of comparative zoology.

glabella tapers into a spine, this feature not being correctly shown by
Weller's figures. I had at first believed that this specimen repre-
sented a distinct species, but fortunately was able to see the type in
Mr. Teller's collection. That the species belongs to Trochurus in-
stead of Dicranopeltis is shown by the dorsal furrows, which curve
inward instead of outward, at the posterior ends.

Formation and locality: — The specimen is from the Niagaran at
Wauwatosa, Wisconsin.

ODONTOPLEURIDAE Burmeister.

Ceratocephala goniata Warder.

Plate 3, fig. 3-5.

Ceratocephala goniata Warder, Amer. journ. sci., 1838, ser. 1, 34, p. 378, fig.
Clarke, 44th Rept. N. Y. state mus. nat. hist., 1892, p. 91-100, pi. 1,
f. 1. Kindle, 2Sth Ann. rept Dept. geol. and nat. res. Indiana, 1904,
p. 480, pi. 24, f. 13. Weller, Bull. Chicago acad. sci., 1907, no. 4, pt. 2,
p. 255, pi. 23, f. 1-2. Raymond, Bull. Victoria mem. mus., 1913, 1, p. 38.

Acidaspis danai HaU, Geol. surv. Wise, 1862, 1, p. 432 (no description); 18th.
Rept. N. Y. state cab. nat. hist., 1865, p. 28, adv. sheets; 20th Rept. N. Y.,
state cab. nat. hist., 1868, p. 333, pi. 21, f. 8-9; 1870, rev. ed., p. 423,
pi. 21, f. 8, 9.

Acidaspis ida WincheU and Marcy, Mem. Boston soc. nat. hist., 1865, 1, p.
106, pi. 3, f. 13.

The cephalon of this species is, thanks to Professor Weller, now well
known, but hitherto the thorax and pygidium have not been noticed.
The Day collection in the M. C. Z. contains parts of two pygidia and
a fragment showing a portion of five segments of the thorax. The
pygidium is like that of Acidaspis portlocki Barrande, with a large
median spine, on each side of which are two smaller spines, then a
large spine, and finally an outer small spine. Each spine gives off
small thorns on each side. The thorax is similar to that of C. verneuili
Barrande. However, as only the central portion has been seen this
part of the body would not be expected to show specific characteristics.
The cephalon seems to be more nearly allied to C. vesiculosa than to C.
verneuili, and a cephalon in the M. C. Z. gives indication that marginal
spines were present on the free cheeks. The pygidium found with the
cranidia at Wauwatosa are unlike those ascribed to C. vesiculosa by

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 25

Barrande, and, on the other hand, thou.^h the pygidia in the Day
collection are much like that of A. portlocki, their cephalons are very
unlike those assigned to A. portlocki by Barrande.

Formation and locality: — The specimens figured are from the
Niagaran at W,ajiwatosa, Wise.

ENCRINURIDAE Angelin.

Encrinurus reflexus, sp. nov.
Plate 3, fig. 7, 8.

The Day collection contains several pygidia of an Encrinurus which
is larger and differs in various ways from any species of this genus
heretofore described from Silurian strata of this country.

Cephalon and thorax unknown.

Pygidium large for the genus, triangular, pleura bent abruptly
downward, the posterior end somewhat turned up. Axial lobe very
long, tapering backward to a point. It is marked by about thirty
rings which are prominent and sharp toward the front, but very faint
at the posterior end. With the exception of two at the anterior end
the rings do not cross the axial lobe, but leave a narrow smooth space
along the median line. Along this smooth lane are disposed eight
pustules, approximately evenly spaced. On the pleural lobes are
eight pairs of broad flat ribs which curve backward, making a rather
abrupt turn near their outer ends. They do not reach quite to the
margin, and end in blunt free spines which project from the sides a
little above the margin. At the posterior end the last two ribs from
each side converge alongside the axial lobe, and, with a small median
spine, project beyond the end of the axial lobe. The ribs have small
pustules scattered somewhat irregularly over them, and not aligned
in longitudinal rows. Nearly every rib has a pustule at its inner end
•and one near the middle. Some of the ribs have only these two, but
the longer ones near the front have another.

Measurements: — The more complete of the cotypes is 28 mm. long,
26 mm. broad at the front. The axial lobe is 9 mm. wide at the front
and 26 mm. long.

Comparison with other species : — Nine species of Encrinurus have
previously been described from the Silurian of America, but most of
them are of much smaller size than the present species and only two

26 bulletin: museum of comparative zoology.

species, E. deltoideus Shumard and E. nereus Hall, have as many as
eight pairs of ribs. Most of the species have seven pairs, one, E.
americanus Vogdes, has six, and E. elegantulus Billings has only five.
The species may be taken up in alphabetical order.

E. americanus Vogdes (Description new Crustacea from Clinton of
Georgia, 18G6, p. 1), has only six pairs of ribs on the pleura, and no
pustules, thus ruling it out at once.

E. deltoideus Shumard (Geol. Missouri, 1855, p. 198, pi. B, f. 10), is
similar in many ways to E. reflexus, having the rings very numerous,
twenty-four in nvimber, and interrupted by a smooth lane. There
are, however, no pustules except very indistinct granules, on this lane,
and the ribs, of which there are eight pairs, are likewise smooth.

E. egani S. A. Miller (Journ. Cine. soc. nat. hist., 1880, 2, p. 254,
pi. 15, f. 1, lb), has a long terminal spine, only seven pairs of ribs,
which are narrow with wide interspaces, and while the axial lobe of
the pygidium is on the same plan as in E. reflexus, it has fewer rings
and tubercles.

E. elegajitulus Billings (Cat. Silurian fossils Anticosti, 1866, p. 62),
has only five pairs of ribs, the median lane on the axial lobe is without
tubercles, but the first eight rings cross it.

E. indianaensis Kindle (28th Ann. rept. Dept. geol. and nat. res.,
Indiana, 1904, p. 482, pi. 24, f. 14, 15), is a very peculiar species, not
at all of the same type as the one under discussion. It has fifteen
rings on the axial lobe and ten pairs of ribs. Each rib crosses the axial
lobe and has three to five tubercles.

E. nereus Hall (20th Rept. N. Y. state cab. nat. hist., 1868, p. 375,
pi. 21, f. 15) has no flattened lane along the top of the axial lobe, and
no nodes on the rings or ribs.

E. ornatus Hall and Whitfield (Pal. Ohio, 1875, 2, p. 154, pi. 6, f . 16) is
c^uite similar to the present species, but has only seven pairs of ribs, and
the ribs themselves are narrower and the spaces between them wider.
There are also only twenty rings and five nodes on the axial lobe.

E. thresheri Foerste (Bull. Sci. lab. Denison univ. 1887, 2, p. 101, pi.
8, f. 26) is a small species which is similar to the last and to E. reflexus.
There are, however, only seven pairs of ribs, which are themselves
exceedingly narrow; there are also only eighteen rings on the axial
lobe, and six pustules on the smooth lane.

E. tuber culifrons Weller (Bull. Chicago acad. sci.. 1907, no. 4, pt. 2,
p. 259, pi. 24, f. 12, 13) is a small form with a short wide pygidium
which is without nodes and the rings of which cross the axial lobe
without interruption.

Formation and locality: — From the Niagaran at Wauwatosa, Wise.

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 27

CALYMENIDAE Milne Edwards.
Calymene niagarensis Hall.

Calymene niagarensis Hall, Geol. N. Y., 1843, pt. 4, p. 102, f. 3 on p. 101; tab.

org. rem. 10, f. 3.
Calymene hlumenbachi var. niagarensis Hall, Pal. N. Y., 1852, pt. 2, p. 307,

pi. 67, f. 11, 12.

Calymenes are difficult fossils to differentiate satisfactorily, but it
is possible to draw a little closer limits to some of the species than has
been done in the past. The Silurian species do not present so difficult
a problem as do those in the Ordovician, specific characteristics being
apparently more fixed and constant in later times. Before venturing
to separate two new species, it is best to direct attention for a moment
to the well-known (in name) Silurian form.

The name was applied originally by Hall to specimens from the
Rochester shale at Lockport, N. Y. The figures and description show
the original specimens to have been of the Calymene hlumenbachi type,
that is with a narrow lip in front of the glabella, three pairs of glabellar
lobes, anrl pygidium with an impressed line on each rib, distinctly
bifurcating the outer portion. The ribs also reach practically to the
margin.

It seems that the species Calymene niagarensis should be restricted
to such trilobites as show these important characteristics of the types
and these may be seen in most of the Calymenes in the Rochester
shale. Another Calymene found at the same horizon, C. vogdesi
Foerste, has the same bifurcated ribs on the pygidium, but a much
longer snout-like lip in front of the glabella. It is also a much larger
form, one of the largest of the Calymenes.

In the Ordovician the common Calymene senaria of the Trenton
has the same type of bifurcated rib, while the later C. meeki Foerste,
so abundant in the Eden and IMaysville at Cincinnati, shows only a
trace of an impressed line on the ribs, and often the line is absent
entirel\'.

Calymene breviceps, sp. nov.

Plate 3, fig. 11.

■Calymene niagarensis Hall, 28th Rept. N. Y. state mus. nat. hist., doc. ed.,
1877, pi. 32, f. 8-15; mus. ed., 1879, pi. 32, f. 8-15; 11th Rept. Dept.
geol. and nat. hist. Indiana, 1882, p. 331, pi. 34, f. 8-15. Hall and
Clarke, Pal. N. Y., 1888, 7, pi. 1, f. 10-14.

28 bulletin: museum of comparative zoology.

The Calymene abundant at Waldron, Indiana, has always been
identified with C. niagarensis, but differs from that species in at least
two marked details. The first and most obvious characteristic is that
there is no lip, nor any furrow between the glabella and the rim, so
that the glabella reaches upon, and in some cases, overhangs the rim,
a feature usual in the Cheiruridae but extremely uncommon among
the Calymenidae. This gives the cephalon the high, short appearance
which suggested the name breviceps. On the pygidium the ribs reach
nearly to the margin but become faint on approaching it. Ordinarily
the ribs do not bear any median impressed line though traces of one
may be seen on some specimens.

This species is in many ways much like C. celebra, the next species
described.

Formation and locality: — This species is so far known only from the
(Silurian) Waldron shale at Waldron, Indiana, where it is very com-
mon.

Calymene celebra, sp. nov.
Plate 3, fig. 9, 10.

Calymene hlumenhachii var. niagarensis Hall, Geol. surv. Wise, 1862, 1, p. 432.

Calymene niagarensis Hall, 18th Rept. N. Y. state cab. nat. hist., 1865, p. 30,
adv. sheets; 20th Rept. N. Y. state cab. nat. hist., 1868, p. 334; 1870,
rev. ed., p. 425 Weller, Bull. Chicago acad. sci., no. 4, pt. 2, p. 261, pi.
23, f. 9-10.

One of the most abundant of the trilobites of the Chicago area and
of southeastern Wisconsin is a Calymene which is constantly identified
as C. niagarensis. It is quite commonly found entire, but always so
far as I have seen in the condition of a cast of the interior. Moulds
of the exterior are common, but seldom complete.

The cephalon is like that of C. niagarensis, with a short lip and
narrow furrow in front of the glabella. The dorsal furrows are always
very deep and sharp, but this is due to the state of preservation. The
glabella tapers rather abruptly toward the front. The basal lobes are
large, rounded, almost isolated; the second lobes small and rounded,
the intermediate "extra lobes" not very prominent. The third lobes
are very small and the fourth ones just barely indicated. The frontal
lobe is short and rather square at the front. The eyes are close to the
glabella and opposite the furrows between the second and third pairs
of lobes.

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 29

The pygidium is the most characteristic portion of the animal.
The axial lobe is narrow, well defined, and has rings. The pleural
lobes show four pairs of narrow ribs, without impressed line, which
reach only halfway to the margin. The fourth of the four pairs are
very faint and short. Each pleural lobe is thus divided into a small
triangular ribbed portion near the axial lobe and a much longer smooth
portion below. This pygidium presents the greatest possible contrast
to C. niagarensis, in which the ribs are more conspicuous near the
margin than near the axial lobe. The peculiarities of the pygidium
have doubtless been noticed before, and probably have been explained
as due to the state of preservation, the specimens all being internal
casts. Internal casts of either cephala or pygidia of trilobites are
practically always less and not more smooth than the exteriors, how-
ever, and cleaned interiors of C. senaria, C. breviceps, and C. meeJci, all
show that Calymene follows the general rule. Calymene celebra shows
a halfway stage to what is achieved in C. clintoni Vanuxem, namely,
a pygidium with smooth pleural lobes. The latter species is too far
removed from the Calymenes with typical ribbed pygidia to be in-
cluded in the same genus.

Formation and locality: — - Calymene celebra is common in the
Niagaran of the Chicago district in northern Illinois, in the same
portion of the Silurian in southeastern Wisconsin, and also near
Madison, Indiana, and Eaton, Ohio.

LiocALYMENE, gen. nov.

Calymeninae (as distinguished from the Homalonotinae) with
distinct glabella, three pairs of glabellar lobes, narrow thorax, pygidium
with ringed axial and smooth pleural lobes. Type, Hemicrypturus
clintoni Vanuxem.

Liocalymene cJintoni, in perfect preservation, appears to be an
exceedingly rare fossil. The single specimen in the M. C. Z. is in
about the same condition as that figured by Hall (Pal. N. Y., 2, p. 298,
pi. A 66, f. 5a), and is from the Clinton shale at Clinton, Herkimer Co.,
N. Y.

30 bulletin: museum of comparative zoology.

CHEIRURIDAE Salter.

Cheirurinae Raymond.

Cheirurus niagarensis (Hall).
Plate 4, fig. 4, 5, 6, 9.

Ceraurus insignis Hall, Pal. N. Y., 1852, 2, p. 303, pi. 67, f. 9, 10.

Ceraurus niagarensis Hall, 20th Rept. N. Y. state cab.' nat. hist., 1868, p. 376,
? Whiteaves, Geol. surv. Canada. Pal. foss., 1884, 3, pt. 1, p. 42; 1895,
3, pt. 2, p. 107. ? Van Ingen, School of mines quarterly, Columbia univ ,
■ 1901, 23, p. 35 (no description). Kindle, 28th Ann. rept. Dept. geol. and
nat. res. Indiana, p. 483, pi. 23, f. 1, 2, pi. 24, f. 8. Weller, Bull. Chicago
acad. sci., 1907, no. 4, pt. 2, p. 263, pi. 24, f. 20, non 21.

The name Ceraurus niagarensis appears for the first time on p. 376
of the first edition (1868) of the 20th Ann. Rept. of the New York
State Cabinet of Natural History. Earlier in the same paper, (p. 335),
Hall referred certain trilobites from the Silurian at Wauwatosa and
other localities in Wisconsin to Ceraurus insignis (Beyrich). On
p. 376 he states that he has reexamined the specimens and considers
them different from C. insignis. The name Ceraurus niagarensis is
used as a heading, but is not designated either as a new name or a new
species. On p. 427 of the Revised edition, published in 1870, the
remarks are reprinted, but the letters N.S. follow the name. In
neither case is there any description of the species given, but the
plates contain representations of an imperfect cranidium and a broken
hypostoma.

The next use of the name by Hall was in 1879, in the 28th Rept. of
the N. Y. State Museum, p. 189. He here describes the pygidium at
some length from specimens obtained at Waldron, Indiana, and
remarks, at the end of his description: "From the above it will be
seen that the separation first made in the revised edition of the 20th
Rept. St. Cab., was necessary, and that it constitutes a distinct spe-
cies." The single figure given represents a pygidium.

After studying collections from a number of localities it becomes
evident that the American forms now referred to Ceraurus niagarensis
include two or three species, and it is therefore, necessary to determine
the type for C. niagarensis. From the absence of description accom-

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 31

panying the first use of the name niagarensis it would appear that Hall
did not apply the name to the Wisconsin specimens alone, but meant
to assign this name to all American forms previously ascribed to
Cheirurus insignis or Ch. hivincronatus . This idea is strengthened
by the fact that he states that the New York and Wisconsin specimens
show the same characteristics. He also refers to Roemer's Ceraurus
bimucronatus from Tennessee. This idea is still further strengthened
by the remark quoted above from his description of the Waldron
fauna. If this is the case, then we should return to Hall's first descrip-
tion of a Ceraurus insigiiis in America to get at his idea of the species.
If we take the other view, that the specimens from Wauwatosa, which
seem to have been the first ones to cause Hall to doubt the correctness
of his reference of all the Silurian cheirurids to the Bohemian species,
are the real types of C niagarensis, we are confronted by the fact that
he did not describe his specimens, and, moreover, he was evidently in
doubt about them, as evidenced by his pleasure at finding distinguish-
ing features in the pygidia from Waldron. This latter description
was the first real description published after the name niagarensis was
proposed, and it might well be argued that the last described of the
group should be the type. It seems simpler, however, to accept what
was Hall's evident intent, and believe that in proposing the name
Jiiagarensis he was merely proposing a new name for the specimens he
had previously described as Ceraurus insignis.

The first description of Ceraurus insignis Beyrich, by Hall occurs
in vol. 2 of the New York State Paleontology, 1852. On page 300
there is mentioned, without description, a glabella from the Clinton
which is figured on plate 66A. On page 306 of the same volume Hall
describes two cranidia from the Rochester shale at Rochester, N. Y.
Both specimens are figured. As these are the first American speci-
mens which are both figured and described, I propose to designate as
the type of Cheirurus niagarensis (Hall) the one represented in fig. 10,
pi. 67, of the above volume. This specimen is in the American Museum
Natural History, No. 1827.

The specimen so designated is a typical Cheirurus, with the glabella
expanding rather rapidly forward, the frontal lobe occupying less than
half the length of the glabella, and the first two pairs of glabellar
furrows nearly straight, and following a direction approximately
parallel to the posterior margin of the cephalon. Their inner ends are
separated by a smooth space equal in width to about half the glabella.
The eyes are near the dorsal furrows, and about opposite the second
glabellar furrows.

32 bulletin: museum of comparative zoology.

This type of cranidium is quite often seen in collections from the
Rochester shales, but one also sees another type, the one which Hall
figured from the Clinton. The cranidium is similar to the one just
described, but the glabellar furrows, instead of being short and
straight, are long, curve backward, and their inner ends almost meet.
This type of head deserves to be recognized as distinct from the other;
it is the type of head figured by Hall from Wisconsin and though both
types are very common there, this is by far the more abundant. To
this same type, though possibly not to the same species, belongs the
cephalon of Ceraurus himucronatus figured by Roemer.

No entire specimen of Cheirurus has, so far as I know, been found
in America, and it is therefore difficult to decide what pygidium shall
be associated with each type of cephalon. It would appear that no
Cheirurus pygidium had been figured from New York. The M. C. Z.
possesses a single small pygidium of a Cheirurus from the Rochester
shale at Rochester, N. Y. It is of the familiar Cheirurus insigiiis type,
with three pairs of long slender spines, and a short median spine. It
is very dift'erent from the pygidium from the Waldron shale ascribed
to Ceraurus niagarensis by Hall, for that specimen was described as
having broad flat spines, each spine with a depressed line on the sur-
face.

Pygidia found at Wauwatosa are like the one from Rochester and
it seems probable that this type of pygidium is to be referred to
Cheirurus niagarensis.

The following description of Cheirurus niagarensis is based on three
glabellas (M. C. Z. €25) and a pygidium (M. C. Z. §24) from the
Rochester shale at Rochester, N. Y., and a cranidium with three
segments attached and an associated hypostoma, from Wauwatosa,
Wise. (M. C. Z. 626). A large cranidium with a part of the thorax
(M. C. Z. 627) from Wauwatosa was also consulted.

Cheirurus niagarensis (Hall) restricted.

A Cheirurus of medium size. Cranidium semicircular in outline,
gently convex, the glabella forming the highest and most prominent
part, but not standing much above the cheeks. The glabella reaches
the front of the cranidium, expands toward the front, and is widest
at the middle of the frontal lobe. Dorsal furrows narrow and sharp,
but not very deep. Glabellar furrows short, sharp, the first two pairs
extending only a short distance onto the glabella. Their direction is

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 33

approximately vertical to the axis of the animal, but they are usually
not absolutely straight, but bend a little forward in mid length. The
posterior furrows run diagonally inward and connect with the neck
furrow, as in the genus generally. The eyes are close to the glabella
and opposite the second pair of glabellar furrows. Free cheeks small,
pitted, with a smooth, convex rim. Fixed cheeks pitted. Glabella
granulose. The associated hypostonia is roughly tetragonal, the
surface with sharp, scattered pustules, and the posterior margin nearly
straight, without spines at the angles. The furrow around the body
portion is wide and deep.

Of the thorax only three segments are known. The axial lobe is
narrow, the inner part of the pleural lobe is crossed by a narrow
diagonal furrow which separates two triangular nodes, and there is a
prominent node at the fulcrum. Beyond the fulcrum the pleuron
projects as a blade-like spine.

The pygidium is short, with three pairs of slender spines which are
oval in section and unfurrowed. The last pair extend further back
than the ones ahead of them. A median spine is present, but very
short. The axial lobe is narrow, cone-shaped, with the point backward,
bearing three rings and a node. The pleural lobes are narrow, and
.show a single short divided rib on either side at the anterior end.

This species is very much like Cheirurus insignis Beyrich. The
glabella seems to be a little shorter and wider in the American form,
and the Bohemian species has the eyes further from the glabella and
has eyelines. Of the latter, however, the specimens in the M. C. Z.
show a trace. The hypostoma of the Bohemian form is similar to that
of the American species, but the posterior margin is somewhat wider
and more flattened. The pygidia are practically the same, though
the median spine is a little stronger in Ch. insignis.

It will be seen from the above description, that if we restrict Cheiru-
rus niagarensis to those forms which Hall first identified with Cheirurus
insignis, we eliminate both the forms which caused him to change his
mind about the identification, and propose the new name niagarensis.
This would seem to vitiate the argument above, but it must be re-
membered that Hall did not recognize that he was dealing with more
than one species, and he did not apply the new name to any definite
specimens. In fact, it would seem that he did not become fully con-
vinced that a new name was needed till he studied the pygidium from
Waldron, and if the name niagarensis is not to be interpreted as has
been done here, it would be almost impossible to decide whether the
Wisconsin or the Waldron specimens should be selected as the types.

34 bulletin: museum of comparative zoology.

In spite of the similarity of Cheirurus niagarensis, as above defined, to
Cheirurus insignis, I believe that a separate specific name should be
maintained, especially as the discovery of further material may show
unsuspected differences between the two.

Measurements: — A cranidium from Rochester (M. C. Z. 625).
Length 20 mm., width 35 mm. ; width of glabella at neck-ring 12 mm.,
at front 16 mm.; length of frontal lobe 10 mm. A cranidium from
Wauwatosa, (M. C. Z. 626). Length 9 mm., width 18 mm.; width
glabella at neck-ring 5 mm., width at front 8 mm. ; length of frontal
lobe, 3 mm.

Formation and locality: — Rochester shale at Rochester and Lock-
port, N. Y., Niagaran in Indiana, Wisconsin and Illinois.

Cheirurus welleri, sp. nov.
Plate 3, fig. 6; Plate 4, fig. 7, 8, 10.

Ceraurus insignis Hall, Pal. N. Y., 1852, pt. 2, p. 300, pi. 66A, f. 4. Geol.

surv. Wise, 1862, 1, p. 433; 18th Rept. N. Y. state cab. nat. hist., 1865,

p. 31, adv. sheets; 20th Rept. N. Y. state cab. nat. hist., 1868, p. 335.
? Ceraurus bimucronatus Roemer, Silurian faun. west. Tenn, 1860, p. 80, pi. 5,

f. 19.
Ceraurus niagarensis HaU, 20th Rept. N. Y. state cab. nat. hist., 1868, p. 376,

pi. 21, f. 10-11; 1870, rev. ed., p. 427, pi. 21, f. 10, 11 ; 11th Rept. Dept.

Geol. and nat. hist. Indiana, pi. 33, f. 10, non pi. 34, f. 16. WeUer, Bull.

Chicago sci., 1907, no. 4, pt. 2, pi. 24, f. 21 (non 20).

The most abundant American Silurian cheirurid in the collections
of the M. C. Z. is one whose glabellar furrows nearly cross the glabella,
and is therefore the nearest approach to a Crotalocephalus so far found
in America. This type has long been known, but constantly confused
with Ch. niagarensis or Ch. insignis. As has already been mentioned,
Hall figured one in 1852, and Roemer a similar one in 1860. The two
species sometimes occur together, as at Wauwatosa, but Ch. welleri is
easily recognized by its long glabellar furrows. A second distinction
is that the posterior margin of the hypostoma is rounded in Ch.
welleri and straight in Ch. niagarensis. Further, the hypostoma of the
latter species is tuberculated and of the former smooth.

A large Cheirurus with approximately semicircular cephalon. The
glabella is long, expands gradually forward. The frontal lobe occu-

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 35

pies less than half the length, and the furrows are long, curve back-
ward, the last pair meeting, as usual, and the inner ends of the other
pairs being very close together. The free cheeks are small, the eyes
opposite the ends of the second pair of glabellar furrows and near the
dorsal furrows. The fixed cheeks are wide, with coarse pits, and the
genal spines are short and slender. The glabella seems to be devoid
of granulation.

Of the thorax, only five segments have been seen. It seems to be
in all respects like that of Ch. niagarensis. No pygidium can defi-
nitely be assigned to this species. All the specimens found at Wauwa-
tosa seem to agree with the pygidium from Rochester which has been
assigned to Ch. niagarensis. Weller has, however, figured a cheirurid
pygidium found near Chicago, in which the central spines of the
pygidium are shorter than the others. This pygidium is in this respect
unlike the one here assigned to Ch. niagarensis and may belong to Ch.
welleri. In cephalon and thorax, Ch. welleri is closely allied to Ch.
quenstedti Barrande, of Bohemia. This species has the inner spines
so short that the pygidium appears to possess only two pairs of spines.
It may be that Ch. welleri has a pygidium intermediate in form be-
tween Ch. insignis or Ch. niagarensis, and Ch. quenstedti.

Measurements: — A large cranidium, one of the cotypes (M. C. Z.
14) is 33 mm. long and 62 mm. wide; the glabella is 19 mm. wide at
the neck-ring and 25 mm. wide at the frontal lobe; the frontal lobe
is 15 mm. long. This has about the same length as Ch. dilatatus, but
the disparity of the other dimensions should be noted. The figures
for Ch. welleri are always given first. Length, 33, 35, width, 62, 52;
width glabella at neck, 19, 20, at frontal lobe, 25, 31. The longer
specimen is the narrower, and has a wider glabella, thus showing a
great reduction of the cheeks. The cranidium of a smaller cotype
(M. C. Z. 630) is 14.5 mm. long, 25 mm. wide, the glabella is 7 mm.
wide at the neck-ring, 11 mm. wide at the frontal lobe, and the frontal
lobe is 7 mm. long. Large specimens of this species seem to have
been abundant, and the largest glabella in the collection is 45 mm. long,
and, so far as I know, the largest American cheirurid. Restored with
the proportions of Ch. insignis, this trilobite would have a length of
145 mm. or nearly 6 inches. The Bohemian Eccoptochile claviger
(Beyrich) equals this size.

Formation and locality: — The types are from the Niagaran at
Wauwatosa, Wise. The species occurs also in the Clinton of New
York, the Waldron of Indiana, Silurian of Tennessee, and Guelph of
Ontario. It is probably the most cosmopolitan species of Cheirurus.

36 bulletin: museum of comparative zoology.

ChEIRURUS DILATATUS, Sp. nov.

Plate 4, fig. 1, 3.

Sphaerexochus romingeri ? Hall, 28th Pi,ept. N. Y. state mus. nat. hist., 1877,

doc. ed., pi. 32, f. 16.
Ceraurus (Cheirurus) niagarensis Hall, 28th Rept. N. Y. state mus. nat. hist.,

1879, mus. ed., p. 189, pi. 32, f. 16; 11th Ann. rept. Dept. geol. and nat.

hist. Indiana, 1882, p. 335, pi. 34, f. 16, non pi. 33, f. 10.

In the discussion of Cheirurus niagarensis (p. 30) frequent mention
has been made of the pygidium from Waldron which Hall figured.
This pygidium differs radically from the pygidia which have been
referred to Ch. niagarensis and Ch. welleri, in having broad, short
spines, each marked by a depressed line. This appearance of the
spines is probably due to crushing, in so far as the depressed line is
concerned, but the spines are decidedly shorter and broader than
those of the species previously described.

Cheirurids seem to be rare at Waldron for a search through an ex-
tensive collection from that locality in the M. C. Z. has revealed only
one good cranidium, one poor one, and a pygidium. The best speci-
men is a fairly well-preserved cranidium, having much the general
appearance of Ch. niagarensis, only larger. On a closer examination
of the proportions, however, it is seen that this form is longer and
narrower than the typical specimens of Ch. niagarensis, and the
glabella makes up a larger proportion of the cephalon. In a specimen
of Ch. niagarensis from Rochester (M. C. Z. 625) the length is .57 of
the width and in the specimen from Waldron (M. C. Z. 628) it is .67.
In the first species the width of the frontal lobe of the glabella is less
than half the width of the cephalon (.46). In the latter specimen it
is somewhat more (.60). These appear, in figures, relatively small
differences, but when the areas involved are compared it is at once
seen that the glabella of the Waldron specimen is much larger than
that of the specimens of Ch. niagarensis.

It is of course uncertain whether the pygidium described by Hall
is to be associated with the cranidium here discussed. It may belong
to the same species, and they are provisionally associated. The
cranidium is, however, made the holotype of the species, and the
pygidium a paratype.

A large Cheirurus. The cranidium is dominated by the glabella,
whose frontal lobe is more than one half as wide as the total width

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 37

of the head. The glabella expands rapidly forward, the frontal lobe
occupies less than one half the length, and the glabellar furrows are
like those of Cheirurus niagarensis. The free cheeks are small, the
eyes opposite the ends of the second pair of furrows, and a little
further from the glabella than in Ch. niagarensis. The surface of the
cheeks shows numerous pits, while the glabella is granulose. Genal
spine rather long and straight.

Only one kind of Cheirurus pygidium has been found at Waldron.
The spines are short and subequal in length, broad, and flat. The ribs
on the pleural lobes are distinctly separated by furrows, and the first
rib on each side has a short sharp diagonal furrow. The median un-
paired spine at the back is about one half as long as the ones adjacent
to it.

Measurements: — The type is 35 mm. long, 52 mm. wide; the
glabella is 31 mm. wide at the front and 20 mm. wide at the neck-ring;
the frontal lobe is 17 mm. long.

Forviaiion and locality: — Found only in the Waldron shale at
Waldron, Indiana.

Cheirurus patens, sp. nov.
Plate 4, fig. 2.

This species, like the previous one, is notable for the small size of
its cheeks. The glabella is of the insignis type, with short furrows,
but is rather more convex and prominent than in that species. It
expands rapidly toward the front, and the frontal lobe is a little longer
than in any other species of the genus, forming a trifle more than one
half the length of the glabella of the type. The outline of the frontal
lobe is almost exactly semicircular, and thus differs from most of the
other specimens seen, the ordinary outline being that of half an ellipse.
Since the first pair of glabellar furrows are a little further back than
on most species, the fixed cheeks which terminate here are shorter
than usual, and the eye is a little further back.

The cheeks are full of pits, and the glabella granulose. The pos-
terior glabellar lobes are not strictly triangular but pentagonal, a
statement which is true of most species of Cheirurus, but more obvious
than usual in this species.

Measurements: — Length about 31 mm., width, about 53 mm.;
width glabella at frontal lobe 26 mm., length frontal lobe 17 mm.

Forviaiion and locality: — The type is a single imperfect cranidium

38 bulletin: museum of comparative zoology.

from the "Niagara" at Cicero, Illinois (M. C. Z. 629). A second
specimen, tentatively referred to this species, is in the Museum of the
Geological Survey of Canada, and is from the Guelph of Hespeler,
Ontario.

Cheirurus tarquinius Billings.

Cheirurus tarquinius Billings, Proc. Portland soc. nat. hist., 1863, 1, p. 121,
pi. 3, f. 22.

This is a little-known species of the Ch. insignis group. It has a
short, wide cephalon, narrow triangular basal lobes on the glabella, a
short frontal lobe, and the first two pairs of furrows turn backward,
are quite straight, and are intermediate in length between those of
Ch. niagarensis and Ck. welleri. The species is especially characterized
by the forward position of the eyes, which are opposite the second
glabellar lobes, and the consequent small free cheeks and long fixed
cheeks. The genal angles appear to be spineless.

The type is No. 3081 in the Museum of the Geological Survey of
Canada. Associated with it is a pygidium from the same locality.
It is of the insignis type, with three pairs of spines, but the median
spine is shorter and more rounded than in either Ch. insignis or Ch.
niagarensis. A poorly preserved hypostoma in the same collection
has the posterior end more rounded than that of Ch. niagarensis, and
thus more like that of Ch. welleri.

Measurements: — The type (G. S. C. 3081) is 19 mm. long, 35 mm.
wide; and the glabella is 12 mm. wide at the neck-ring and 17 mm.
wide at the frontal lobe.

Formation and locality: — Middle Silurian at Port Daniel, Bay,
Chaleur, P. Q., Canada. Also reported by Billings from Masardis,
Maine.

Cheirurus hydei (Weller).

Ceraurus hydei Weller, Bull. Chicago acad. sci., 1907, no. 4, pt. 2, p. 264, pi. 24,
f. 22.

This species is of more than ordinary interest, from its resemblance
to a Ceraurus. The cephalon and thorax are those of a typical
Cheirurus, but the pygidium is that of Ceraurus. This at once raises
the question as to whether this is a Cheirurus which has developed a
Ceraurus-like pygidium, or whether it is a Ceraurus whose cephalon
and thorax have developed in a manner paralleling that of Cheirurus.

baymond: new and oll silurian trilobites. 39

There is still a third possibility, namely, that Ceraurtis hydei is the
young of Cheirurus niagarensis, with which it occurs. On this third
point, Weller states that the fixed cheeks of C. hydei lack the pitted
surface characterizing C. niagarensis, and that C. hydei has a border
all around the cephalon, while C. niagarensis lacks it in front of the
glabella. These facts seem to be borne out by the type which is now
before me, and Professor Weller might have added that the glabella
expand more rapidly in the young of C. niagarensis than in C. hydei,
and has deeper glabellar furrows. The eyes too, of the young of C.
niagarensis are much further back than those of C. hydei.

Against these differences we may, however, place the fact that the
thorax is alike in the two species, and more similar to the thorax of
Cheirurus insignis Bey rich than to any of the Ordovician species of
Ceraurus. In both Cheirurus niagarensis and Ceraurus hydei, the
part of each pleural lobe between the dorsal furrow and the fulcral
line is very much reduced, the diagonal furrow is very short, and the
two small nodes which it separates are narrow, and one directly in
front of the other, a point not brought out in Weller's somewhat
generalized figure. On the fulcral line there is a row of nodes, and
just inside this row is a longitudinal furrow parallel to the dorsal
furrows. Beyond the fulcral line, the pleura are free, not contiguous
as shown in Weller's figure. These same characteristics are shown in
two specimens of Cheirurus niagarensis from Wauwatosa, Wisconsin.

On the whole, it does not seem very probable that C. hydei is the
young of Cheirurus niagarensis, especially as there is another species,
Ceraurus nuperu^ (Billings) which has a Cheirurus-like cephalon and
Ceraurus-like pygidium. The choice seems to lie between calling
it a Ceraurus or a Cheirurus. Theoretically, it would seem that the
Ceraurus pygidium was more specialized, and, therefore, less apt to
be dupHcated than the Cheirurus head. Most of the other Cheiruri-
dae, except Ceraurus, have all the spines of the pygidium approxi-
mately equal.

In Ceraurus -pleurexantherrvm there is a tendency in some specimens
to have the basal lobes of the glabella triangular instead of square,
and in Ceraurus viisneri the glabella occupies a large part of the
cephalon, and the cephalon is long. Further, Ceraurus reaches the
climax of size and abundance in the Trenton, the late species being
smaller, and the specimens rarer. As to the thorax, I have shown that
this portion of the test changes in parallel directions in many lines of
the Asaphidae, and the same might well happen in the Cheirurudae.
On the other hand, one would not expect a decadent race to show new
characters similar to those of a race which is at its best.

40 bulletin: museum of comparative zoology.

Though no other cheirurid exactly dupHcates the Ceraurus pygi-
diuni, there are numerous cases among the trilobites referred to the
genus Cheirurus in which there is a reduction of the inner pairs of
spines of the pygidium. Thus, Weller has figured a Cheirurus pygi-
dium from Lemont, near Chicago, in which spines of the inner pair
are shorter than the others. In Cheirurus qucnsiedti there are only
two pairs of spines, the inner pair being reduced to mere rudiments.
In Cheirurus hawlei there is a still further reduction, so that there is
only one pair of long spines, thus producing a pygidium which is a
parallel to that of Ceraurus, though differing considerably from it in
detail. The typical number of segments in a Cheirurus pygidium
seems to be five, a protopygidium and four pairs of coalesced segments
which originally had free spines. Among the species referred to
Cheirurus by Barrande may be seen Ch. minutus Barrande with four
pairs of spines, Ch. bifurcatus with four pairs, the central pair partly
united, Ch. insignis and many others with three pairs and a central
spine. In England, Ch. biniucronatus with three pairs without the
central spine. In Bohemia again, Ch. quenstedti Barrande with two
pairs of spines and two rudiments, Ch. hawlei Barrande with one pair
spines and four rudiments, and in America Ch. hydci Weller and Ch.
nuperus Billings with one pair of spines and three rudiments. Differ-
ences in the cephalon show that this is not a progressive (or regressive)
series, but apparently a number of cases of parallel development by
the loss of the posterior inner pairs of spines.

In view of this general tendency among the cheirurids to a reduction
of the spines of the pj^gidium, it seems that more weight should be
given to the cephalon then to the pygidium in determining relation-
ships, and Ch. hydei and Ch. nuperus are therefore referred to Cheiru-
rus. It may be proper, when the family has been more fully studied,
to erect a new genus for these peculiar species. The M. C. Z. has
recently acquired a fairly complete specimen (M. C. Z. 631) of this
species, of which only two other specimens are now known. The
specimen is from an unknown locality near Chicago, 111. This speci-
men shows that the eye is very far forward, opposite the first pair of
glabellar furrows. Both the genal and pygidial spines are longer than
had been supposed, and as pointed out above, the pleura of the thorax,
beyond the line of nodes denoting the fulcra, are free blade-like spines.

Measuremejits: — The specimen figured by Weller is 24 mm. long,
14.5 mm. wide at the genal angles, and the glabella is 5 mm. wide at
the back.

Formation and locality: — Known onl\' from the Niagaran near
Chicago, Illinois.

RAYMOND: NEW AND OLD SILURIAN TRILOBITES. 41

I

Cheirurus nuperus Billings.
Cheirurus nuperus Billings, Cat. Silurian fossils Anticosti, 1866, p. 60, f. 20.

This species was described from an isolated glabella and pygidium
from Div. 3 at East Point, Anticosti. Schuchert and Twenhofel have
listed it with a query from the upper part (D9) of their Gun River
formation, where it is associated with Bilobites hilohus and Triplecia
ortoni, in strata of Clinton age.

Like Ch. hydei, this species shows the Cheirurus type of basal lobes
on the glabella and the pygidium shows three pairs of spines. The
outer pair or great spines are large, flat and not so long or so much
curved as in most of the species of the genus. C. hydei has the great
spines much more slender and further apart than in C. nuperiis.

The type of this species is lost, and no further specimens have been
described.

Sphaerexochus romingeri Hall.

S-phaerexochus romingeri Hall, 20th Rept. N. Y. state cab. nat. hist., 1868,
p. 375, pi. 21, f. 4-7. (See Weller, Bull. Chicago acad. sci., 1907, no. 4,
pt. 2, p. 209, for earher and later references to this species).

This is an exceedingly common species in the Niagaran in the
Chicago and Wisconsin areas but the pygidium is rare and usually
incorrectly figured. Hall started the misrepresentation figuring the
pygidium as having three spines on each side and a rounded projection
at the back.

As a matter of fact, the margin of the pygidium is entire, and the
spines figured by Hall are the ribs on the pleural lobes. Weller
produced practically a similar figure, and one of Kindle's is about
the same, but the other, being of a mould, is more correct. Other
describers of the species have refrained from figuring the pygidium.

PLATE 1.

Raymond. — New and Old Silurian Trilobites.

PLATE 1.
All figures natural size.

1. Bumastus decipiens Raymond. A specimen lacking the free cheeks and

with the posterior portion of the pygidium incomplete. Niagaran
at Wauwatosa, Wise. M. C. Z. 641.

2. The same species. Side view of the holot>'pe, showing the large eye and

the absence of lip on the free cheek. Wauwatosa, Wise. M. C. Z.
642.

3. Bumastus niagarensis (Whitfield). An entire specimen. Wauwatosa,

Wise. M. C. Z 643.
4,5. A ctinolobus americanus R&ynioud. Two views of the holotype. From

the Racine dolomite (Niagaran), Racine, Wise. M. C. Z. 644.
6, 7. Bumastus tenuis Raymond. Two cranidia, to show the flattened form

and the wide concave lip. Wauwatosa, Wise. M. C.Z. 645, 646.
8; 9. Bumastus dayi Raymond. Dorsal and side views of the holotype.

Wauwatosa, Wise. M. C. Z. 647.

10. The same species. A large cephalon. Wauwatosa, Wise. M. C.Z. 648.

11. Bumastus tenuis ? Raymond. A pygidium referred with doubt to this

species. Wauwatosa, Wise. M. C. Z. 649.

BULL. MUS. COMP. 200L.

Raymond.— Trilobites, Plate 1

3

4

W

8

II.

E. f!. FISCHER, DEL.

PLATE 2.

Raymond. — New and Old Silurian Trilobites.

PLATE 2.

\

Bumastus indeterminatus (Walcott) The holotype. From the Leray-
Black River at Newport, Herkimer Co., N. Y. Natural size. M. C. Z. 650.

BULL. MUS. COMP. ZOOL.

Raymond.— Trilobites, Plate 2

E. N. FiSCHER, DEL.

Batmond. — New and Old Silurian Trilobites.

PLATE 3.

All figures natural size.

1,2. Trochurus nasutus (WeUer) Dorsal and side views of a cranidium from
the Niagaran at Wauwatosa, Wise. M. C. Z. 632.

3. Ceratocephala goniata Warder. Mould of the exterior of a pygidium.

From the Racine dolomite, Racine, Wise. M. C. Z. 633.

4. The same species. A fragment of a thorax. Racine, Wise. M.C. Z.

634.

5. The same species. A natural cast of the central part of a pygidium.

Racine, Wise. M. C. Z. 635.

6. Cheirurus welleri Raymond. A cephalon and parts of five thoracic

segments. Wauwatosa, Wise. M. C. Z. 630.
7, 8. Encrinurus reflexus Raymond. Dorsal view of one, and profile view of
the other of the cotypes. From the Niagaran at Wauwatosa, Wise.
M. C. Z. 636, 637.
9, 10. Calymene celebra Raymond. The cephalon of one entire specimen and
the pygidium of another, the two specimens being the cotypes of the
species. From the Niagaran at Grafton, Illinois. M. C. Z. 638, 639.
11. Calymene breviceps Raymond. An entire specimen from the Waldron
shale (Silurian), at Waldron, Indiana. Holotype. M. C. Z. 640.

BULL. MUS. COMP. ZOOL.

Raymond.— Trilobites, Plate 3

3

n>\jrx

/?>^^:::^'Cn

70

E. N. FISCHER, DEL

PLATE 4.

Batmond. — New and Old SUurian Trilobites.

PLATE 4.

1. Cheirurus dilitatus Raymond. The holotype. From the Waldron shale,

Waldron, Indiana. Natural size. M. C. Z. 628.

2. Cheirurus patens Raymond. The holotype. From the Niagaran at

Cicero, Ilhnois. Natural size. M. C. Z. 629.

3. Cheirurus dilatatus Raymond. A pygidium from the Waldron shale.

X 2. M. C. Z. 1273.

4. Cheirurus niagarensis Hall. Cephalon and part of thorax from Wauwa-

tosa, Wise. Natural size. M. C. Z. 627.

5. The same species. A pygidium from the Rochester shale at Rochester,

N. Y. X 3.4. M. C. Z. 624.

6. The same species. An hypostoma from Wauwatosa, Wise. Natural

size. M. C. Z. 1278.

7. Cheirurus welleri Raymond. A cotype. Wauwatosa, Wise. Natural

size. M. C. Z. 14.

8. The same species. The other cotype. The same specimen is shown in

Plate 3, fig. 6. Wauwatosa, Wise. X f. M. C. Z. 630.

9. Cheirurus niagarensis Hall. A pygidium from Wauwatosa, Wise. X 2.

M. C. Z. 1277.
10. Cheirurus welleri Raymond. An hypostoma. Wauwatosa, Wise. The
specimen is attached to a glabella. X 1.45. M. C. Z. 1269.

BULL. MUS. COMP. ZOOL.

Raymond — Trilobites, Plate 4

W^xM

GEORGE NELSON. PHOTO.

JAN 26 iBlu

Bulletin of the Museum of Comparative Zoology
AT HARVARD COLLEGE.
Vol. LX. No. 2.

THE AUSTRALIAN ANTS OF THE GENUS
ONYCHOMYRMEX.

By William Morton Wheeler.

With Two Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

January, 1916.

No. 2. — The Australian Ants of the Genus Onychomyrmex.

CONTRIBUTIONS FROM THE ENTOMOLOGICAL LABORATORY OF THE
BUSSEY INSTITUTION, HARVARD UNIVERSITY. NO. 104. '

/

By William Morton Wheeler.

Twenty years ago Emery described a singular ponerine ant taken by
Podenzana on Mt. Bellenden-Ker, in Queensland, as the type of a
new genus under the name of Onychomyrmex hedleyi, in honor of Mr.
C. Hedley, a distinguished Australian naturalist. The worker, which
was the only phase seen by Emery, exhibited an unusual combination
of characters, especially in the shape of the mandibles, clypeus,
petiole, and middle and hind tarsi, the terminal joints, pulvilli, and
claws of which were conspicuously enlarged. He regarded the
affinities of the genus as obscure. "Its mandibles and petiole," he
says, "recall the species of Amblyopone and related genera, but the
frontal carinae, approximated and dilated in front, resemble those of
Ponera and Leptogenys. The tibiae without spurs are not found in
any other Ponerinae. The tarsi, with their enormous claws and pul-
villi, have no analogue, to my knowledge, except in the Dorylinae
{Aenictus, Anomma), but the insertions of the antennae and the
structure of the thorax lead me to think that these resemblances do
not indicate a true relationship." Ashmead (Can. ent., 1905, 37,
p. 382) regarded the genus Onychomyrmex as constituting a distinct
tribe of Ponerinae (Onychomyrmicini). In his recent revision of the
subfamily in the "Genera Insectorum" (1911), Emery adopts Ash-
mead's name as that of a sixth and last subtribe in the tribe Ponerini.

While collecting in the rich tropical "scrub" in the neighborhood
of Kuranda, Queensland during the autumn of 1914, I succeeded in
finding not only 0. hedleyi, which had not been recorded for nearly
twenty years, but also two additional species of the genus. On
returning to Boston I learned that Dr. E. Mjoberg had anticipated
me in finding 0. hedleyi and one of the other species, while he was
collecting for the Swedish scientific expedition to Australia during
1910-1913 and that Forel had just described the latter as 0. mjobergi.
The third species is described in the following pages as 0. doddi, in

46 BT lletin: museum of comparative zoology.

honor of Mr. F. P. Dodd, the well-known observer and collector of
Queensland insects. I was so fortunate as to discover the females
of mjobergi and doddi and the larva of the former species. The female
Onychom^Tmex is of such an unusual type that it seems advisable to
revise the genus in such a way that some of my Australian friends may
be able to recognize all the known species at a glance and to make
additions to our knowledge concerning their habits.

Unfortunately the male Onychomyrmex is still unknown and will
have to be found before the precise status of the genus in the sub-
family Ponerinae can be ascertained. Forel does, indeed, describe a
male ponerine taken by Mjoberg as that of 0. hcdicyi, but he says that
he does this "with a very great interrogation point." He has, in fact,
no evidence that the specimen is an Onychomyrmex, except the very
inconclusive fact that it was taken in the same locality (Malanda,
Queensland) as the worker of hcdleyi. I deem it advisable, therefore,
to assume that the male is unknown till it is actually taken in nests
with the workers. Such observations as I was able to make on the
habits of the three species of Onychomyrmex are recorded below in
connection with the taxonomic descriptions. So far as at present
known all the species of the genus are confined to Queensland, and all
live in red rotten logs in the tropical rain-forest (" scrub ").^

Onychomyrmex Emery.

Emery, Arm. Soc. ent. Belg., 1895, 39, p. 349; Genera Insectorum,
1911, fasc. 118, p. 96; Forel, Arkiv. f. zool., 1915, 9, p. 2.

Worker. Small, slender, monomorphic. Mandibles rather long,
narrow at the base, broadest in the middle, with long, curved, acute
tips, their inner borders armed with a number of unequal teeth, some
of which, near the middle of the series, are directed backward. Both
the maxillary and labial palpi very short, 2-jointed. Clypeus very
short, abrupt, with rounded, entire anterior border beset with a regular
row of minute teeth. Frontal carinae small, prominent, closely
approximated, enlarged and dilated anteriorly, separated by a very
narrow groove. Frontal groove lacking. Eyes very small, consisting
of about 6 or 8 ommatidia, situated behind the middle of the head.

1 Emery believed that the Anommi erratica of Frederick Smith from New Guinea
might be an Onychomyrmex, but the description mentions none of the distinctive
characters of this genus and was, perhaps, drawn from an Aenictiis.

wheeler: AUSTRALIAN ANTS. 47

Antennae 12-jointed, funiculus filiform, not clavate or conspicuously
enlarged at the tip. Thorax slender, with very distinct promesonotal
and mesoepinotal sutures; mesonotum small, discoidal, with distinct
sutures on all sides. Petiole with a short peduncle in front and a
large, prominent compressed ventral projection, the node rounded,
scarcely narrowed behind where it articulates hy means of its whole
posterior surface with the postpetiole. Postpetiole large, convex
below, separated by a pronounced constriction from the gaster, which
is rather short. Sting very long and well-developed. Legs long;
middle and hind tibiae without spurs; terminal joints of the middle
and hind tarsi conspicuously elongated and incrassated, with very
large, strongly curved, simple claws and large pulvilli.

Female. Apterous and ergatomorphic. Head broadened in front
and more depressed at the anterior corners than in the worker. Eyes
very small; ocelli absent. Mandibles more falcate, not abruptly
curved at the tips, with only a few short, blunt teeth. Mesonotum
somewhat longer than in the worker. Petiole differing from that of
the worker in being much broader, with a very short and narrow
peduncle and lacking the ventral projection. Constriction between
the postpetiole and gaster much less distinct than in the worker.
Gaster much larger, elongate elliptical, sting somewhat smaller. In
other respects like the worker.

Larva. Slender, smooth and nontuberculate, with twelve very
distinct postcephalic segments, the constrictions between which are
everywhere deep and conspicuous, even at the posterior end of the
body. Head short, rounded, with well-de^'eloped, slender, acute,
falcate mandibles, destitute of teeth. Clypeus rather long, project-
ing. Antennae very small. Maxillary sensillae long and prominent.
Head sparsely, remainder of body more densely and uniformly covered
with short, straight, stiff hairs or bristles.

Genotype: Onychomyrmcv hedleyi Emery.

The discovery of the ergatoid female of Onychomyrmex only adds
to our perplexity in regard to the precise taxonomic position of the
genus. Similar females are known to occur in a few other ponerine
genera, notably in Acanthostichus, Paranomopone, which I recently
described from .Queensland, and Leptogenys (subgenus Lobopelta),
but all of these, together with Onychomyrmex, belong to very differ-
ent sections of the subfamily, and the resemblances between them
seem to be due to "convergence" and not to morphological relation-
ship, or common phylogenetic development. The thorax of the female
has simply assumed the structure of that of the worker, while the

48 bulletin: museum of comparative zoology.

gaster is greatly expanded to accommodate the voluminous ovaries.
On closer examination it is found that in each of the four genera men-
tioned the female differs from the congeneric worker in certain peculiar
characters. This will best be seen by a comparison of the worker and
female Lobopelta with the corresponding phases of OnychomjTmex.
Many years ago I called attention to the fact that the female Lobo-
"pelia elongata Buckley of Texas has no winged female, but that each
colony contains a single egg-producing individual, which agrees in
all respects with the worker, except in the larger size of the abdomen
and the somewhat more compressed petiolar scale. While at Kuranda
I succeeded in finding two females of another species (Leptogenys
(Lobopelta) fallax Mayr subsp. fortis Forel), a small-eyed form which
lives, like the species of Onychomyrmex, in red rotten logs in the
primeval rain-forest. One of these females was the mother of a
flourishing colony of perhaps 300 workers, the other was isolated in a
small cavity in a large log and was, therefore, about to start a colony.
I have figured one of the specimens (Plate 2, fig. 8, 9), with the worker
(fig. 6, 7) to show the difference between them (in this case greater
than those obtaining between the female and worker of Lobopelta
elongata) and between the corresponding phases of Onychomyrmex
mjobergi and doddi (Plate 1, fig. 3-6; Plate 2, fig. 3-5). It will be
seen that in the Lobopelta female the petiole is very much more
compressed and more curved forward than in the worker, the thorax
more convex and furnished with a small scutellar sclerite and that the
head is more orbicular and less rectangular and has distinctly larger
eyes and a single ocellus. In the female Onychomyrmex the eyes are
not larger than in the worker, there are no traces of ocelli, the head is
dilated anteriorly, with rather straight, posteriorly converging sides,
and with very different mandibles, while the petiole exhibits a peculiar
modification as compared with that of Lobopelta, being greatly
swollen behind and much contracted in front. The female Acantho-
stichus differs from the worker, according to Emery, in its rounded
head, larger eyes, the presence of three ocellar pits, more widely
separated frontal carinae, broader thorax, much larger abdomen, the
absence of prickles on the sides of the pygidium, and a different
pubescence on the abdomen. The only external differences between
the female and worker Paranomopone are the presence of a median
ocellus in the former and a larger postpetiole and gaster. These
comparisons all point to the conclusion that in each of the four genera
ergatomorphic females have been developed independently by simpli-
fication, or atrophy from the primitively winged type of female during

wheeler: AUSTRALIAN ANTS. 49

the long phylogenetic history of the ponerine subfamily. It is also
pi'obable that the very similar " diehthadiif orm " females of the ants
belonging to the subfamily Dorylinae have had a like independent
origin and development.

The larva of Onj^ehomyrmex (Plate 1, fig. 7; Plate 2, fig. 1, 2), in
the very distinct segmentation of the body and in the structure of
the head, seems to be of a rather primitive type and resembles the
larvae of the Dorylinae (Eciton) and lower Ponerinae (Acanthostichus,
Cerapachys), but the larvae of ants have not been sufficiently studied
to enable us to draw satisfactory conclusions concerning the phylo-
genetic relationships of the various genera.

A study of the worker Onychomyrmex certainly reveals a number
of highly specialized characters. Such are particularly the shape of
the mandibles, the vestigial condition of the palpi, the small size of the
eyes, and the enlargement of the terminal joint, claws, and pulvilli of
the middle and hind tarsi. The degenerate visual organs show that
these ants belong to the hypogaeic series and that they pass their lives
concealed in the logs which gradually decompose in the moist shade
of the dense tropical jungle. The powerful, toothed mandibles, long
sting and great hooked claws indicate that their possessors do not feed
habitually on small feeble insects like termites, but on much larger
creatures such as the larvae of passalids and scarabaeids and possibly
on adult myriopods and scorpions. This I found to be the case in a
colony of 0. mjobcrgi, for when the log containing it was broken open,
many of the workers were detected in the act of biting and stinging to
death a huge lamellicorn beetle larva more than two inches in length,
which they had just found in a cavity in the wood. It is not improba-
ble that the colonies move from place to place in search of their prey,
like the colonies of the subterranean Dorylinae (Eciton coecuvi and
Dorylus), which the}' very closely resemble in behavior, color, sculp-
ture, and pilosity.

The species of Onychomyrmex are far from common even in Queens-
land, and the few colonies I secured were the reward of many hours of
search and of the destruction of many old logs in places where I was
frequently attacked by land-leeches and saw quite a number of the
deadly black snakes (Pseudechis porphyriacus) . Perhaps it would be
possible for the collector to attract colonies by placing large beetle or
cossid larvae in holes in the rotten logs usually found along the paths
through the " scrub."

50 bulletin: museum of comparative zoology.

Onychomyrmex hedleyi Emery.

Plate 1, fig. 1, 2.

Emery, Ann. Soe. ent. Belg., 39, 1895, p. 350, f. 2. ^ ; Gen.
Insect., 1911, fasc. 118, p. 97, pi. 3, f. 9, 9b; Forel, Arkiv. f. zool.,
1915, 9, p. 3, pi. 1, f. 3 ^ cf (?).

Worker. Length 3.5-4 mm.

Head about I5 times as long as broad, subreetangular, a little
broader in front than behind, with straight sides and posterior border
and rounded posterior corners. Clypeus with the anterior border
slightly flattened, arcuately rounded in the middle, sinuate at the sides,
its edge beset with about 20 minute, regular teeth. Eyes with about
6-8 minute ommatidia, situated | the length of the head from the
anterior margin. Mandibles with long, abruptly incurved apical
tooth and seven basal teeth of different sizes, the two in the middle
of the series largest and directed backward. Antennae slender,
scapes fully | as long as the head, first and last funicular joints twice
as long as broad, remaining joints about I2 times as long as broad.
Thoracic sutures all strongly impressed; pronotum convex above,
especially in front, with convex sides, a little longer than broad;
mesonotum nearly twice as broad as long ; epinotum longer and nar-
rower than the pronotum, longer than broad, Avith feebly convex sides
and separated in dorsal view from the pronotum by a pronounced
impression on each side. In profile the thorax is distinctly impressed
at the mesonotum, the base of the epinotum is nearly twice as long
as the straight declivity into which it passes through an obtuse
angle. Node of petiole in profile with rather straight anterior slope
and convex summit, slightly concave at the posterior border; from
above the node is as long as broad, rounded in front, with straight
posterior border; ventral projection long and blunt, compressed and
somewhat translucent. Postpetiole as long as broad, first gastric
segment a little longer than the postpetiole. Legs slender.

Very smooth and shining; mandibles, clypeus, and cheeks sub-
opaque, the mandibles finely striated, the cl.A'peus and cheeks finely
rugulose-punctate. Bodj- with fine, sparse, piligerous punctures,
which are most numerous on the head, especially on its sides.

Hairs delicate, pale yellowish, short, suberect, covering not only
the whole body, legs, and antennal scapes but also the funiculi;

wheeler: austil\liax ants. 51

somewhat longer and sparser on the thorax, abdomen, and legs than
on the head and antennae.

Black; thoracic sutures, sides and terminal segments of abdomen,
clypeus, cheeks, and anterior portion of gula reddish castaneous,
mandibles, except their teeth, antennae, and legs paler, brownish red,
middle portions of femora and tibiae more or less infuscated.

Queensland: Mt. Bellenden-Ker, type locality (Podenzana); Ma-
landa (E. ]Mj6berg) ; Kuranda (Wheeler and F. P. Dodd).

I took two small companies of this ant, unaccompanied by larvae
or females, Oct. 24 and 28, e\ddently on foraging expeditions in the
heart of rotten logs. One of the companies comprised a dozen, the
other about two dozen workers. Later Mr. Dodd sent me eight
workers which he had taken in the same locality. The ants moved
rather slowly and were easily captured.

OXYCHOMYRMEX MJOBERGI Forel.

Plate 1, fig. 3-7; Plate 2, fig. 1, 2.

Forel, Arkiv. f. Zool., 191.5, 9, p. .3, pi. 1, f. 7; text fig. 1, S .

JVorker. Length 3. .5-4 mm.

Head subrectangular, not more than \ longer than broad, scarcely
broader in front than behind, with feebl\' and evenly convex sides,
feebly concave posterior border and rounded posterior corners.
Clypeus with broadly arcuate anterior border, sinuate on each side,
minutely and evenly denticulate. Eyes scarcely smaller than in
hedleyi, situated about § the distance from the anterior to the posterior
border of the head. ^Mandibles similar to those of hedleyi. Antennae
shorter, scapes only § as long as the head, first and last funicular joints
nearly twice as long as broad, remaining joints not longer than broad,
the more basal joints a little broader than long. Thorax differing
from that of hedleyi in being stouter and in haAing the dorsal outline
nearly straight in profile, the pronotum being convex only at the ex-
treme anterior end and the mesonotum less impressed. Thoracic
sutures very distinct but less impressed than in hedleyi. Mesonotum
fully three times as broad as long. Petiole with very short peduncle,
anterior surface of node more concave, its upper surface seen from
aboAe distinctly broader than long, with very convex sides. Con-
striction between the postpetiole and gaster somewhat deeper than
in hedleyi, legs stouter.

52 bulletin: museum of compakative zoology.

Smooth and shining; mandibles shining, not striate but sparsely
punctate, like the remainder of the body. Punctures on the head
coarser than in hedleyi, and more abundant, especially on the cheeks
and sides of the front. Clypeus subopaque, rugulose-punctate.

Hairs similar to those of hedleyi but coarser and of rather uneven
length, pale yellow.

Rich ferruginous red, clypeus darker; tarsal claws, sutures of thorax
and gaster, articulations of antennal funiculi dark brown, mandibular
teeth black; legs and anal segments of gaster paler and more yellowish.

Female. Length nearly 5.5 mm.

Head a little longer than broad and nearly as broad in front as long,
with prominent, depressed anterior corners, the sides converging
posteriorly, with two transverse impressions, one half-way between
the anterior corner and the eye and one at the eye. Eyes as small as
in the worker, but more elongate. Mandibles with less abruptly
incurved tips than in the worker and with only two indistinct teeth.
Thorax more robust than in the worker, the pro- and epinotum with
more convex sides and the pronotum more convex above, so that the
megonotum is more impressed in profile. From above the mesonotum
is scarcely twice as broad as long. Petiole much larger than in the
worker, with very short, slender peduncle, without ventral projec-
tion; node large, very convex in front, from above more than twice
as broad as long, broader than the epinotum and nearly half as broad
as the postpetiole. Gaster very much larger than in the worker,
more than twice as long as broad, suboblong, flattened dorsoventrally.

Sculpture as in the worker, but the piligerous punctures, especially
on the head, much coarser, almost foveolate and somewhat elongated
on the sides of the front. Cheeks and sides of epinotum subopaque,
finely rugulose-punctate.

Hairs coarser and longer, especially on the body, than in the worker.

Color more brownish ferruginous; mandibles, antennae, and legs
more yellowish ; pleurae, sides of petiole, and sutures of gaster brown-
ish yellow.

Queensland: Herberton (type locality), Atherton and Cedar Creek
(E. Mj5berg); Kuranda (Wheeler).

October 24, I found two fine colonies of this species in rotten logs.
One comprised at least 400 workers, a single queen, with the abdomen
greatly distended with eggs, and a large number of nearly mature
larvae but no pupae. The other colony was somewhat less populous
but also contained many larvae. The ants moved rather slowly in
long files through the cracks in the wood, evidently endeavoring to-

wheeler: AUSTRALIAN ANTS. 53

keep in close touch with one another by means of their antennae, after
the manner of the DoryHnae. They stung severely for such small
insects.

The worker of 0. mjobergi is readily distinguished from that of
hedleyi by its paler color, shorter head, antennal scapes and funicular
joints, the straight dorsal profile of the thorax, broader epinotum and
petiole, deeper constriction between the postpetiole and gaster, and
smooth, shining, and sparsely punctate mandibles.

Onychomyrmex doddi, sp. nov.
Plate 2, fig. 3-5.

Worker. Length: 2-2.5 mm.

Head subrectangular, about \ longer than broad, scarcely broader
in front than behind, with nearly straight lateral and posterior borders
and rounded posterior corners. Clypeus with broadly arcuate, finely
denticulate anterior border, sinuate on the sides. Eyes very similar
to those of the preceding species, situated about | the distance from
the anterior to the posterior border of the head. Mandibles with the
long terminal tooth less abruptly bent inward, remaining teeth rather
small. Antennal scapes | as long as the head; first and terminal
funicular joints fully twice as long as broad, remaining joints scarcely
longer than broad. Thorax rather stout, shaped much as in mjobergi,
with straight, horizontal dorsal outline, the pronotum longer than
broad, rising rather abruptly from the neck, but posteriorly flattened
above, its sides only feebly convex. Mesonotum somewhat more
than twice as broad as long. Thoracic sutures very distinct. Epino-
tum in profile with the base feebly and evenly convex and longer than
the declivity which is sloping and distinctly concave. Petiole in
profile with a short basal peduncle and prominent, compressed,
somewhat translucent ventral projection; the node with subequal
anterior and dorsal surfaces, both feebly convex; seen from above as
long as broad, subrectangular, with rounded sides and straight,
subequal anterior and posterior borders. Postpetiole as long as broad,
very convex below and separated by a pronounced constriction from
the gaster. Legs as in mjobergi.

Smooth and shining, covered with small piligerous punctures, which
are most abundant on the head and especially on the cheeks. Mandi-
bles, clypeus, and cheeks opaque, the mandibles finely and sharply

54 bulletin: museum of comparative zoology.

striate and sparsely punctate, the clypeus densely transversely rugu-
lose.

Pilosity pale yellow, much as in hedleyi but shorter.

Color also like that of hedleyi, deep castaneous, nearly black;
mandibles, except the teeth, clypeus, and frontal carinae deep brown-
ish red; antennae, legs, and tip of gaster yellowish brown; coxae and
middle portions of femora and tibiae darker.

Female. Length nearly 4 mm.

Resembling the female of mjbhergi in form, but the head is pro-
portionally broader behind and without lateral impressions; differing
from the worker in the shape of the head, which is broadened in front,
the feebly dentate, less curved mandibles and the stouter thorax and
larger petiole, postpetiole, and gaster. The sides and dorsal surface
of the pro- and epinotum are more convex than in the worker and the
promesonotal and mesoepinotal sutures are more impressed so that
the dorsal outline is much less straight and continuous. Mesonotum
not more than twice as broad as long. Petiole like that of the female
mjohcrgi, the peduncle very small, the node very large, convex and
rounded in front and on the sides, with straight posterior border;
seen from above it is only a little more than 1^ times as broad as long,
scarcely broader than the epinotum and more than half as broad as
the postpetiole. The latter is separated by a very slight constriction
from the gaster, w^hich is large and shaped much as in the female
mjohcrgi.

• Sculpture and color as in the worker, hairs considerably longer and
coarser, especially on the postpetiole and gaster.

Queensland: Kuranda (Wheeler).

I found onh- one colony of this ant (November 1), consisting of a
female and nearly 50 workers, but without larvae, in a small log in a
damp, shady spot in the dense "scrub."

The worker is readily distinguished from both hedleyi and mjobergi
by its smaller size and less abruptly curved mandibles; from mjobergi
by its color, longer head, striated mandibles and finer pilosity; from
hedleyi by the straight dorsal outline of the thorax and less convex
pronotum, shorter petiole, scapes, and funicular joints.

i

PLATE 1.

Wheeler. — Aiistralian Ants.

\

PLATE 1.

1. Onychomyrmex hedleyi Emery. Worker, lateral view.

2. Head of same; dorsal view.

3. O. mjobergi Forel. Worker, lateral view.

4. Head of same, dorsal view.

5. 0. mjobergi Forel. Female, lateral view.

6. Head of same, dorsal view.

7. Adult larva of 0. mjobergi, lateral view.

BULL. MUS. COMP. ZOOL.

Wheeler. — Omychomyrmex, Plate 1

W. M. WHEELER, Del.

PLATE 2.

Wheeler. — Australian Ants.

PLATE 2.

1. 0. mjobergi Forel. Head of larva, lateral view.
la. Mandible of same.

2. Head of same, dorsal view.

3. 0. doddi Wheeler. Worker, dorsal view.

4. Thorax and abdomen of same, lateral view.

5. O. doddi Wheeler. Female, dorsal view.

6. Leptogenys {Lohopelta) fallax Mayr subsp. fortis Forel. Worker, lateral

view.

7. Head of same, dorsal view.

8. L. (L.) fallax Mayr subsp. fortis Forel. Female, lateral view.

9. Head of same, dorsal view.

BULL. MUS. COMP. ZOOL.

Wheeler. — Onychomyrmex, Plate 2

W. M. WHEELER, Del.

FFB 12 1916

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LX. No. 3.

THE SPERMATOGENESIS OF PHRYNOTETTIX MAGNUS,

WITH SPECIAL REFERENCE TO SYNAPSIS AND THE

INDIVIDUALITY OF THE CHROMOSOMES.

By D. H. Wenrich.

With Ten Plates.

CAMBRIDGE, MA8S., U. S. A.:
PRINTED FOR THE MUSEUM.

February, 1916.

No. 3. — The Spermatogenesis of Phrynotettix magnus, with special
Reference to Synapsis and the Individuality of the Chromosomes.

By D. H. Wenrich.

contributions prom the zoological laboratory of the
museum op comparative zoology at harvard

COLLEGE. No. 266.

TABLE OF CONTENTS.

I. Introduction

A. Outline of the problems

B. Material and acknowledgments

II. Observations

A. Outline of spermatogenesis: nomenclature

a. Introductory .....

b. Outline of successive stages

1.

2.

3.
4.
5.

Spermatogonia
Primary spermatocytes
Secondary spermatocytes
Spermatids
Spermatozoa
c. Additional features

B. Synapsis
a. The postspireme stages

1. Chromosome-pair A

2. Chromosome-pair B

3. Chromosome-pair C
h. The conjugation of chromosomes .

1. The formation of leptotene threads

2. The zygotene stages

3. The pachytene stages

C. The individuahty of the chromosomes .

a. The selected chromosome pairs

1. Chromosome-pair A

2. Chromosome-pair B

3. Chromosome-pair C

b. The accessory chromosome

c. The spermatogonia! divisions

d. The somatic nuclei

D. Summary of observations

PAGE.

68
58
60
60
60
60
61
61
61
63
64
64
64
65
65
66
69
70
72
72
74
75
76
76
76
79
84
86
87
91
92

58

bulletin: museum of comparative zoology.

III. Discussion .......

A. Synapsis and the maturation divisions .

a. Results from Orthoptera

b. Recent work on synapsis

B. Individuality .....
a Constancy of metaphase chromosomes

1. Constancy in number

2. Constancy in size and shape .
b. Persistent organization of chromosomes

1. The selected chromosomes

2. The heterochromosomes .

3. Plasmosomes and nucleoli .

4. Persistence of chromosomes between mitoses

C. Chromosomes and heredity .

a. Mendelism and maturation

b. Some experimental evidence

D. Summary of conclusions

IV. Bibliography ....
V Explanation of plates

Page.

95

95

95

98

105

105

105

106

110

110

111

112

114

118

118

122

125

126

135

I. INTRODUCTION.

A. Outline of the Problems.

Two of the most important subjects which have claimed the atten-
tion of cytologists for many years are the two named in the subtitle
of this paper. Every species of animals and plants is thought to have
a definite number of chromosomes, which is characteristic for the
species. In the process of maturation this typical, or diploid, number
becomes reduced so that each functional gamete contains only half
that number, the haploid number. It is generally believed that the
process of reduction is initiated by a pairing of the chromosomes in
the prophase of the first maturation division. It is also generally
admitted that of the two chromosomes which united to form a single
pair, one has been derived from the maternal, and the other from the
paternal ancestor, and that these become separated again at one of the
two maturation divisions. But there has been a considerable amount
of disagreement as to how the pairing of the chromosomes takes place,
and also differences of opinion as to which of the two maturation divi-
sions results in their separation.

As to the process by which the pairing of chromosomes is accom-

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 59

plished, the two opposing views that have been most wudely discussed
are: — (1) that homologous chromosomes unite side-by-side (para-
synapsis), or (2) that they unite end-to-end (telosynapsis). The
adoption of either view, however, involves the very important assump-
tion that there is a continuity between the chromosomes that appear
in the earlier divisions and those that conjugate. Doubt has been
expressed by some writers as to the existence of any such continuity, or
individuality, of the chromosomes, and the question is regarded as
one that is still unsettled. Many geneticists, on the other hand, are
readily inclined to correlate the behavior of the chromosomes in
maturation with the behavior of Mendelian factors in heredity. And
in some cases an organization of the individual chromosomes has been
assumed of such a nature that a definite portion or region of a chromo-
some is concerned with the transmission of a particular factor. Such
assumptions call for an analysis of the individual chromosomes to
determine their inner constitution or architecture.

The author of the present study has sought to throw light on all
these problems. That as to how synapsis takes place was the first
considered; it was taken up from the standpoint of the origin and
constitution of the chromosomes of the first spermatocytes. Early
in the work it was found that the only method by which conclusive
results could be obtained was that of following the history of individual
chromosomes. Owing to the favorableness of the material, at least
three chromosome-pairs were found that possessed individual peculiari-
ties by which they could be recognized through all stages from the
growth-period to their division in the first spermatocyte mitosis.
Pursuit of this method naturally led to a consideration of the problem
of the individuality of the chromosomes, and it was found to be possi-
ble to recognize one pair of chromosomes at all stages from sperma-
tognia to spermatids. A further study of chromosome-individuality
led to the interesting discovery that each chromosome has a definite
organization, or architecture, which appears at the same stages in all
the animals studied.

In the following description, I have not followed the usual method
of adhering to the chronological seqvience of events, but have adopted
the order in which the problems presented themselves. I believe I
have been able through a study of this material to demonstrate that
in Phrynotettix (1) parasynapsis occurs, (2) usually the first matura-
tion is equational, (3) each chromosome preserves its individuality
throughout the spermatogenic cell-generations, and (4) at least cer-
tain chromosomes, and probably all, have a recognizably constant
organization.

60 bulletin: museum of comparative zoology.

B. Material and Acknowledgments.

Phrynotettix magnus belongs to the subfamily Oedipodinae of the
orthopteran family Acrididae. The specimens that furnished the
basis for this investigation were collected in 1907 near the Santa Rita
Mountains of southern Arizona, by a collecting party from the Uni-
versity of Kansas. The testes were dissected out and fixed in Flem-
ming's stronger solution. Sections were cut 6-12 micra thick and
stained either by Heidenhain's iron-haematoxylin, or by Flemming's
tricolor, method. Material from thirteen animals was available and
consisted partly of the slides used by Miss Pinney as the basis of her
paper of 1908, partly of other slides prepared in Dr. McClung's labora-
tory, and lastly of material sectioned and stained by the writer.

The work was begun in 1911 at the University of Kansas under
the direction of Prof. C. E. McClung, to whom I am indebted for
the material used and for advice and kindly interest throughout. The
greater part of the work was done at Harvard University during the
years 1912-1915 under the direction of Prof. E. L. Mark, to whom I
owe my warmest thanks for valuable criticism and suggestions and
for sympathetic interest at all times. I am also indebted to Miss
Eleanor Carothers, formerly a fellow student, for some collaboration,
especially with reference to the so-called "plasmosomes."

II. OBSERVATIONS.

A. Outline of spermatogenesis: Nomenclature.

a. Introductory.

There is some confusion in the literature on maturation in regard
to the use of the terms applied to the various steps and processes in
the history of germ-cells undergoing development into gametes. This
is due in part to differences in the details of the processes in the various
forms investigated, and in part to different interpretations of similar
stages by different authors. It therefore seems necessary, or at least
expedient, to explain the terms that one wishes to use in description.
A brief outline of acridian spermatogenesis follows, in connection with
which the nomenclature employed will be explained. In addition,

WENEICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 61

enough of the pecuUarities of Phrynotettix will be described to render
clear any new terms made necessary by them.

Wilcox ('94) and Davis ('08) have both given detailed descriptions
of the structure of the acridian testis and have given figures or dia-
grams to show the topography of the follicles of which the testes are
composed. It will therefore be unnecessary to reproduce such figures
and descriptions here.

b. Outline of successive Stages.

1. Spermatogonia. — The spermatogonia of Phrynotettix (Plates
1, 2, fig. 1-20) behave in a manner typical for the Acrididae as de-
scribed especially by Sutton ('00), and by Davis ('08). As Pinney
('08) has shown, there are 23 chromosomes, of which 22 can be
arranged in pairs, leaving an odd one, the accessory chromosome
(McClung, '99), or monosome (Montgomery, '06). The paired chro-
mosomes may be referred to as the autosomes (Montgomery, '06).
All the divisions of the spermatogonia are mitotic and are usually
considered as equivalent to somatic mitoses. A detailed account of
these divisions is given on pages 87-90.

2. Primary spermatocytes. — The daughter cells produced at the
final spermatogonial division, as is well known, are characterized,
among other things, by the growth-period and by the formation of
the reduced, or haploid, number of chromosomes. For distinguishing
the different stages in the prophases, Davis ('08) employed a non-
descriptive method, designating successive stages by the successive
letters of the alphabet. Here it seems advisable to use largely the
terminology introduced by Winniwarter and by Gregoire.

The telophase of the last spermatogonial division embraces a series
of processes similar to those in the telophases of the earlier spermato-
gonial divisions. Following the telophase, a series of changes takes
place which results eventually in the formation of fine single threads.
This fine-thread stage may be called the leptotene stage (Winniwarter,
'00). Between the telophase and the leptotene stages occur changes
which are of the utmost importance in any attempt to solve the prob-
lems of synapsis and the individuality of the chromosomes. These
stages may be called the jjreleptotene stages (Gregoire, '07). There
may be distinguished an earlier (Plate 2, fig. 23, 24) and a later (Plate
3, fig. 25-27) preleptotbue stage.

When the leptotene threads are first formed, they seem to be greatly

{)2 bulletin: museum of comparative zoology.

tangled and lack definite arrangement. This condition, which may
be called early leptotene (fig. 28), is followed by a later leptotene
(fig. 29), in w^hich the threads become oriented with one end attached
at one side (the polar side) of the nucleus. Soon there appear among
the single threads others which are double and twice the width of the
single ones. The proportionate number of double threads gradually
increases until all the threads appear double. The stage during w^hich
the doubling takes place (fig. 30, 31) is the zygotene stage of Gregoire
('07). When all the threads have become double the pachytene stage
(Winniwarter, '00) has been reached (Plate 3, fig. 32-34). This term
continues to be applicable throughout the relatively long growth-
period, and until the spireme ^ breaks up into the haploid number of
segments, which become tetrads. The number of pachytene threads
seems to be much less than that of the leptotene threads.

The stages characterized by the appearance of separate segments of
the spireme may be designated by the term diplotene of Winniwarter
('00). This term is used for the sake of consistency with the others
employed, although the conditions in Phrynotettix differ somewhat
from those described by Winniwarter for mammals. He describes
the longitudinal split as disappearing in the pachytene stages, on
account of the threads becoming twisted, and reappearing in the diplo-
tene stage. In Phrynotettix the longitudinal split remains visible
and little or no twisting occurs.

Soon after becoming independent, a second longitudinal split occurs
in the spireme segments at right angles to the first, thus forming
typical tetrads, each composed of four chromatids (McCIung, '00).
The first longitudinal split, which persists from the pachytene stage,
may be called the priviary split, and the one at right angles to it may
be called the secondary split. From the time of their formation until
the succeeding metaphase, the tetrads undergo a gradual shortening
and thickening. During this period they pass through the well-
known figures, X's, K's, 8's, rings and crosses (Plate 3, fig. 38). The
stage during which these changes occur is frequently referred to as
the diakinesis stage (Hacker, '95*), but it may be simpler to call it
the postspircine stage (Gregoire, '07), or the tetrad stage."^

The postspireme stages end with the establishment of the tetrad-

' The term spireme will be used to embrace the stages included under the names leptotene,
zygotene, and pachytene without, however, implying anything as to the existence of a con-
tinuous thread.

2 I have avoided the use of the term, prophase, in connection with the postspireme stages
because it might properly be applied to the whole series of stages from the preleptotene to the
metaphase.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 63

chromosomes upon the mitotic spindle of the first spermatocyte divi-
sion. The number thus appearing on the spindle is twelve (Plate 4,
fig. 39). One of them is the accessory chromosome, which is a dyad
and passes to one pole undivided (Plate 4, fig. 41, X). The eleven
tetrads represent the other twenty-two spermatogonial chromosomes
arranged in pairs. One daughter cell of each spermatocyte receives
eleven dyads and the other receives twelve, the additional one in the
latter case being the accessory chromosome.

In the anaphase all the chromosomes appear as V's, thus showing
their dyad constitution (fig. 42 and 43). Before this, in the meta-
phase, the separate chromatids are not discei-nible, but early in the
anaphase they separate from each other at the end opposite that
which is attached to the spindle-fiber, in this way giving rise to the
V-shaped figures. The V-shaped arrangement persists until the
metaphase of the succeeding division is reached.

3. Secondary spermatocytes. — The secondary spermatocytes pre-
sent only a short resting stage. For this stage, between the formation
of the secondary spermatocytes and their division, we may employ
the term interkinesis (intercinese) proposed by Gregoire ('05). The
extent of diffusion reached by the dyads in interkinesis is much greater
than that usually described for this stage, as will be seen from figures
46 and 47 (Plate 4). The djads reappear however, in the same
orientation and relative positions that they had before dift'usion.

In the metaphase the dyads show the same double structure that
they did in the anaphase of the immediately preceding division (Plate
5, fig. 50-52). The two monads composing each dyad are separated
from each other in the metaphase, and in the anaphase are carried to
the poles of the spindle (fig. 54). Half of the secondary spermato-
cytes show in the plates of the metaphase eleven chromosomes and
the other half twelve chromosomes, as was to have been expected
owing to the non-division of the monosome in the division of the
primary spermatocytes. Figure 50 (Plate 5) shows eleven and
figure 51 shows twelve chromosomes.

The term reductio^ial will be used to designate that one of the two
maturation divisions which results in the separation of the chromo-
somes that conjugated in synapsis. Correspondingly the term equa-
tional will be applied to the division in which the halves of whole
chromosomes are separated. Employing the terminology of Korschelt
und Heider ('03), we may use the terms prereduction when the first
maturation division is reductional, and postreduction when the second
division is reductional.

64 bulletin: museum of comparative zoology.

4. Spermatids. — The spermatids, daughter cells of the secondary
spermatocytes, undergo gradual transformation, without further divi-
sion, into the mature spermatozoa. Their chromosomes undergo dis-
solution, having, however, first formed a network not unlike that found
in the telophase of ordinary mitoses.

5. Spermatozoa. — This term is used, as usual, to designate the
functional male gametes — the end products of all the preceding
processes.

c. Additional Features.

In Phrynotettix, as pointed out by Pinney ('08), there appear in
many of the stages of spermatogenesis condensed and deeply staining
granules at the ends of the chromosomes. These granules are recog-
nizable in the stages where the greater part of the chromatin is ex-
tended or diffuse, so that their density, contrasted with that of the
rest of the chromatin, brings them into view. Figures 8, 10, 12
(Plate 1) and 14-20 (Plate 2) show them for the spermatogonia;
figures 28-38 (Plate 3) for the primary spermatocytes; figures 45-48
(Plate 4) for the inter kinesis stage, and figure 55 (Plate 5), for the
spermatids. These granules appear at that end of each chromo-
some — including the accessory — to which the spindle-fiber attaches.
They were named accordingly by Miss Pinney polar granules. In the
case of certain clu'omosomes, as noted by her, similar granules also
occur at the end of the chromosome opposite that to which the spindle-
fiber attaches. The chromosomes are thus seen to exhibit polarity
and it will therefore be convenient to designate the two ends by differ-
ent terms. In the. absence of better terms, I shall call the end to
which the spindle-fiber attaches the proximal or synaptic end, and the
opposite one the distal end.

At various stages there is a tendency for some of the polar granules
to fuse together, as noted by Pinney '08, forming what I shall call
coviposite granules. These are to be seen in the telophase and pro-
phase of the spermatogonia (Plate 1, fig. 12; and Plate 2, fig. 14, 16), in
the spireme stages of the primary spermatocytes (Plate 3, fig. 33-36),
and even in the connective-tissue nuclei (Plate 9, fig. 108-110).
They are particularly noticeable in the pachytene stages, for during
that period quite large masses of chromatin may form by the coales-
cence of a nimiber of these polar granules. The number of granules
making up a composite granule is variable, but may usually be de-
termined by the number of spireme threads attached to it. These

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 65

sometimes radiate out from the composite granule like the spokes of a
wheel from its hub. Such a stage corresponds to the bouquet stage
of Eisen ('00). At the end of the pachytene stage the granules
composing the composites separate out again, apparently without
having changed their identity (Plate 3, fig. 35-37).

The tendency of the polar granules to remain on one side of the
nucleus may be interpreted as evidence of a somewhat persistent
polarity of the nucleus as a whole. It will therefore be convenient to
speak of that region of the nucleus where the majority of the polar
granules are congregated as the ■proximal pole, and the opposite side
as the distal pole, of the nucleus.

In my description of the leptotene and zygotene stages it will have
been noticed that no mention is made of the contraction, or synezesis,
stage (McClung '05). Such a phenomenon has not appeared in my
material and, as has been claimed by McClung ('00, '05), Davis ('08)
and others, is probably not normal in the Orthoptera.

I shall use the term synapsis in the same sense in which it was
originally used by Moore (*95), that is, to indicate the process of
coupling or conjugation of the chromosomes of the last spermatogonia
to form those of the jSrst spermatocyte. Following Wilson ('09), I
shall use parasynapsis to denote side-by-side conjugation, and telo-
synapsis to denote end-to-end conjugation.

For the purpose of determining more accurately the history of the
changes undergone by the chromatin through the successive stages
outlined above, three individual autosome-pairs have been selected
for detailed study. To distinguish them from the other autosomes,
I shall call them the selected chromosomes.

B. Synapsis.

a. The Postspireme Stages.

Of the various methods by which the diploid series of chromosomes
could unite in pairs to form the haploid, or reduced, series, the two
which have been more frequently defended are: — (1) that by which
the members of each pair unite end-to-end (telosynapsis), and (2)
that by which they unite side-by-side (parasynapsis). Evidence in
favor of both methods has been gained from observations on orthop-
teran material. The writer, without prejudice in favor of either view,
undertook to discover which of these processes occurs in Phrynotettix.

66 bulletin: museum of comparative zoology.

Efforts were first directed to a study of the postspireme stages in
the hope of discovering how the segments of the pachytene spireme
became the tetrads exhibiting the shapes of V's, X's, 8's, crosses and
rings. Such a variety of shapes and forms presented themselves at
any one of the tetrad stages, however, that it was impossible to decide
which were the more primiti^'e and which the derived forms. Figure
38, a-g (Plate 3); for example, shows some of the different shapes of
tetrads seen in a single stage and, indeed, in the same cyst. The only
method that seemed to offer a means of securing decisive evidence on
the problem was that of following the history of individual chromo-
some-pairs through a large number of stages. For this purpose it was
necessary to find pairs which possessed individual characteristics by
which they could be recognized in all the stages concerned. Fortu-
nately, at least three pairs were found which fulfilled these require-
ments. For convenience in description they have been designated
'M," "5," and "C."

1. Chromosome-pair A. — This element was first distinguished in
the pachytene stage, where it is a very deeply stained spireme segment.
Examples of it are shown in figures 56 and 57, (Plate 5). Its differen-
tial staining property is so marked and constant that it can be recog-
nized by this character alone up to the later postspireme stages. But
there is an additional means of identification. Like most of the pachy-
tene threads, this one normally makes a loop the two ends of which
approach to, or attach at, the proximal pole of the nucleus (Plate 5,
fig. 56). One or both ends may become free from entanglements, but
more frequently only one. In the latter case the free end, or if both
ends are free, one of them, is nearly always terminated by two knobs,
of which one is usually larger and less deeply stained than the other
(Plate 5, fig. 57.(7; P'ate 10, fig. 113). These knobs, I believe, may be
identified as the polar granules described by Miss Pinney ('08). But
in this instance, as shown by numerous observations, the more promi-
nent granules occur at the distal end of the chromosome instead of
the proximal end, where they are found on the majority of the other
chromosomes. That the expanded condition of one of the granules
furnishes a means of identification, will be apparent from an examina-
tion of figure 62 (Plate 6).

As an exceptional occurrence these two terminal granules may be
equal in size, neither one being expanded. In order to test the relative
frequency of these two conditions, some counts were made and
tabulated for both the spireme and postspireme stages, as follows: —

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 67

Stage Total Expanded Not expanded % Expanded

Spireme HI 94 7 93.06

Postspireme 162 146 16 90.12

Both stages 273 240 23 91.26

"f^

It will be seen from this table that approximately 90% (examples
counted at random) have one of the granules in the expanded condi-
tion. In the postspireme stages this peculiarity appears less like an
expanded single granule than as a group of closely associated small
granules, typically three in number. This condition will be discussed
more fully in another place (p. 112). In both the spireme and the
postspireme stages the modified polar granule furnishes a ready means
of identification of chromosome-pair A, especially when its staining
qualities, already described, are taken into consideration. The con-
stant relative size of A in the tetrad stages is also a help in identifica-
tion.

Figure 62 (Plate 6) indicates clearly the processes by which the
spireme loop becomes first transformed into a typical tetrad, and then
condensed to a metaphase chromosome. From the zygotene stage
onward, there is a gradual shortening of the spireme loops or segments.
The later stages of this process are to be seen in figure 62. Through-
out the pachytene stage the spireme loops exhibit a median longi-
tudinal cleft, usually referred to as the longitudinal split. I shall call
this the primary longitudinal split. Occasionally paired granules, or
chromomeres, appear to be fused together, but as a general rule, the
split is continuous throughout the length of the loop. In my opinion
this so-called longitudinal split is really the space between two spireme
(leptotene) threads which have conjugated side-by-side. Fm-ther evi-
dence for this belief will be presented later.

Figure 62, c, indicates the first step in the process of forming the
four chromatids of the tetrad. A second longitudinal split, at right
angles to the first or primary split, begins at the proximal end (upper
end in the figures) of the free spireme segment (fig. 62, c) and gradually
proceeds toward the distal (lower) end (fig. 62, c-c). It will be seen
from these figures that as the separation produced by the secondary
split proceeds distally, the separated chromatids at the same time
reunite along the plane of the primary split. The separation due to
the secondary split gradually increases until the diverging pairs of
chromatids extend in opposite directions, thus forming a rod-like
element the two ends of which correspond to the proximal pole of the

68 bulletin: museum of comparative zoology.

original spireme segment, and its middle point to the distal pole. The
rod-shaped tetrad becomes oriented in the spindle of the first matura-
tion division with its long axis parallel to the spindle-axis, and at
metaphase separates in the middle. In other words, the plane of the
secondary split becomes the plane of the first maturation division,
which is therefore equational. If now we may assume that the longi-
tudinally split spireme segment has represented a pair of chromatin-
threads which had conjugated side-by-side throughout their length,
the plane of the primary split must be the plane of the reductional
division, which becomes effective in the second spermatocyte mitosis.

The tetrad A also forms rings, as shown in figure 62, j, k, I (Plate 6).
I have not been able to trace these rings into the metaphase to de-
termine their orientation on the spindle, and furthermore I am quite
uncertain whether the ring shape persists as far as the metaphase.
Most of the metaphase figures show one tetrad in the form of a rod
with its axis parallel to the spindle-axis, and with a constriction in the
middle, as shown in figure 62, i and figure 79, A (Plate 7). Sometimes
two or more rod-shaped tetrads are to be seen in the same spindle
and with the same orientation. However, one of them is alwavs in a
more advanced stage of division than the others, and I have been in-
clined to identify this precocious one with tetrad A. Figure 62, c-t,
indicates that such a conclusion is justified. Since the straight-rod
condition is so characteristic of the metaphase, it may be that the
rings also become transformed into straight rods by the time the
metaphase is reached.

The rings seem to have been formed either by a failure of the proxi-
mal ends to separate during the formation of the secondary longi-
tudinal split, or by a secondary union of these ends, i. e. after the split
had begun. For example, if a tetrad in the condition of figure 62, c,
has the secondary split completed without the separation of the proxi-
mal ends, a ring would result. So also would a ring be formed by a
secondary union of the two proximal ends of a stage such as is seen in
figure 62, d or e. In either event the region within the ring would
represent the space formed as a result of the secondary longitudinal
split. If the chromatids should now begin to separate at the proxi-
mal end along the plane of the primary split, as seems to be indicated
in figure 62, k and /, and if this process should be continued until a
metaphase chromosome such as that shown in i is produced, there is
every reason to believe that it would result in a separation of the
original conjugants of the pair, and therefore constitute a reductional
division. On the other hand, it is possible that the separation along

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 6^

the plane of the primary split is never completed, but that the chroma-
tids again become separated at the proximal end, assuming the forms
shown at /to i, figure 62, and that the first division is therefore always
equational. However, the possibility of an occasional reductional
division as a result of the ring-formation must be taken into considera-
tion.

2. Chromosome-pair B. Figure 63 (Plate 6) presents a series of
stages for B corresponding to those in figure 62 for A. This series of
stages of B supports the conclusions reached from a study of A in
regard to: — (1) a probable parallel, association in the pachytene
stages of pairs of threads, each representing individual chromosomes;
(2) the formation of the tetrad by, first, a separation along the plane
of conjugation (?". e., the primary longitudinal split) and, secondly,
by a splitting of each of the original conjugants (the secondary longi-
tudinal split) ; and (3), as a result, an equational division of the tetrada
at the first division.

This chromosome-pair (B) is characterized by the presence of large
and well-marked polar granules at both ends and by a similar large
granule not far from the middle, though always somewhat nearer the
distal end. Leaving aside the formation of rings, the chief difference
in behavior between A and B is that jn the former the plane along
which the greatest separation takes place before metaphase is that of
the secondary longitudinal split, while in the latter the greatest separa-
tion takes place along the plane of the primary split. This results
in A becoming extended in the direction of the spindle-axis, as already
described, while B becomes extended at right angles to this axis. In
the latter case the separation along the plane of the primary split does
not become complete at the expense of the separation along the plane
of the secondary split, but the latter separation persists for a short
distance, giving rise to a cross with unequal arms (fig. 63, g, h). The
short arms terminate in the proximal or synaptic ends of the chroma-
tids, while the longer arms terminate in the distal ends.

However, these differences in behavior between A and B are not
fundamental, since the final result, an equational division, is the same
in both cases. But they are indications of the individual peculiarities
of these elements. It should also be pointed out that such differences
could easily be misinterpreted, if only parts of the histories of the
pairs were known.

It is important to note that the drawings of the series shown in
figure 63 were all taken from sections of a single testis. In searching-
for the same element in other individuals, I was surprised to find the

70 bulletin: museum of comparative zoology.

condition shown in figure 64, a-h. In this series are found the same
differentiating characters that have ah-eady been described for B,
except that one member of the large pair of granules at the distal end
is lacking. In other words, we have to do here with a pair, composed
of unequal elements, which differs from its homologue in another
individual, composed of equal elements, by the absence of a definite
part of one of the components. Examination of all the thirteen
individuals demonstrated that eleven of them possessed this second
or unequal type, while only two showed the equal type.

If there could have been any doubt about the sequence of events
in the transformation of a spireme segment into a tetrad and the sub-'
sequent equational division in the case of chromosome-pair A or the
equal type of B, the beha^•ior of this unequal type of B, as shown in
figure 64, must certainly make the subject clear. In this instance,
on account of the difference between the two members, it is possible
to identify them in such a way that there can be no question as to the
two planes of longitudinal splitting. The figures have in all cases
been made with great care with the aid of a camera lucida and are
faithful reproductions of the conditions seen under the microscope so
far as they can be represented by the method of reproduction used.

In the early stages of the transformation of the spireme segments
into tetrads, the separate chromatids are not distinguishable through-
out the whole length of the segment. This is due in part to a closer
association of the chromati<^ls anfl in part to the fact that one of the
longitudinal splits becomes more pronounced at one end and the other
split at the other end of the tetrad. Somewhere between the ends,
therefore, there is a crossing or apparent chiasma. At the point
of the crossing the chromatids at first appear to be fused together
(figs. 63, d and 64, d). Very soon, however, the confusion disappears,
the chromatids become distinct, and their relationships easily discerni-
ble, as shown in figures 68, r, and 64, r. In botli these cases the wide
separation at the proximal end has been along the plane of the second-
ary longitudinal split, and that at the distal end along the plane of
the primary split. The resulting crossing, or apparent chiasma, is
a perfectly normal and natural result of these processes and indicates
nothing in the wa^^ of a breaking or recombining of the parts of
chromatids.

3. Chromosome-pair C. — Figure 65 (Plate 6) shows one form of
the third of the three selected chromosome-pairs. In this case the
two components are very unequal in size, one of them possessing a
very large, condensed mass, or granule, of chromatin at its distal end,

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 71

while the other has none. It will be noticed (fig. 65, b, c) that the
details of the two components are quite homologous up to the large
distal granule, and that the point of attachment of this large granule
seems to correspond to the distal end of the smaller component. These
considerations would lead us to suppose that here, too, as in the un-
equal type of B, the difference between the members of the pair may be
due to the loss by one of them of a definite part possessed by the other.
In this case, however, no such equal pair has been found as occurs in
B when both members possess the part in question. The side-by-side
association of the members of this pair is as evident as it was for A
and B and the relations of the two longitudinal splits are the same.

In regard to the mode of distribution of tetrad parts in the first
maturation division, however, we meet in this case a curious exception
to the general rule. This pair divides equationally, as shown in
figure 65, h-j; but it sometimes divides reductionally as shown at
k-m, same figure. From casual inspection it appeared that the divi-
sion occurred as frequently in the one manner as in the other. But
in order to test the relative frequencies of the two methods, 928 cases
chosen at random were counted and it was found that of this number
472, or 50.8%, were in process of reductional division, while 456, or
49.2%, were dividing equationally. It would seem from these counts
that the method by which the tetrad divides is a matter of mere
chance. This is the more apparent when we take into consideration
the fact, brought out by extended observations, that the two methods
occurred side-by-side in the same cysts. It may be that the shape or
position of the tetrad when it is first brought under the influence of
the mitotic spindle determines the mode of division.

The fact that this unequal pair divided in the first division reduc-
tionally a part of the time made it possible to study the distribution
of the two conjugants with reference to the accessory chromosome,
which goes to one pole undivided. It was soon found that either
member of the pair could accompany the accessory into the second-
ary spermatocytes. Consequently counts were made to determine
whether the two kinds of distribution occurred with anything like
equal frequency. Out of 421 cases counted at random 216, or 51.3%,
were found to show the larger member going to the same pole as the
accessory (Plate 10, fig. 121, C), while in 205, or 48.7%, of the cases
the srnaller member was going with the accessory to the same pole (fig.
120, C) . These results seem to furnish a good example of chance dis-
tribution of clu-omosomes at maturation.

The behavior of these three selected clu-omosome-pairs, as described

72 bulletin: museum of comparative zoology.

above in detail, seems to me to establish very definitely that the
association of paired eliromosomes in the pachytene stages is one in
which the members lie side-by-side throughout their entire length,
and therefore exliibit parasynapsis. I should further add that while
I have not singled out any other members of the complex for individual
study, a careful analysis of the other spireme segments and the deriva-
tive tetrads indicates that the condition of parasynapsis is realized
for the entire series. I was thus able to analyze the stages of the
complex as a whole after following the history of the selected indi-
vidual pairs, whereas previously I was unable to reach a definite
conclusion.

As to the method of division in the first spermatocytes, the evidence
presented indicates that B, always, and A, in most cases, divide equa-
tionally, while C divides either reductionally or equationally and with
equal frequency by each method. My study of the other tetrads
leads me to think that, as a general rule, they divide equationally in
the first division. Where the first division is equational the second
is regarded as reductional, and we therefore have postreduction.
The general rule has its exceptions, however, as already noted in the
case of C and possibly sometimes in the case of A.

b. The Conjugation of Chromosomes.

1. The formation of leptotcne threads. — The evidence for para-
synapsis derived from a study of the postspireme stages, as presented
in the preceding paragraphs, has not embraced the actual process of
conjugation; and it therefore remains to be demonstrated that a
side-by-side conjugation does take place. But it is even more im-
portant to show that the conjugants are actually chromosomes, the
morphological descendants of the telophase chromosomes of the final
spermatogonial division. Figure 21 (Plate 2) shows a side view and
figure 22 a transverse (optical) section through the chromosomes of
cells nearing the end of the telophase of the last spermatogonial divi-
sion. The side view shows the chromosomes already partly diffused,
but each one occupies a definite territory, so that there is no question
as to their persistent individuality, except for the coalescence of some
of the polar granules. But, as I shall point out later, the polar gran-
ules do not necessarily lose their identity when they unite into the
compound masses. The optical section, figure 22, shows even more
plainly the persistent individuality of the chromosomes up to this

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 73

point, for there still can be seen the remnants of the vesicular walls
which surrounded each chromosome in the earlier telophase.

There are 21 of these chromatin-masses, or "blocs" (Janssens, '01),
shown in this optical section, and that is sufficiently close to the total
number, 23, to indicate that all the chromosomes are still independent,
except for the union at the polar ends, as already mentioned.

In figure 21 it will be noticed that the diffusing chromatin is dis-
posed roughly in the form of spirals. Figures 23 to 29 indicate what
becomes of these spirals in the " blocs" of chromatin. I am not quite
sure of the exact succession of stages here, but believe they are about
as shown in the successive figures. It is possible that figures 23 and
24 — which are side view and optical section, respectively, of the same
stage — are no earlier than the stages shown in figure 25 (Plate 3).
However that may be, the evidence seems to indicate that each of the
blocs at stages such as those shown in figures 21 and 22 gives rise to a
single fine thread, at first much coiled but later much elongated.

The side view shown in figure 23 is at a stage the casual examination
of which might lead one to suppose that the chromatin was in a hope-
less tangle without any definite arrangement whatever. But careful
focussing and patient study revealed what I have tried to show in
figure 23, viz., that the chromatin is still disposed, for the most part,
in separate blocs, but that a very much coiled and convoluted thread
is forming within each one of these territories. Some have unraveled
to a considerable extent, and have become extended in various direc-
tions through the nuclear sap. But each seems to be a continuous
thread, despite some tendency for the ragged edges at times to be
connected with adjacent threads. In the optical section of this stage
(fig. 24) it will be seen that the blocs have remained in place and
separate from each other for the most part, though some anastomosis
of the linin fibers has taken place at the periphery of the blocs. On the
other hand, there are still some remnants of the previously existing
vesicular walls, as shown in the left side of the figure. When one
focusses up and down on such a cell, it is possible to follow in some cases
the tliread which is differentiating out of the net-like structure of each
bloc, but in optical section the reticulum is more apparent than the
continuous thread. The section shows nineteen blocs, which number
is not far from the somatic number of chromosomes (23). Figures
23 and 24 represent what I have called the preleptotene stage.

At the stage shown in figure 25 (Plate 3), which I believe to be
slightly more advanced than the one in figures 23 and 24, the amount
of anastomosis between adjacent chromatic elements seems consid-

74 bulletin: museum of comparative zoology.

erably greater than in the stage last described. The anastomosis is
to be seen more particularly at the sides of figure 25. Through the
middle of this figure the individual spiral threads seem to be more
easily distinguishable, and I am inclined to believe that the two which
stain more deeply than the others are the members of the A pair of
chromosomes. The stages including and following this reticular
stage are hard to represent in a drawing of the kind employed, owing
to the difficulty of portraying in their natural relations the parts seen
at different planes of focus. Careful study has always convinced me,
however, that the uncoiling and elongating threads are single, con-
tinuous, and not united into an indiscriminate network. I have
selected in figm-es 26 and 27 views favorable for di'awing where some
of the threads, at least, are definitely separate and continuous across
the diameter of the nucleus.

At the stage represented in figure 28 (Plate 3) the unwinding of the
coiled threads has been completed, but the threads have as yet no
definite orientation. At the somewhat later leptotene stage shown in
figure 29 the threads are finer and less homogenous than in the earlier
stage, the substance of the thread seeming to have become more
distinctly differentiated into a linin fiber and cliromatic granules, the
latter scattered at irregular intervals along the fiber. Moreover, in
this later stage the threads appear to be definitely oriented, with one
end attached at the proximal pole of the nucleus. The threads then
take a course thi'ough the center of the nucleus or near its periphery,
extending wholly or partly across and then turning back with a wide
curve.

2. The zygotene stages. — In figure 30 (Plate 3) some of the threads
are double, others are single, and it would be difficult to decide from
a casual examination of this stage alone whether or not the double
threads had arisen by a splitting of the single ones. In the case of
one or two of the double threads, however, as may be seen at the left
side of the nucleus, the double condition does not continue throughout
the whole length, but towards the distal end of the nucleus the thread
is seen to branch into two single threads. I interpret tliis branching
thread as one in which the parallel conjugation has not yet been
completed. Another instance of the same kind may be seen in figure
31, which represents a stage somewhat more advanced than that of
figure 30. These appearances lead me to believe that conjugation
begins at the side of the nucleus corresponding to the proximal ends
of the leptotene threads, and proceeds gradually toward their distal
ends. It is further evident from these figures that conjugation is

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 75

not a simultaneous process for all the chromosome-pairs, but that
it is a gradual process, some conjugating earlier than others. Just
how the members of the different pairs are enabled to select their
mates is a very puzzling question, but probably the stretching and
orientation of the threads as shown in figure 29 might facilitate this
process. That some of the pairs conjugate quite early, is shown in
figure 30, where it may be seen that in selected pair B conjugation is
complete. In figure 27, which is of a very much earlier stage, there
are to be seen two of the still hazily defined threads lying side-by-side.
They are similar enough in their constitution to be regarded as the
two members of a pair, and it would not be surprising if conjugation
should begin at a stage as early as this.

As an additional detail it should be pointed out that the bead-like
granules which are strung along the threads of the leptotene and
zygotene stages are not always of exactly equal size in the two con-
jugating elements. Tn figure 58 (Plate 5) the example of chromo-
some-pair B well illustrates the disparity in size between the two
members of some of the pairs of granules. This condition may well
answer the criticisms of those who hold that the accuracy with which
the granules are paired could be accounted for only on the assump-
tion that they arose by a splitting of single granules into equal parts.
I am able to show in this case that the members of each pair of granules
are not always of equal size.

3. The pachytene stages. — ■ Figures 32 and 33 (Plate 3) are of early
pachytene stages. It sometimes happens that even at such stages
there may remain one or two pairs of threads that are not fully conju-
gated, though I have not added a drawing of such a condition. In the
case of some of the pachytene threads of figure 32, complete loops have
been formed, both ends being attached at the polar region. The
formation of such loops is not necessarily the rule, however, as has
been indicated already in connection with the spireme loops of the
selected chromosome-pairs (fig. 56-61, Plate 5). In figure 33 a scat-
tered arrangement of the polar granules is to be seen, though they have
coalesced to form several composite granules. Figure 34, of a later
pachytene, exhibits one of the large composite granules. Figures
35-37 indicate how the composite granules break up into their com-
ponent polar granules. A comparison of the examples of chromo-
some-pair B in figures 30 and 35 will indicate the extent of the process
of gradual shortening which takes place during the pachytene stages.
It will be noticed that the line of separation between the threads
which have conjugated (i. e., the primary longitudinal split) remains
visible throughout the pachytene stages.

76 bulletin: museum of comparative zoology.

C. The Individuality of the Chromosomes.
a. The selected Chromosome Pairs.

The method adopted in the study of the subject of synapsis — that
of following the history of individual chromosome-pairs — has natu-
rally led to a consideration of the subject of the individuality of the
chromosomes, that is, their persistence as morphological entities
tlirough all the stages of nuclear activity. I have already attempted
to demonstrate that each of the chromosomes of the last sperma-
togonial division gives rise to a single leptotene thread and that these
single threads conjugate two-by-two in the zygotene stage. It will
be more convincing, however, if we can follow some particular chromo-
some-pair through these difficult stages.

1. Chromosome-pair A. — As the chief characteristic by which the
chromosome-pair A could be recognized in the pachytene and later
stages of the first spermatocytes, I have already described its great
density and staining capacity. If there is a persistence of individual
chromosomes from the spermatogonia to the spermatocytes, we should
expect to find in the former a pair of chromosomes exhibiting th( same
peculiarities that the pair did in the later generation. Such a pair
can, indeed, be found in the telophases not only of the last spermato-
gonia! division but of the earlier spermatogonia as well. Figures 66
and 67 (Plate 6) show such pairs of chromosomes more deeply stained
than their fellows. Figure 66 shows one of the earlier generations of
spermatogonia, as is indicated by the vesicular condition of the ac-
cessory chromosome, and figure 67 represents a telophase of the last
spermatogonial division, as is shown by the condensed accessory at
this stage.

It is difficult to follow all the changes that these chromosomes
undergo in their transformation into pachytene tlu'eads, but I believe
that most of the stages are represented in the series of drawings,
figures 67-78. Figure 67 corresponds to a stage midway between
those shown in figures 21 and 23 (Plate 2). Figure 68 is of a stage
corresponding very closely to that in figure 25 (Plate 3). In figures
68 and 25, two bands or "blocs" of chromatin can be seen which are
more deeply stained than the other chromatin-blocs. The accessory
chromosome is distinguishable by its characteristic density and its
position at the periphery of the nucleus. The polar granules are also
distinguishable. The chromatin in these darker blocs {A in both

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 77

figures 67 and 68) shows a more or less well-defined spiral condition.
This spiral is better shown in figure 69, where it is more unravelled.
Very soon after the process of uncoiling gives rise to the leptotene
threads, stages in conjugation may be seen. Figure 70 shows an early
leptotene stage with two threads which stain more deeply than the
others, having conjugated as far as they can be traced in this particular
section. I think we may identify these denser threads as the members
of the chromosome-pair A.

The two sides of figure 70 are drawn differently. The left side is
diagrammatic and is intended to represent the apparent entanglement
of the leptotene threads. On the right side an attempt has been made
to follow individual threads. Careful study makes it evident that
the tlireads, instead of anastomosing, as they appear to do when one
makes only a superficial examination of them, are really continuous
and distinct for certain distances. The difficulty in following in-
dividual threads is due to the fact that after the early leptotene stage
the chromatin collects into chromomeres, which are strung along a
linin fiber, so fine and stainless in some places that it is scarcely trace-
able. When two such fibers cross each other in close proximity it is
sometimes almost impossible to trace the independent courses of the
two in the region of the apparent intersection.

There is less difficulty, however, in tracing the threads of ^. At the
stage shown in figure 71 — which corresponds with that in figure 29
(Plate 3) the threads are very fine and well oriented. In this nucleus
there can be seen a loop of heavier threads (A), which have the ap-
pearance of being two, loosely wrapped around each other. The ac-
cessory, as shown at X, also forms a he^vy spireme loop at this stage,
but it is so much heavier than the one described that there can be no
confusion between the two. The deeply staining loop of interlaced
threads I interpret to be the spireme of the chromosome-pair A. In
figure 73 is shown an yl-spireme which has not completed its conjuga-
tion. It will be noticed that of the other threads in this nucleus
some are double and some are single; and, furthermore, that the
double ones are twice the width of the single ones. In figure 74 the
spiremes of the pair A have completely conjugated, though the general
appearance of the cell indicates that the stage is no further advanced
than that shown in figure 73. Figures 75-78 (Plate 7) show the pair
A in various stages of conjugation at stages closely corresponding to
those shown in figures 73 and 74.

I have already traced the pair A from the pachytene stage to the
metaphase of the first spermatocyte division, so that it now remains to

78 bulletin: museum of comparative zoology.

examine only the stages following that division. Figure 80 (Plate 7)
shows a telophase of the first spermatocyte division as seen when
looljing from the equator toward the centrosome. There are eleven
dyads here, and since the whole number could easily be counted, the
accessory is not present. One of these dyads is more deeply stained
than the others, and, judging from its size relations, I think we may
identify this dyad as one from chromosome-pair A. This conclusion
receives still stronger support from figures 81 et seq. Figure 81 is of a
stage slightly later than the one in figure 80, and here we can see the
dyad A in addition to the accessory dyad, which is less deeply stained
than the others and is surrounded by a well-defined clear space, as
indicated by the dotted line. In figure 82 is drawn a telophase in
which the dyad A is shown in both the daughter cells. From these
figures (80-82) it is apparent that this element cannot be confused
with the accessory at these stages. In figures 83 and 84, however, it is
less easy to distinguish between them. But a long and careful study
has convinced me that the accessory, having early passed through a
stage of greatest diffusion, soon becomes condensed, while the other
dyads are undergoing dissolution. Dyad A, on the other hand, at
first remains more condensed than the others and then gradually be-
comes diffused like them. Figure 83 shows an early interkinesis stage
in which the large accessory dyad (X) is more condensed than that
shown in figure 81, but where dyad A is still more dense. In figure 84,
which is of a stage not much further advanced, the accessory is seen
to be the most condensed dyad (X), whereas A has gone far toward its
stage of diffusion corresponding to that of the other chromosomes.
That the accessory remains condensed throughout interkinesis is
further shown in figures 46 and 48 (Plate 4). Figure 48 further shows
that in the prophase of the second spermatocyte the A dyad condenses
earlier than any other dyads except that of the accessory.

It was impossible to trace the A dyad into the metaphase of the
secondary spermatocyte, but in the telophase it may again be recog-
nized by its characteristic deeper staining and by its size relations.
In figure 55 (Plate 5), which is a polar view of such a telophase, three
deeply staining chromatic masses are shown. The larger one (X) is
probably the accessory, the next in size, the monad of A, and the
smallest, a monad of B (p. 79). Figure 85 (Plate 7) shows a somewhat
later telophase, in which diffusion has progressed a little beyond that
seen in figure 55. About the same relative staining qualities and
relative sizes are seen as in figure 55. The accessory appears in only
half of the secondary spermatocytes and spermatids, however, and

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 79

figure 86 is of a pair of spermatids in which the accessory does not
occur. The diffusion process has here proceeded beyond that shown
in figure 85, but the two more deeply staining masses, representing
the monads of A and B, can readily be distinguished.

We may on the strength of this evidence say that the chromosome-
pair A can be traced from the spermatogonia to the spermatids, thus
demonstrating a case of morphological identity through all these
generations and stages.

2. Chromosome-pair B. — If morphological continuity is the gen-
eral rule, and if the peculiarities of the chromosome-pairs B and C
are distinctive enough, we should be able to trace the latter as we have
traced A. In many stages, however, these smaller pairs are not so
easily recognizable as was the pair A, but it has been possible to ob-
tain good evidence for individuality even through them.

I have called attention to a dyad in interkinesis, and a monad in
the spermatids, which seem to satisfy requirements for identification
as the element B. In figures 80 and 83 (Plate 7), for example, is seen
a dyad smaller than A, which stains almost as deeply as the latter.
An element with similar properties is to be seen in figures 55 (Plate 5),
85 and 86 (Plate 7). This element {B in the figures) has such size
relations when compared with A and the smallest element (as seen in
figure 80) as we should expect in B; when we consider, further, that
in the postspireme stages B stained more deeply than the majority
of the other tetrads, the staining qualities exhibited in these later
stages should also furnish a means of identification.

When we look at the spermatogonia! telophases of the same indi-
vidual from which figure 63 was taken, that is, one in which the
components of pair B are equal, we can readily find a pair of chromo-
somes that possesses the chief characteristic by which B was recog-
nized in the postspireme stages, namely, the presence of a prominent
polar granule at each end and a third not far from the middle, though
nearer the distal end. Examples of such spermatogonia! telopliases
are sliown in figures 87-96 (Plate 8). A further consideration of
these stages is given on page 83.

The study of cliromosome-pair B in the growth-period has furnished
some of the most interesting data on the subject of chromosome
individuality that I have secured. An analysis of this pair in its
extended condition in the pachytene stages of the first spermatocyte
was made for one of the specimens (no. 772) and then comparisons
drawn between the conditions in this and those in all the other animals
in the series studied.

80 bulletin: museum of comparative zoology.

Figure 58 (Plate 5) shows the element during the zygotene stages —
as indicated by the incompletely conjugated pair of threads near the
middle of the figure — in a condition of complete parallel association
for the two conjugants, but a condition in which the members of the
pairs of granules, the chromomeres, are distinct. A close examination
of this spireme of B discloses a series of cliromomeres in addition to,
and smaller than, the three already mentioned as characterizing the
element.

For convenience in description the more prominent granules or
chromomeres will be given separate designations. The five granules
which I wish to mention more particularly will be numbered in order
from the proximal (no. 1) to the distal end (no. 5, figure 58). I shall
also call attention to the two paii's of small granules between numbers
3 and 4 and to the two pairs of still smaller ones between 2 and 3. I
should not omit to direct attention to the series of granules between
numbers 1 and 2 and between numbers 4 and 5, but detailed considera-
tion' of those already mentioned will probably suffice for the purpose
in view.

I was at first impressed by the constancy in relative size and posi-
tion with which some of these granules recurred in different examples
of B and at different stages in a single individual (no. 772). It then
occurred to me to compare the same element at about the same stage
for all the thirteen animals from which material was available. Figure
97 (Plate 8) is the result, each of the separate draAvings having been
taken from a different animal. The constancy with which the minute
details of size and arrangement of the parts of this pair were repeated
in all of the individuals was surprising. Not only are the five more
prominent chromomeres repeated in approximately the same relative
sizes and positions, — as shown in figure 97, where corresponding
granules are connected by dotted lines, — but there is likewise a strik-
ing correspondence in the more minute details. For example, the
segment between the granules numbered 3 and 4 always contains two
pairs of granules of about the same relative size, though they vary
somewhat in relative position. On the other hand, the segment be-
tween 2 and 3 is characterized by the entire absence of any prominent
granules. In some cases, however, as in/, i, and k, figure 97, granules
can be made out in this segment, and when this is possible there are
always two pairs of very small ones in the same relative positions.

It is true that there are some variations in the appearances of the
segments between granules numbered 1 and 2 and between 4 and 5,
as well as differences in the actual size of the numbered granules.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 81

These variations may be due to one or more of several causes: — (1)
DiflFerenees arise on account of slightly different reactions to the
fixatives and stains. (2) There is a tendency for adjacent granules to
fuse, thus causing apparent variations in number and relative size.
(3) There is a slight difference in appearance at different stages. (4)
The different positions assumed by the element with reference to the
optical axis of the microscope may account for some variation in
appearance. (5) Some individual variation from animal to animal
might be expected.

It will be noticed that the distal granule (no. 5) is single in all the
individuals except those represented at b and c, where it is double.
This is in accordance with the statement previously made (page 70)
that chromosome-pair B is unequal in eleven and equal in only two
of the thirteen animals studied. It will also be noticed that the
granules at the proximal end (no. 1) frequently become associated
with other polar granules in a composite granule (a, h, i, j, I, m, fig. 97),
and that with one exception (/) the distal end is free. The formation
of composite granules is a characteristic feature of this material, as
already noted on page 64.

One of the granules of the proximal pair (no. 1) of individual /,
figure 97, is seen to be enlarged and less deeply stained than its mate.
Another example may be seen at k. I believe this to be an example of
a modification similar to that described in connection with the distal
granules of the pair A (p. 66). In B, this condition appears with
much less frequency, for in a count of 84 cases taken at random from
one individual only 14 (16|%) had one of the granules in the expanded
condition. This modification may persist into the tetrad stages, as
was the case with A. No case was found in which both granules were
expanded.

In order to test the variability of the details of constitution of the
element 5 in a single animal, a study was undertaken with this object
in view. Sixteen drawings (fig. 98, a-p) were made of examples taken
at random from a single slide. Comparison shows about the same
degree of constancy in the composition of the elements here as in the
set from different animals. Some of the variations may be pointed
out. For example, the relative lengths of the segments 1-4 and 4-5
in example a, figure 98, are somewhat different from those in example
h. I think we may assume that the spireme threads possess some
elasticity and that the variation in arrangement, association, and posi-
tion of the several segments of the spireme may frequently bring about
stresses which may stretch some of the threads or parts of threads to a

82 bulletin: museum of comparative zoology.

greater or less degree. The tendency for adjacent granules to fuse
probably accounts for some of the variations to be noticed. If one
will compare in order the examples I, m, a, and c, (fig. 98) the different
steps in the fusion of granule no. 4 with the smaller, yet prominent,
granule close to it will be seen. As the threads shorten during the
later pachytene and postspireme stages, this coalescence of adjacent
granules becomes more noticeable and the individual granules all
finally lose their visible identity in the compact metaphase chromo-
somes.

It will be observed that the members of a pair of granules may also
appear to be fused together into a single ma|SS. An example of this is
seen in figure 98, n, granules 4 and 5. This fusion must be very
temporary in character, since it is not the general rule, and since the
granules separate again in the postspireme stages, as shown in figure
63 (Plate 6) ; yet so close an association of these granules apparently
offers opportunity for the exchange of chemical substances between
them. In the case of the proximal granules (no. 1), the members may
not only fuse with each other but, as previously noted, characteristi-
cally unite with the polar granules of other chromosome-pairs to form
the composite granules. The association is fully as close as in that of
any single pair, for frequently all traces of the outline of individual
granules is entirely lost, as, for example, in figure 34 (Plate 3). Al-
though the individual granviles separate out again in the postspireme
stages, if we admit that there is an exchange of chemical substances
between members of a single pair of granules, I think we must also
assume it for the polar granules of the different chromosome-pairs.

At j (fig. 98, Plate 8) may be seen another example of an expanded
polar granule, such as has already been mentioned. The possible
significance of this peculiarity will be discussed on page 112.

It will be instructive to compare the members of particular pairs
of granules. Figure 58 (Plate 5), as already mentioned, represents a
zygotene stage. The paired chromatic threads near the middle have
jiist begun to conjugate, while in the case of chromosome-pair B, in
the left half of the figure, the two conjugating threads have only
recently come to lie side-by-side, for the members of the different pairs
of granules are yet distinct. This condition fortunately gives us an
opportunity to compare the relative sizes of the members of each pair.
On examination it will be seen that the members of the pair numbered
4 are not equal in size. This is also true for the pair numbered 3. In
the case of number 4, the disparity in size between the two granules
is considerable, and it is interesting to observe that this difference in

WENKICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 83

size can frequently be noticed throughout the pachytene stages.
Examples of this may be seen at b, e, i, I, and o in figure 98 (Plate 8),
which are drawn from the same individual as figure 58. Similar
conditions are also to be found in other individuals, as will be seen in
figure 97,. a, c, e, j, and /.

This pair of chromosomes can be recognized in the spermatogonia
by the presence, in the telophase, of the three most prominent granules,
those I have numbered 1, 4, and 5 in the pachytene stages. Examples
of such telophases are represented in figures 87-96. In two cases,
where the cliromosomes had become considerably elongated in the
general diffusion process of the telophase, I was able to make out
granules 2 and 3 also with their characteristic relative positions and
sizes. These are shown in figures 95 and 96. Where both chromo-
somes of the pair are recognizable in the same nucleus, there seems to
be in every case a difference in size between the two middle granules
(no. 4). This difference is probably directly related to the difference
noted in the zygotene stage (fig. 58) and the pachytene stages (figs. 97
and 98).

Thus, aside from finding a striking degree of correspondence in the
minute organization of the chromosome-pair B for all the individuals
studied (in the pachytene stages), it has also been possible to trace the
pair through all the stages from the spermatogonia to the spermatid,
except in the preleptotene and leptotene stages. Figures 30 (Plate 3)
and 58 (Plate 5) show that conjugation is completed at a relatively
early stage in the zygotene. This precocious conjugation is possibly
facilitated by the relatively small size of this pair. The failure to
recognize the pair in the leptotene and immediately preceding stages
is probably due to the fact that it has not so great a differential stain-
ing capacity as has pair A, and to the lack of sufficiently long con-
tinued study with this object in view.

A further peculiarity of chromosome-pair B may be seen upon an
examination of figures 99 and 100 (Plate 9). There it will be seen
that one end of the tetrad has a peculiar roughened or brush-like
appearance, to which McCIung ('14) has already called attention.
It will be noticed in the same drawings that the accessory chromosome
also presents a similar appearance. Furthermore, a like condition
is to be seen at the longer end of C, as shown in figure 100, and at the
end of some of the other autosomes, as seen in figure ^9. The rough-
ened contour of the accessory in both metaphase and anaphase of
the first spermatocyte division was noted for Phrynotettix by Miss
Pinney ('08), and has been described for other species of Orthoptera,

84 bulletin: museum of comparative zoology.

for example by Davis ('08) for Dissosteira and Stenobothrus, and
McClung ('14) for various Acrididae. But no one, so far as I am
aware, has described such a condition for any of the autosomes.
Figure 99 is from a sUde that had been treated with Heidenhain's
iron-haematoxyhn stain, but the destaining process had been carried
farther than in most of the other sUdes. Figure 100 is from another
individual, the slides of which had been stained by Flemming's tri-
color method, but had not been excessively differentiated. It will be
noted that the autosomes in this figure do not exhibit the roughened
synaptic ends that are seen in figure 99. It seems probable, therefore,
that differences in the staining process may have much to do with
the appearance or non-appearance of the roughened condition. In
heavily stained slides even the accessory, as well as the tetrads B and
C, may appear with a smooth contour. In this connection, I may call
attention to these several points: — (1) Tetrad B is unequal in both
the cases figured and the roughened end corresponds to the large
distal granule on the larger conjugant (see fig. 64, Plate 6). (2)
Tetrad C is likewise unequal and the roughened end also corresponds
with the large distal granule at the end of the larger of the two com-
ponents (see fig. 65). (3) The polar granules usually occur at the
proximal end, i. e. the end to which the spindle-fibers attach, and
therefore the roughened tips of the autosomes in figure 99 probably
correspond to the polar granules of these elements. (4) The accessory
chromosome and the polar granules have the common property of
remaining condensed while the rest of the chromatin is diffuse, as well
as the common property exliibited in these two figures (99 and 100,
Plate 9). The suggestion therefore offers itself that there may be
some common physical or chemical properties underlying the corre-
spondence in behavior between the accessory and the polar granules.

3. Chromosovw-pair C. — The drawings of chromosome-pair B in
figure 64 (Plate 6) and those of C in figure 65 were made from sections
cut from the same testis. An examination of the spermatogonial
telophases of tliis individual revealed the larger members of each of
these pairs very well defined, as indicated in figures 101-105 (Plate 9).
No attempt was made to recognize the smaller members of these pairs,
because they lacked characteristics, other than size, distinctive enough
to make recognition certain. With the larger members of these pairs,
however, the distinguishing features are so pronounced that I think
there can be no doubt about the identification.

I did not attempt to follow these elements through the preleptotene
and leptotene stages, but I have no doubt that careful enough study

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 85

would enable one to trace them, as was done in the case of chromo-
some-pair A. It is a matter of no small importance, I believe, that
each of the "selected" chromosome-pairs has been recognizable by
means of one or both its members, in the spermatogonia as well as in
the spermatocytes.

On the other hand, when I came to search through the postspireme
stages of the other individuals for tetrad C, I was able to find the
condition shown in figure 65 in only two instances; but a careful study
of these stages in the rejnaining animals of the series revealed, in
place of the large unequal type shown in figure 65, two other types,
which are shown in figure 107, c~m. Figure 106 presents an example of
tetrad B from each of the thirteen animals from which material was
available for study, and figure 107 a similar series of tetrad C. The
corresponding letters, a, h, c, etc., in the two series represent the same
animal. We may therefore, speak of the different animals as a, b, c,
etc. Cliromosome-pairs B and C are the smallest in the whole com-
plex and it will be seen from these two series of drawings that, except
in a and b, the pair C is the smaller of the two. In a and b, C is slightly
larger than B, as was determined by numerous comparisons in the
metaphases of the first spermatocytes. The difference in quantity of
chromatin in these two cases is quite small, however, and differences
in shape and 'behavior were largely depended on for identification.

For convenience in description, we may designate the three types
of chromosome-pair C as Ci, C2, and C3. By Ci will be indicated
the type, previously described, which is represented in figure 65, and
at a and b in figure 107. The type shown in figure 107, c-h, may be
designated Co, and that shown in figure 107 at i-vi, as C3. Thus it
will be seen that (with a possible exception yet to be discussed) of the
thirteen animals studied, two exhibited the type Ci, six the type C2,
and the remainder, five, the type C3.

If now we compare types Ci and C2, it will be apparent at once that
both members of the pair C2 resemble the smaller member of Ci.
The homology is striking if one notices the polar granules and the pair
of granules close to them, both of which appear in about the same
relative size and position in all the examples of both types (except h).
It is therefore not difficult to believe that type Co does actually repre-
sent a pair of chromosomes homologous to the smaller conjugant in
type Ci.

Turning to type C3, as shown in figure 107, i-m, it will be observed
that this is quite different from either Ci or C2. It represents an
unequal pair but the larger member is very different from the larger

86 bulletin: museum of comparative zoology.

one in type Ci. Furthermore the prominent chromomere near the
polar granules does not seem to be present, except possibly at vi (fig.
107). On the other hand the smaller eonjugant resembles those in 0%
in size and otherwise except for the prominent granule already men-
tioned. We might therefore be led to suppose that the smaller com-
ponent in Cs is homologous to the smaller one in Ci and the two small
ones in Ci. But if the example at h (fig. 107) be regarded, it will be
seen that this is a small pair lacking any prominent cliromomere near
the polar granules, and might therefore be thought to be homologous
with the smaller eonjugant in type Cz, if it be considered different from
those in type C2. However, even if the somewhat questionable posi-
tion of example h, be disregarded as to homologies, it still must be
admitted that we have at least three different types of chromosomes
appearing in these examples of tetrad C. I may again point out that
there is no chance of making a mistake as to the identity of these
elements, for the chromosome-pairs B and C are the smallest pairs in
the complex, and the different types of C are mutually exclusive, that
is, no two of them are ever found in the same animal. I might further
add that all the drawings were carefully outlined with a camera
lucida and the details filled in so as to represent as accurately as
possible the actual conditions as seen in the microscope. The matter
of the possible recombination and redistribution of these different
types is discussed on page 121.

h. The Accessory Chromosome.

The accessory chromosome has not been made an object of special
study here. Since it has been so thoroughly and so frequently de-
scribed for orthopteran material, it will suffice to give only a brief
account of it in this connection. In the first place, it should be stated
that the accessory can be recognized as a distinct chromatic individual
at practically every stage from the primary spermatogonia to the
spermatid. The fact that it forms a large and faintly staining vesicle
or "sac" in all the spermagonia except the last, probably accounts for
the occasional statement that it can be first recognized in the telo-
phase of the last spermatogonia, where it appears as a condensed mass
of chromatin, or a chromatin nucleolus.

Two points deserve to be emphasized: — (1) The accessory, more
than the other chromosomes, maintains an exclusive individuality in
nearly all stages. However, it sometimes does become associated

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 87

with other chromosomes, especially in the growth-period. Here its
polar granule may unite with those of the other chromosomes to form
a composite granule. (2) Its behavior, while unique in many respects,
differs from that of the autosomes in the degree and the chronology,
rather than in the kind, of its changes. The autosomes form vesicles
in the telophase of the spermatogonia, as Sutton ('00) long ago pointed
out, just as does the accessory, but they are not quite so large or
persistent as with the latter. In the growth-period the accessory
forms a looped spireme, just as the autosomes do (see fig. 71 and 72,
Plate 6), but its thread is much more dense and heavily stained than
the others. Although it fails to find a mate in synapsis, its behavior
is very like that of the autosomes and its spireme loop may occupy the
entire circumference of the nucleus. The process of shortening and
thickening, which all the chromosomes undergo, occurs very early in
the case of the accessory and it passes through most of the growth-
period as a rather compact mass of chromatin. In the postspireme
stages, at the time when the chromatids separate from each other by
the formation of the secondary longitudinal split, the accessory forms
a more or less bent or twisted rod, which often shows a longitudinal
split. This split must be homologous to the secondary split seen in
the autosomes, which divides longitudinally each of the chromosomes
united in synapsis. In the anaphase of the first spermatocyte division
its halves separate at the distal end, so that it forms a dyad similar in
all respects to those of the autosomes, except for its more roughened
condition. In fact, the accessory dyad cannot always be distinguished
from the others in the late anaphase. In the metaphase of the second-
ary spermatocytes it divides along with the autosomes and usually
is indistinguishable from them. Its behavior may therefore be more
nearly parallel to that of the whole series of chromosomes than we are
sometimes led to suppose.

c. The Spermatogonial Divisions.

Let us now consider the subject of persistent chromosomal organiza-
tion from the standpoint of the spermatogonial divisions. Figures
1-20 (Plates 1 and 2) are intended to represent the most important
stages included in the cycle of changes from one cell division to the
next. In this description no reference will be made to the selected
chromosomes, but the general behavior of the chromatin material
will be considered. We shall also leave out of account the mechanics

88 bulletin: museum of comparative zoology.

of the division process and concern ourselves chiefly with the fate of
the chromosomes after their division and separation has been accom-
phshed.

In my account of the accessory chromosome, I have already men-
tioned the formation of sacs or vesicles in the telophases of the sperma-
togonia. In an early telophase, such as is shown in figure 5 (Plate 1),
the chromosomes are clumped together in a rather compact mass at
the pole of the spindle. But the distal tips of the larger chromosomes
may be seen projecting in various directions. Following the clumped
condition, stages occur during which the chromosomes begin to ex-
pand and to separate from one another. iVt the same time there is
developed about each chromosome a hyaline area, at first small in
extent, but gradually enlarging as the chromosomes continue to ex-
pand. These conditions are shown in figures 6-9. Figure 6 is a side
view and figure 7 a transverse (optical) section of the same stage.
Figures 8 and 9 are likewise side view and optical section, respectively,
of a later stage. At this later stage it will be seen that a membrane
has been formed at the boundary between the hyaline area and the
cytoplasm. We are therefore probably dealing with sacs or vesicles
similar to those described by Sutton ('00) for Brachystola.

What is the origin of these sacs? Does the hyaline region as it
first appears represent material from the cytoplasm, or from the chro-
mosomes, or is it an artifact resulting from the contraction of the
chromatin under the influence of the fixative? That it is not an arti-
fact, will be apparent, I believe, from the following considerations: —

(1) The chromosomes themselves, at the stages shown in figures 6 and
7, are larger than in the earlier stages represented in figures 3 and 4.

(2) The chromosomes continue to expand and the vesicles expand still
more rapidly, as will be seen from the later stages (fig. 8 and 9). (3)
The hyaline region as seen in figures 6 and 7 appears more highly
refractive than the cytoplasm which would not be the case if it were a
space produced by shrinkage of the chromatin.

A comparison of the conditions shown in figures G and 7 with those
shown in figures 10 and 12 will, I believe, show that the expansion of
the vesicles has been at the expense of the cytoplasm. The relative
volume of the space within the vesicle as compared with the volume of
the cytoplasm, is much less in these earlier stages (fig. 6 and 7) than
in the later stages (fig. 10 and 11). Further, it will be seen that the
expansion of the vesicles is accompanied by: — (1) an increase in the
size of the cell-body, (2) a diffusion of the chromatin into a kind of
reticulum within the space of each sac, (3) the breaking down of the

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 89

vesicular membranes between adjacent vesicles within the group,
especially at the polar end, (4) the formation of an irregular nuclear
membrane from the outer walls of the vesicles, (5) the apparent
anastomosing of the edges of the networks arising from the diffusion
of the chromosomes in adjacent vesicles. The walls bounding the
original vesicles are still to be seen in figures 8-13, and this is particu-
larly true of the accessory chromosome, the vesicle of which persists
till a late prophase.

What can we now say as to the continuity of the individual chromo-
somes? Let us first follow the changes undergone by the accessory
chromosome. Figures 1 and 2 are of metaphases, in which all the
chromosomes except the accessory are compact and smooth in outline.
This is roughened in outline and seems to have already begun the
process of expansion which characterizes its behavior immediately
after division. In figure 4 a hyaline area of considerable extent has
already been formed about the accessory, and close examination
reveals also a narrow hyaline area just beginning to develop around
each of the autosomes. By the time the stage shown in figure 6 is
reached, the substance of the accessory has become distributed through
the entire space of the vesicle which accompanied the formation of the
hyaline area. In its distribution within the vesicle, the chromatic
substance is more concentrated on the periphery of the sac, than
through the central space. The vesicle continues to expand along
with the expansion of the nuclear material as a whole, until the stage
of greatest diffusion of the autosomes has been reached (fig. 12 and 13).
At the stage shown in figures 14a and 14b (Plate 2) the chromatin has
begun to concentrate towards the axes of the sacs, but this process
seems to be less advanced in the accessory {X, fig. 14b) than in the
autosomes. These are the earliest of the prophases. In the later
prophases, as shown in figures 17 and 20, the accessory becomes
concentrated as a coiled thread running down through the middle of
the vesicular space. The wall of the vesicle persists longer than does
that of the nucleus as a whole or that of the other autosomes (fig. 20).
There can be no question, it seems to me, that the accessory main-
tains a persistent individuality through all these stages.

If now the changes undergone by the autosomes be followed, we
shall find for them also evidences of persistent individuality. I think
no one would deny a persistent individuality up to the stages shown in
figures 8 and 9 (Plate 1) . In these figures the chromatin has become
reticular, but the masses representing individual cliromosomes are
still quite distinct and surrounded for the most part by the persisting

90 bulletin: museum of comparative, zoology.

walls of the vesicles. The method of formation for these vesicles
parallels very closely that described for the accessory, the chief
difference being that in the case of the accessory the process is much
more rapid. In figures 10 and 11 we find the chromatin much diffused
and occupying most of the space within the original vesicles. The
vesicular walls are no longer visible, however, except on the periphery
as an undulating nuclear membrane, and around the accessory. In
spite of this fact, the chromatic masses or blocs, each of which has
arisen from a single chromosome, are still recognizable as distinct from
one another. This is especially well shown in the optical section
drawn in figure 11. There are only eighteen masses shown in this
section, but the apparent reduction in number need cause no appre-
hension as to the fate of the other members of the complex. It fre-
quently happens that the chromosomal vesicles do not all lie parallel
to each other, so as to be represented in a single transverse section,
and some may even assume a position at right angles to the axis of the
majority. Such a case is shown in the upper left-hand corner of figure
10. If, now, we examine the stages shown in figures 12 and 13, which
are of the period of greatest diffusion that I have been able to find, we
may still see, both in optical section (fig. 13) and in side view (fig. 12),
the positions of the individual chromosomes represented by a more
condensed band or core. In the case of the optical section, nearly
the complete number of chromosomes, as represented by these denser
masses or cores, can be counted. It is true that there seems to be an
anastomosing system of fibrils connecting the adjacent masses, but
this need not mean that there has been a loss of chromosome-iden-
tity in a common nuclear mass.

An early prophase is represented by figures 14a and 14b (Plate 2),
which show the two sections of a single cell. We see at this stage the
beginning of the process of chromatin concentration which results,
finally, in the formation of the condensed chromosomes ready for the
next division. The chromatic material of each chromosome first con-
centrates near the middle of the region that it occupied in the nucleus in
the diffuse condition. There is thus formed a loosely reticulated core
(fig. 14a and 14b), out of which there develops a spirally coiled thread,
as shown in figures 15a and 15b. The two stages represented in figures
14 and 15 are very close together in time, for they occurred side-by-
side in the same cyst. These coiled threads are at first rather small
in diameter, but they rapidly thicken and shorten, as indicated in
figures 16-20 (Plate 2). During the process of shortening and thick-
ening the outlines of the vesicular walls become more distinct. This is

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 91

especially true of the distal pole, as shown in figure 18. It would seem,
therefore, that the vesicular membranes first became formed, then
largely disappeared, and later reappeared in part. I am inclined to
believe that they actually persist to a greater extent than is apparent.
There cannot be any doubt, however, that the vesicles do coalesce at
the polar end of the nucleus, for there the individual polar granules
frequently fuse to form composite granules, such as may be seen in
figure 12 (Plate 1) and figures 14, 15, 16, and 19 (Plate 2).

The first indication of the longitudinal split which forecasts the next
mitosis was discernible at a stage such as is shown in figure 17. From
this stage on to the metaphase, however, the split was clearly visible.

I believe that the evidence here presented furnishes very good
grounds for believing that the chromosomes do not lose their individ-
uality in passing through the so-called 'rest-stage' between the two
successive cell-di\'isions.

d. The somatic Nuclei.

Only slight attention has been given to somatic cells in connection
with the subject of the individuality of chromosomes, but some points
were noted which it seems worth while to record. The connective-
tissue nuclei within the follicle always divide by the indirect or mi-
totic method. The details are similar to those just described for the
spermatogonia, except that individual chromosome-vesicles, even for
the accessory, are less conspicuous — in fact, in my limited study of
these cells I have not recognized the accessory chromosome with
certainty. The only evidence of amitosis is a lobulated condition of
the resting nuclei ; that condition is a very characteristic one, but has
no more significance as to amitosis than the lobulated appearance of
the spermatogonia! nuclei. In the diffused chromatin-stages — telo-
phase, rest-stage, and early prophases — the polar granules appear,
coalesce more or less to form composite granules, and separate out
again just as they do in the spermatogonia. Furthermore, it is possi-
ble to find chromosomes in the telophases that exhibit all the chief
characteristics of the "selected" chromosomes. For example, in
Plate 9, B, figures 108, 109 and 110, are to be seen diffusing chromo-
somes with the characteristic features of one of the larger members
of chromosome-pair B. It would seem from this evidence that the
same morphological constitution of individual chromosomes persisted
even in these somatic cells.

92 bulletin: museum of comparative zoology.

Going outside the follicle, it is of interest to note what appears in
the nuclei of the follicular investment. This investment is a thin
membrane inclosing the follicle, forming the outer of the two laj^ers
composing the follicular wall. In this membrane the nuclei are very
much flattened, so that the chromosomes lie nearly all in one plane.
Figures 111 and 112 indicate the chromatic conditions in two such
nuclei. It is, I believe, a significant fact that the chromatic masses
to be found in these nuclei are in number approximately equal to the
unreduced number of chromosomes found in the spermatogonia.
Exceptions, it is true, occur; adjacent cliromosome-masses may be-
come intunately associated, or one individual mass may become
divided into partially separated masses. These nuclei are fully
differentiated and are destined never to undergo another cell-division.
They must gradually lose their functions and will finally " die in their
tracks." The different conditions of the chromatin in the different
nuclei suggests that the process of senescent degeneration may have
already set in. The important fact still remains, that the individual
chromosomes have a tendency to remain distinct from each other,
even in these highly differentiated nuclei in a period not only of 'rest'
but perhaps of senescence.

D. Summary of Observations.

1 . The general topographical relations of the different generations
of male sexual cells in the testes of Phrynotettix magnus are typical
for the Acrididae.

2. For purposes of accurately following the history of the changes
undergone by the chromosomes from the pachytene stages of the first
spermatocyte to the time of mitosis, three individual chromosome-
pairs were selected, each of which possessed characteristics by which
it could be recognized in all the stages concerned. These three pairs
were designated, for convenience, "A," " B," and "C." A study of
these three chromosome-pairs showed : — (a) that there is a longi-
tudinal split in the pachytene stages, which persists into the tetrad
and later stages (this is called the primary longitudinal split); (b)
that a tetrad is formed out of a spireme segment by (1) a separation
along the primary split, and (2) the appearance of a secondary longi-
tudinal split along the middle of each of the two parts separated by
the primary split.

3. Tetrad " ^ " opens out along the plane of the secondary split.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 93

the proximal ends separating and moving about 90 degrees apart, so
that a rod-shaped element is formed the middle of which represents
the distal end of the original segment. The rod, thus extended, be-
comes oriented with its long axis parallel to that of the spindle and it
separates in the middle, thus bringing about an equational division.
Tetrad "A" also forms rings, but these were not traced into the meta-
phase, and their later behavior is not known.

4. Tetrad " B" occurs in one or the other of two forms: either (1)
as an equal pair (in two of the thirteen animals), or (2) as an unequal
pair (in the other eleven animals) . The unequal pair differs from the
equal in the absence of a large terminal granule at the distal end of
one of its members. Both types show the same behavior, opening
out at both ends of the segment so that a cross is formed. The
separation along the plane of the primary split is the greater and occurs
at the distal end; but the cross becomes so oriented on the spindle
that the short arms (i. e. the proximal end of the original segment) are
attached to the spindle-fibers. Separation in metaphase is therefore
along the plane of the secondary split, thus constituting an equational
division.

5. Tetrad "C" occurs in three forms, designated Ci, C2 and C3.
Ci is composed of very unequal elements, the larger of which possesses
a relatively very large terminal knob or granule that is not present
on the other. C2 is a pair with equal members each of which appear
to be homologous to the smaller member of Ci. C3 is a pair of unequal
elements neither member of which appears to be exactly homologous
to the components of Ci and C2. The smaller member resembles
those of Co and may be homologous to them. The larger member is
midway in size between the two members of Ci. Co and C3 divide
equationally in the first maturation mitosis, but Ci divides half the
time equationally and half the time reductionally in this first division.
When dividing reductionally the two unequal dyads follow the law of
chance in their distribution with reference to the accessory chromo-
some, which passes to one pole undivided.

6. Study of the early growth-stages of the first spermatocyte shows
that each of the chromosomes of the telophase of the last sperma-
togonial division forms a long spirally coiled thread, which uncoils
and stretches out to form the leptotene threads of the primary sperma-
tocyte. The leptotene threads conjugate side-by-side (parasynapsis)
to form the double threads of the pachytene stage.

7. It was possible to recognize the chromosome-pair A in the
spermatogonia as two separate chromosomes (telophases) and to

94 bulletin: museum of comparative zoology.

trace the pair through all the stages from the spermatogonia to the
spermatids, thus constituting a demonstration of a case of continuous
identity, or individuality, through these stages. It was also possible
to recognize chromosome-pairs B and C in the spermatogonial telo-
phases as well as in the second spermatocytes and spermatids.

8. In the earlier pachytene stages, chromosome-pair B was found
to have a definite arrangement of granules or chromomeres, and it was
shown that the relative sizes and positions of these chromomeres re-
mained constant for similar stages, not only in different cells of a
single individual, but also in all the thirteen animals.

9. The spermatogonial divisions showed that each chromosome
forms a sac or vesicle in the earlier telophases, and that it expands
and becomes diffused within these vesicles; that, although the vesicles
appear to coalesce, there is always a remnant of each chromosome
visible in the center of the region occupied by the vesicle, and that in
the prophase the chromatin concentrates about this remnant or core
and there forms a spirally coiled thread, which develops into a prophase
chromosome.

10. Study of somatic cells showed:— (1) that chromosome B
could be recognized in the connective-tissue cells within the follicle,
and (2) that cells of the follicular envelope, which are probably in a
state of senescence, still preserved the normal number (23) of chromatic
masses.

11. The polar granules are constant features of the organization
of the individual chromosomes, as was shown by Pinney ('08); but
in some cases (chromosome-pairs A and B) they may become modified
to give rise to expansions which resemble the "vesicles" described by
Carothers ('13), as well as the "plasmosomes" of most authors. The
polar granules tend to unite into composite granules at all of the diffuse
stages of chromatic evolution.

12. The accessory chromosome behaves in the manner that is
typical for the Acrididae. It forms a large separate sac or vesicle in
the earlier spermatogonial generations and a peripheral compact mass
in the telophase of the last spermatogonial division. During the
leptotene and zygotene stages it may unravel into a long loop, which
in some cases is equal in length to a great circle of the nucleus. In the
pachytene stages it reassumes a compact form, but may b 5 attached
by its polar granule to the polar granules of other cln-omosomes and
thus become attached to a composite granule. It passes to one pole
undivided in the first maturation division but divides in the second.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 95

III. DISCUSSION.
A. Synapsis and the Maturation Divisions.

It is very difficult to separate the subjects of synapsis and the matura-
tion divisions from the subject of chromosome-individuaHty. Yet
for the sake of clearness it seems best to make such an artificial separa-
tion. It might also be possible to separate from each other the sub-
jects of synapsis and maturation divisions, but since the two are so
intimately related, it seems better to discuss them at the same time.

Anything like a complete review of the literature on the subjects of
synapsis and reduction divisions will not be attempted here, in view
of the extensive general reviews in the monographs of Gregoire ('05,
'10) and Vejdovsky ('11-12), and the reviews relating particularly
to orthopteran spermatogenesis by Davis ('08) and McClung ('14).

a. Results from Orthoptera.

McClung ('14) has so recently reviewed the literature on Orthop-
tera dealing with this subject that it will suffice here to summarize
briefly the results. The different views may be classified as follows : —

I. Synapsis not considered.

a. Both maturation divisions reductional.

1. Wilcox ('94, '96, '97, '01), Caloptenus.

b. Both maturations equational.

1. De Sinety ('01), various Orthoptera.

2. Granata ('10), Pamphagus.

c. First division transverse.

1. Vom Rath ('92, '95), Gryllotalpa.

2. Farmer and Moore ('05), Periplaneta.

3. Jordan (*08), Aplopus.
II. Synapsis described or assumed.

A. Telosynapsis described or assumed.

a. First maturation division reductional.

1. Montgomery ('05), Syrbula.

2. Stevens ('05), Blatta.

3. Wassilieff ('07), Blatta.

4. Zweiger ('06), Forficula.

96 bulletin: museum of comparative zoology.

5. Davis ('08), Acrididae and Locustidae.

6. Buchner ('09), Gryllus, Oedipoda.

7. Stevens ('10b), Forfieula.

8. Brunelli ('09, '10), Gryllus, Tryxalis.
b. Second maturation division reductional.

1. Sutton ('02, '03), Brachystola.

2. Baumgartner ('04), Gryllus.

3. McClung ('05, '08a, '14), various Orthoptera.

4. Stevens ('05), Stenopalmatus.

5. Nowlin ('08), Melanoplus.

6. Finney ('08), Phrynotettix.

7. Robertson ('08) Syrbula.

8. Carothers ('13), Acrididae.
B. Parasynapsis assumed or described.

a. First maturation division reductional.

1. Gerard ('09), Stenobothrus.

2. Morse ('09), Blattidae.

3. Stevens ('12a), Ceuthophilus.

4. Robertson ('15), Tettigidae.

b. Both divisions equational.

1. Vejdovsky ('11-12), Locustidae.

c. Division neither reductional nor equational.
1. Otte -('07), Locusta.

This classification ^ is interesting from two points of view. In the
first place, it indicates the diverse results that have been obtained by
the various investigators working on a limited group within which one
might reasonably expect to find a high degree of uniformity in chromo-
somal behavior. In the second place, the results that I have obtained
do not come under any of the classes in the above outline. As stated
on previous pages, I have shown (1) that the spermatogonial chromo-
somes develop into the fine leptotene threads, which conjugate by
paras;>aiapsis without the conjugants losing their identity, that is, the
line of conjugation is visible tlu-oughout the growth-period as the
'primary longitudinal split'; (2) that a second longitudinal split at
right angles to the first occurs in the early postspireme stages; and
(3) that the tetrads become so oriented on the first maturation spindle
that the resulting division is equational. Each of the dyads of the
second spermatocj^tes consists of parts of the two original conjugants,

' I have omitted reference to some papers which were non-committed on the points under
discussion.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 97

and these conjugant-halves become separated in the second matura-
tion mitosis, the result being, therefore, a reductional division. I am
thus able to support the careful studies of McClung and his students
as to the orientation of the tetrads in the first maturation spindle,
where the spindle-fibers become attached at the so-called 'synaptic'
or proximal ends, and therefore bring about an equational division.
I can likewise support the findings of those investigators who describe
parasynapsis. If we accept the view that one of the longitudinal splits
is in reality the line of separation between parallel conjugants, we can
also accept the observations of De Sinety ('01) as to the existence of
two longitudinal divisions.

McClung and those of his students who have worked on orthopteran
material have derived their results from studies confined largely to
spermatogonia and the postspireme stages. There is nothing in any
of their figures, however, which would be incompatible with parasynap-
sis. And the figures by Sutton ('02, fig. 5a, 5b, 6 and 7) of early
postspireme stages in Brachystola are much more satisfactorily inter-
preted from the standpoint of a preexisting parasynapsis than from
the standpoint of telosynapsis. I may also state that I have recently
examined some Brachystola material and am well satisfied that the
conditions there are quite comparable to those prevailing in Phry-
notettix. McClung in his latest paper ('14) accepts the possibility
of parasynapsis, and Robertson, who in 1908 argued foi- telosynapsis
in Syrbula in no uncertain terms, has recently found parasynapsis in
the Tettigidae (Robertson, '15).

A glance at the outline of the results of orthopteran studies given
above reveals the fact that parasynapsis has relatively few adherents.
I believe the failure to recognize this important stage has been due (1)
to the general unfavorableness of these synapsis, or lepto-zygotene,
stages foi" the elucidation of the conditions and a consequent failure
properly to interpret them, or (2) to attention having been largely
confined to the postspireme stages. That a study of the latter stages
could allow of quite diverse interpretations, I am keenly aware, for it
was not till I undertook to follow the history of individual chromo-
somes that I was able to arrive at any satisfactory conclusion as to the
sequence of events. I am confident that the use of the same method
on other material will reveal conditions similar to those that I have
described for Phrynotettix.

Where the chromosomes differ among themselves as to shape, as
they do in Stenobothrus, another source of confusion is encountered,
for very few authors have recognized the fact that cliromosomes of

98 bulletin: museum of comparative zoology.

different shape may behave differently in their orientation on the
maturation spindles. McClung has recently gone over this matter
in a very painstaking way, and I can agree with his conclusion that,
in general, the chromosomes with the spindle-fiber attachment termi-
nal, that is, rod-shaped chromosomes, are oriented in the first matura-
tion spindle so as to produce an equational division, while those wliich
have the spindle-fiber attachment non-terminal, that is, at the apex
of V-shaped cln-omosomes, become oriented so as to bring about a
reduction at the first division. This general rule is of course violated
when the pairs of rods are of unequal length, which usually (Baum-
gartner, '11; Payne, '12; Carothers, '13; Robertson, '15), but not
always (Ci, described in this paper), divide reductionally in the first
division. Davis ('08) sought to establish the behavior of the V-
shaped chromosomes of Stenobothrus as the type for the Orthoptera in
general. He correctly described the behavior of these chromosomes
in the maturations, but fell into error by attempting to make the rod-
shaped chromosomes conform to the same type of behavior. He also
failed to recognize parasynapsis. I have recently made a study of
the conditions in Stenobotlirus and may say that I found parasynapsis
for both forms of cln-omosomes, and that the V-shaped chromosomes
divide reductionally in the first maturation mitosis, as Davis de-
scribed, but that the rod-shaped chromosomes divide equationally
in the first division, as I found that they did in Phrynotettix.

>

b. Recent Work on Synapsis.

That parasynapsis has a wide occurrence, is evident from a glance
at the cytological literature, especially within recent years. Gregoire
in his two admirable monographs ('05, '10) has reviewed most of the
previous literature bearing on the subject of the behavior of the
chromosomes in maturation, and has endeavored to find a common
type of behavior for both plants and animals. He says ('10, p. 384) :
"Dans un bon nombre d'objets animaux et vegetaux, les cineses de
maturation s'accomplissent suivant le type d'une prereduction hetero-
homeoty pique preparee par une pseudo-reduction prophasique par
parasyndese ou zygotenie." In this " heterohomeotypique " scheme,
however, Gregoire has failed to distinguish the difference in behavior
between the chromosomes with terminal and those with non-terminal
spindle-fiber attachment. Since the publication of Gregoire's later
monograph, a considerable number of investigators have reported
the existence of parasynapsis.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 99

De Saedeleer ('13) finds in Ascaris all the typical stages of the
growth-period: — leptotene, zygotene, pachytene, and diplotene; he
consequently believes that parasynapsis occurs.

Among the Crustacea parasynapsis has been found by McClendon
and by Kornhauser for Copepoda and by Fasten for Canibarus.
McClendon ('10) found parasynapsis in both the oogenesis and the
spermatogenesis of Pandarus sinuatus, but could not decide which of
the maturation divisions were reductional. Kornhauser ('15) gives
a very full account of a careful study of the process of parasynapsis in
Hersilia apodiformis, thus confirming the earlier results, as to the
existence of parasynapsis in Copepoda, of Lerat ('05), Matschek ('09),
and McClendon ('10). In this paper he clears up the uncertainty in
regard to this group brought about by the unique theories held by
Hacker ('92) and his followers. Kornhauser demonstrates very
clearly that the so-called ' Querkerbe,' which Hacker and his followers
interpreted as the point of end-to-end union, is nothing more than the
synaptic point of the chromosomes which have a median or non-
terminal spindle-fiber attachment. The Copepoda are thus brought
into line with the majority of other forms. Fasten ('14) finds para-
synapsis in Cambarus, and although he is dealing with a very large
number of clu-omosomes (the diploid number is about 200), his figures
of the leptotene and zygotene stages are quite convincing.

With respect to work on insect material, I have already mentioned
that on Orthoptera. The results of Stevens are unusual in that she
has described telosynapsis for Blatta ('05), Stenopalmatus ('05), and
Forficula ('10b), while in Ceuthophilus ('12a) she found parasynapsis.
In the last mentioned article she says (p. 227) " I should not be sur-
prised if the range of variation should prove to extend from (a) cases
where there is nothing that could be called conjugation, but merely
such a pairing without contact even, as will secure segregation of
homologous maternal and paternal chromosomes to different daughter
cells, through (b) an intermediate condition of telosynapsis and less
intimate paras;yTiapsis, to (c) cases where homologous chromosomes
are so completely fused in parasynapsis that it is impossible to tell
whether the resulting cliromosomes which are segregated in mitosis
are identical with those that went into synapsis or not." It may be
that more intensive studies will reveal greater uniformity of behavior
than Stevens advocated.

Payne ('14), in a brief description of tetrad formation in Forficula
sp., reaches only tentative explanations and conclusions. He finds a
variable number of chromosomes in the two maturation divisions and
suggests that this might be accounted for b;^- supposing that some of

100 bulletin: museum of comparative zoology.

the spermatogonial chromosomes had failed to pair. He describes
two methods of ring-formation. The correctness of his conclusions
as to the succession of stages in some of his series might be questioned
on the ground that they are not different stages of the same chromo-
some. The series shown in his figures 2 to 11 probably represents a
normal method of ring-formation, viz., by the opening out of a para-
synaptic spireme segment along one of the longitudinal splits, with
the ends remaining in contact. The series in figures 12 to 16 might
also easily be derived from a parasynaptic segment. It is extremely
questionable whether the figures in the series 18-20 are arranged by
the author in their natural sequence; the reverse order is more likely
to be the correct one. His figures 19 to 27 (Plate 2) doubtless repre-
sent different shapes of the same chromosome-pair, the so-called
"middle granule" serving to identify the element. I would suggest,
however, that the stage that he represents in figure 19 may have
resulted from an opening out of a parasynaptic segment in the same
way that I have described for chromosome-pair A, in which case the
"middle granules" would be polar granules instead of "middle" ones.
In view of the rather far-reaching conclusions that Robertson ('15)
has drawn from his work on the Tettigidae, I would call attention to
some differences, as well as similarities, between his work and mine.
In the first place, he describes parasynapsis in the early stages of the
growth-period, as I have done, but in the postspireme stages he as-
sumes that the conjugants separate along the plane of conjugation
(primary longitudinal split), the separation beginning at the proximal
end. He shows in the metaphase of the first spermatocyte most of
the chromosomes as elongated rods with appearance and orientation
similar to that seen for my tetrad A. It will be remembered that in
the latter case the separation from the proximal end of the spireme
segment toward the distal end is not along the plane of the primary
split, but along the plane of the secondary split. I believe that
Robertson may have overlooked a similar behavior in the chromo-
somes of his material. Curiously enough the unequal elements that
he describes are very similar to the unequal type of chromosome B
and of chromosome C, in PhrynotettLx. I have shown that the
behavior of the cliromatids in B and C is similar to that in A. And
that in the cases in which C divides reductionally in the first division
the spindle-fiber attachment is necessarily shifted to the distal ends.
I believe that such a condition probably occurs in the unequal tetrads
described by Robertson, and if so, theoretical explanations, as to
how the elements came to be related to each other in the way they are.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 101

would be unnecessary. Robertson says he believes that the unequal
tetrad in Tettigidea -parvipennis has arisen by a loss at the distal end.
And he finds in other individuals an homologous pair each member
of which is equivalent to the larger member of the unequal pair.
This is just the condition that is presented by B in my material. And,
furthermore, by analysis of the pachytene and postspireme stages, I
am able to say just what part has been lost.

Another striking analogy between my observations and those of
Robertson occurs in connection with the larger unequal pair that he
has found in Acridium granulatus. He finds in some individuals a
small equal pair and in two individuals an homologous unequal pair,
the smaller component of which corresponds to either of the two
elements of the equal pair. This, again, is precisely the relationship
between types Ci and C2 in Phrynotettix. Here, too, we both failed
to find the other possible combination, namely, that of a pair of the
larger conjugants. These striking similarities lead me to think that
the elements described by Robertson may be explained in the same
way that I have explained them in Phrynotettix. If such be the
case, then the various assumptions as to doubling and "sesquiva-
lent" chromosomes will be unnecessary. I believe, further, that
without a doubt the unequal tetrads described by Robertson do
divide reductionallv in the first maturation division, but that the
spindle-fiber attaches at the distal and not at the proximal end; and,
furthermore, that there is a very good chance that the other chromo-
somes may behave as does chromosome-pair A in Plirynotettix, and
therefore divide equationally, as it does.

Of recent works on Hemiptera, the most interesting from the stand-
point of synapsis are the papers by Montgomery, Wilson, and Korn-
hauser. Montgomery ('11) advocated telosynapsis for many years,
but in this late paper, in which he described the spermatogenesis of
Euschistus, he concludes that pairing is by a process of parasynapsis.
I believe it to be a highly significant fact that Montgomery, at the
end of his very active career as a cytologist, and with his wide ex-
perience back of him, should reverse his former position on the subject
of synapsis and should find parasynapsis in this insect, which he had
studied and reported on at an earlier date ('01). He says (p. 743):
" In the growth period through the pachytene stage there is no longi-
tudinal splitting, for what I had previously ('01) interpreted as such,
I now find to be the line of conjugation .... Frequently at certain
points along a geminus the chromatin granules appear accurately
paired. But this does not appear until a rather advanced stage of the

102 bulletin: museum of comparative zoology.

strepsinema and is by no means regular" .... Further on (p. 753) he
says : — " During the past year I have also convinced myself of the
occurrence of parasyndesis in Plethodon, such as Janssens had de-
scribed for this object and the Schreiners for Salamandra."

Wilson ('12), in his critical study of the subject, first states the
questions that he believes must be answered in connection with synap-
sis and then gives his reasons for believing in the wide occurrence of
parasynapsis. He regards the following questions as still awaiting a
satisfactory answer: — "1. Is synapsis a fact? Do the chromatic
elements actually conjugate or otherwise become associated two-by-
two? 2. Admitting the fact of synapsis, are the conjugating ele-
ments chromosomes, and are they individually identical with those
of the last diploid or premeiotic division? 3. Do they conjugate
side-by-side (parasynapsis, parasyndesis), or end-to-end (telosynapsis,
telosyndesis), or in both ways? 4. Does synapsis lead to a partial
or complete fusion of the conjugating elements to form 'zygosomes'
or 'mixochromosomes,' or are they subsequently disjoined by a
reduction division?"

Wilson finds his own material (hemipteran) not altogether favorable
for a solution of the problems enumerated, but has been able to study
the preparations of Tomopteris, supplied by the Schreiners, and of
Batrachoseps supplied by Janssens. He studied also some orthop-
teran material, including Phrynotettix, secured from McClung.
After a study of Tomopteris and Batrachoseps he says (p. 384):
"Through the study of Batrachoseps and Tomopteris I have finally
been convinced — for the first time, I must confess, as far as the auto-
somes are concerned — (1) that synapsis, or the conjugation of
chromosomes two-by-two, is a fact, and (2) that in these animals
(perhaps also in the Orthoptera) the conjugation is a side-by-side
union, or parasynapsis." And again (p. 399), "The few observations
I have been able to make on McClung's preparations of Achurum,
Phrynotettix, and Mermiria. . . .lead me to the impression that a
side-by-side union of leptotene threads takes place here also."

Browne ('13), from a comparative study of the spermatogenesis of
three species of Notonecta, regards the evidence, though not abso-
lutely conclusive, as indicating a conjugation by parasynapsis.
Kornhauser ('14), as a result of a very careful study of the spermato-
genesis of two species of Enchenopa, finds conclusive evidence of a
parasynaptic union at the beginning of the growth-period. The
evidence for parasynapsis in the Hemiptera is thus seen to be very
strong.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 103

Of recent work on the Diptera, I may mention that of Stevens and
of Taylor on Culex, and that of Metz on Drosophila. Stevens ('10a)
finds parasynapsis in Culex, not merely in the growth-period of
spermatogenesis, but among other kinds of cells. She says ('10a,
p. 208) : — " That parasynapsis occurs immediately after the last
oogonial mitosis is certain and it is equally certain that the chromo-
somes are similarly paired in earlier generations of the oogonia,"
Again (p. 209) : — " In Culex it is quite certain that parasynapsis
occurs in each cell generation of the germ cells in the telophase," and
(p. 212), "It is interesting to find in Culex a clear case of parasynap-
sis in oogonia, oocytes, spermatogonia, and spermatocyte prophases
and then to see the same chromosome pairs appearing in the first
maturation metakinesis as though united end-to-end." (It is prob-
able that she has overlooked stages in the postspireme showing the
changes undergone by a parasynaptic spireme segment in its trans-
formation into a metaphase tetrad). Miss Stevens found six to be
the somatic number of chromosomes in Culex, the reduced number
being three. The side-by-side pairing of the chromosomes in nearly
all generations of cells studied, gave the appearance of a reduced
number in many situations where it would not have been suspected.
Taylor ('14) states that she found in Culex pipiens only three chromo-
somes in all the stages that she studied, i. e. in both the somatic and
germ-cells of both sexes. This is a very surprising result, but an
explanation may perhaps be found in the conditions observed by Miss
Stevens, namely, the tendency for the chromosomes in all kinds of
cells to pair between mitoses. A poor fixation might easily prevent
one from recognizing the double nature of a closely adhering pair
of chromosomes. Besides, Miss Taylor found some cells with six
chromosomes, and shows figures of some others with more than three.
It would seem more reasonable, then, to regard the prevalence of the
reduced number found by Miss Taylor as the result of the constantly
recurring tendency of the chromosomes to unite side-by-side between
successive mitoses, and possibly to poor fixation.

The results of Metz (*14) on Drosophila are interesting in this
connection, for he reports conditions in these flies similar to those
found in the mosquito. In this he confirms the earlier results of
Stevens ('08). He finds (p. 55) that: — "The chromosomes not only
exhibit a close association in pairs at nearly all times, but that before
every cell division the members of each pair become so intimately
united that they may be said actually to conjugate. Each pair, with
the possible exception of the sex chromosomes, goes through what

104 bulletin: museum of comparative zoology.

amounts to a synapsis in every cell di\'ision, so that in many cases the
figures closely resemble the haploid groups. Apparently this takes
place especially in early prophase, but a second conjugation may
occur dm'ing metaphases, just a short time before division. In the
second, or metaphase, conjugation, at least, it is worthy of note that
the union is unquestionably a side-by-side or parasynaptic one."
Thus we find parasynapsis in a greatly exaggerated form in these
examples from the Diptera.

Of recent studies on Vertebrata, we may note those of Snook and
Long on an amphibian, of Jordan on an opossum, and of Wodsedalek
on the pig.

Snook and Long ('14) find the same kind of evidence for para-
synapsis that has been presented for Batrachoseps by Janssens ('05),
for Salamandra by the Schreiners ('07), and by Wilson ('12) as quoted
at p. 102. This evidence, together with that announced by Mont-
gomery ('11) for Plethodon, forms a series of observations which
renders very probable a general occurrence of parasynapsis among
amphibians.

Jordan ('11) describes in the spermatogenesis of an opossum what
he considers evidence for telos;vTiapsis. His figures, however, are far
from convincing on this point, since they could as readily be inter-
preted in favor of parasynapsis as telosjaiapsis.

Wodsedalek ('13), in his studies of the spermatogenesis of the pig,
is unable to find conclusive evidence on the subject of synapsis. He
says (p. 13), however, that in the synezesis stage, "The thin tlireads
become arranged in a very much tangled mass of loops, which later
appear in about half the original number and twice as thick." Inas-
much as these phenomena accompany every case of demonstrated
parasjTiapsis, the evidence seems to favor the occurrence of this mode
of conjugation in this case.

In conclusion, I think it must be admitted that there is abundant
evidence for a widespread occurrence of parasynapsis, especially as
shown by the most recent investigations. While a majority of the
authors who have worked on orthopteran material have reported
telosynapsis, I believe there is some chance that many of them were
mistaken, or that a more careful analysis of the critical stages would
have given a different result. Whether we accept the hypothesis of
Stevens, that all degrees of s\Tiapsis occur, or the idea of Gregoire,
that parasynapsis is an almost universal phenomenon, we must at all
events admit that the most careful of the recent investigations indi-
cate that the latter condition is widespread throughout the animal
kingdom.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 105

As to which of the two maturation divisions is equational and which
is reductional, no absohite rule can be laid down. The evidence,
however, points to the probability that generally chromosomes with
terminal spindle-fiber attachment are not separated from each other
until the second division, while those that have a non-terminal at-
tachment are separated in the first, and that consequently in the
former the reduction occurs at the second division, in the latter at the
first division.

B. Individuality.

The theory of individuality was early championed by Van Beneden
('83), Rabl ('85), and Boveri ('88). In more recent years the theory
has been supported by many wi'iters, who have accepted as substan-
tial evidence in its favor the constancy in the number, size, and
shape of the chromosomes reappearing in the mitotic spindle of any
one species of animal or plant. On the other hand, some eminent
zoologists have attacked the theory on the ground that the individual
chromosomes cannot be traced through the so-called "rest" period
between mitoses. It will, therefore, be convenient to discuss the two
topics: — (a) constancy in metaphase chromosomes, and (b) persistent
organization of chromosomes.

a. Constajici/ of Metaphase Chromosomes.

1. Constancy in number. — The constancy in number of chromo-
somes for any species is among the most commonplace of cytological
observations. It will therefore be unnecessary to make any exten-
sive references to the literature. Some exceptions to the general
rule occur, however, and should receive attention. Supernumerary
chromosomes have been reported from time to time, and have been
studied especially by Stevens and Wilson. Wilson ('09) found in
Metapodius variations in chromosome-number from 21 to 26, though
the number for each individual animal was constant. The number of
chromosomes was dependent neither on sex, nor locality of habitat,
nor w^as it correlated with constant differences of size or of visible
structures in the adults. But the variation affects only particular
classes of chromosomes (the small idiochromosomes) and all exhibit
the same behavior. Furthermore, Wilson found a few cases of mitoses
in which both members of a pair of small chromosome were going to

106 bulletin: museum of comparative zoology.

the same pole. Using this as a basis, he was able to find satisfactory
explanation of the variation in number, and one which served to sup-
port the theory of the individuality of the chromosomes. Stevens
('12b) found in Diabrotica supernumerary chromosomes varying in
number from 1 to 5, and believed that they had their origin in trans-
verse and longitudinal divisions of the X-chromosome, which normally
divides only longitudinally. These anomalies can therefore be
explained on the basis of some unusual method of distribution of the
chromosomes in mitosis; the fact that such extra cliromosomes persist
in all the cells of the animal in which they are found is a striking bit
of evidence in favor of the belief that they maintain their individuality.

Delia Valle ('09, '12) has attacked the theory of individuality,
declaring that the chromosomes are temporary and variable structures,
which form in the prophase and dissolve in the telophase. He thinks
their number is the quotient of the quantity of chromatin divided by
the average size of the clu-omosomes. The quantity of clu-omatin
is said to vary with conditions of nutrition, and the number of cln-omo-
somes with variations in external conditions. He made counts of
chromosomes from cells of the peritoneum of salamander larvae and
obtained numbers varying from 19 to 27. Montgomery ('10) points
to the following grounds for doubting the accuracy of Delia Valle's
conclusions : — " 1. The chromosomes counted are long, sinuous
ribbons, that overlap and interlace, the most difficult kind to count
with accuracy. 2. He included in the counts some cells in prophases,
where one cannot be certain that all the clu'omosomes have fully
separated. 3. The total number of the chromosomes is so large,
about 24, that the chance of error in enumeration is great. It is but
fair to conclude that while his technique was excellent, his choice of
material was bad, consequently a degree of scepticism might well be
maintained toward his results." Delia Valle in his latest paper ('12)
argues that the chromosomes are variable structures, because he has
been able to find transition stages between mitotic and amitotic
methods of cell-division in the erythrocytes of young salamanders.
It is a well-known fact that amitosis frequently accompanies degenera-
tion, and the figures of Delia Valle present strong indications of being
those of degenerating cells. It is precisely in degenerating cells that
one would look for inconstancy in the behavior of the chromosomes.

2. Constancy in size and shape. — It will be convenient to consider
the subjects of size and shape together. As to shape, we may dis-
tinguish spheres, rods, and V-shaped elements. Spheres are invariably
small and may be regarded as short rods. It will be convenient to

A\ enrich: spermatogenesis of PHRYNOTETTIX MAGNUS. 107

include under the term "V-shaped" all the chromosomes which have
a non-terminal spindle-fiber attachment. They may be regarded as
rods which have become bent at the point of the spindle-fiber attach-
ment. Broadly, therefore, we may look upon all cliromosomes as
rod-shaped, but it will make description easier to distinguish the
types just mentioned.

In attempting to show that chromosomes have a constant size and
shape for each species, as well as a constant number, it will be well to
call attention to the fact, so clearly stated by McClung ('14), that the
point of the spindle-fiber attachment is, as a general rule, constant
and therefore one of the indications of a persistent organization for
each individual chromosome.

Some groups of animals exhibit a high degree of uniformity in the
shape and size of the chromosomes in any species, as for example,
among the Crustacea and the Amphibia, while others show a great
variety of forms (Orthoptera, mammals). In the groups with diverse
shapes and sizes of chromosomes, the striking fact was pointed out by
Montgomery ('01) that there are two of each different size. Mont-
gomery reached the logical conclusion that of the two equivalent
series existing in «ach cell, one had been derived from the maternal
and the other from the paternal ancestor.

That the same series of sizes and shapes reappears in each cell-
generation, is recorded by nearly every observer whose material is
favorable enough to admit of such comparisons. The work of Mc-
Clung ('00, '02, '04, '08b, '14), Sutton ('02, '03), Baumgartner ('04),
Nowlin ('08), Pinney ('08), Robertson ('08) on orthopteran material
has done much to establish this fundamental feature of individuality.
A very interesting series of observations on this point is that of Meek
('12a, '12b) on Stenobothrus. He describes the results of a series of
careful measurements of cliromosome-dimensions in diiTerent genera- "
tions of cells, and as a result of these observations becomes convinced
of the existence of persistent individuality. I may quote some of his
conclusions ('12a, p. 24, ff.): — " (1) In all metaphases the relative
positions of the chromosomes in the equatorial plate appear to be
arbitrary. (2) The rods composing all ordinary chromosomes are
cylindrical with rounded ends, and of an unifonn and constant diame-
ter, viz., 0.83 micra. In each species eight lengths have been found,
and these constitute members of a series in arithmetical progi'ession,
of which the difference between consecutive terms is equal to the
radius of the rod. The heterotropic chromosome does not belong to
this general series, for, although equal in length to the longest rod, its

108 bulletin: museum of comparative zoology.

diameter varies at different points and exceeds 0.83 micra. (3) The
rods are indivisible units, and, since each spermatogonial and second
spermatocyte chromosome is composed of two, and each primary
spermatocyte chromosome of four, their morphological identity is
metrically proved. (4) The eight rod-lengths are not the same in
any two species; the longest and 5 short chromosomes occur in all,
but identity is always established by the two remaining chromosome-
rods. (5) The complexes of a species and its variety appear to be
identical; differences if existing, are too small to be recognized. (6)
The somatic chromosomes are identical with those of the germ cells.
(7) The total volume of ordinary chromosomes is the same in sperma-
togonial and primary spermatocyte metaphases, whereas only half this
amount appears in that of the secondary spermatocyte."

For some zoologists, the fact that for any species the chromosomes
reappear in the different cell-generations possessing the same relations
as to number, shape, and size is merely an expression of the activities
of the cell and signifies nothing as to a persistent individuality of the
chromosomes. Fortunately, we are not dependent on this kind of
evidence alone, the results of studies of the chromosomes of hybrids,
for example, offering still stronger evidence of individuality, as shown
by the work of Moenkliaus on hybrids between Fundulus and Men-
idia, and that on echinoderm hybrids by Baltzer and by Tenent.
Moenkhaus ('04) could recognize the two sets of chromosomes arising
from the pronuclei of the diverse parents by characteristic differences
in size. For the first two or three cleavages the two groups tended to
remain distinct, but in later cleavages the chromosomes of the two
kinds become more and more intermingled, though they are still
recognizable by their characteristic sizes. The value of this evidence
is obvious, and on this point Moenkhaus says (p. 53) : — "As long as
the two kinds remain grouped, as during the first two divisions, this
fact has little added significance (i. c, that two groups of distinctly
different kinds of chromosomes arise), since within each group it
would be perfectly possible for the component chromosomes to ex-
change chromatin granules during the resting period. If, however,
as occurs in the later cleavages, the two kinds of chromosomes become
mingled, the chromatin granules of both kinds must lie mingled to-
gether within the resting nucleus. If from such a nucleus the two
kinds of chromosomes again emerge, it amounts almost to a demon-
stration that the chromatin substance of a given chromosome forms a
unit and that unit persists." Baltzer ('09) was able to recognize in
hybrids between Echinus and Strongylocentrotus certain chromo-

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 109

somes which were distinctive of the species from which they were
derived. Tennent ('08) found in hybrids between Moira and Arbacia
a mixture of two kinds of chromosomes, each variety of which could
be distinguished. It is, indeed, difficult to understand how these
distinctive chromosomes could recur with such definite characteristics
in hybrid embryos, if there is no persistent identity for them.

Variations from the general rule of chromosomal constancy have
been recorded from time to time, for example, in the shapes of tetrad
chromosomes. In many species there is a tendency for each of the
forms of the tetrads to be reproduced in the first spermatocyte meta-
phase. This is particularly true where the chromosomes are all of a
similar shape and size. But even in such cases, there is a variation
in the exact contour presented by different chromosomes. It has
been made apparent by many investigators, especially by McClung
and his students, that the shape of a metaphase tetrad is dependent
upon the extent and character of the movement on each other of the
constituent cliromatids. The work of these authors also shows that
homologous chromosomes tend to assume about the same shape in all
the cells at corresponding stages of mitosis, but that this condition of
similarity has its exceptions. Baumgartner ('04) called attention to
the constancy in the number of rings formed among the tetrads of
Gryllus, and others have noted similar conditions. However, such a
criterion for individuality is not always a safe guide, as was pointed
out by Foot and Strobell ('05). Commenting on Baumgartner's
paper, they say, in regard to chromosomes in Allolobophora foctida: —
" We find no constant form differences of the chromosomes, the simplest
form of the bivalent chromosomes is two rods«attached end to end, and
these present a variety of shapes, rings, figures 8, crosses, etc., without
any regularity or constancy. The free ends of the bivalent chromo-
somes show a tendency to unite into a ring and in some cases nearly
all the eleven chromosomes are rings, and sometimes not a single ring
is formed" (footnote, p. 222). A glance at figures 39 and 40 (Plate 4)
of this paper will also show a variability in shape of the eleven bivalent
chromosomes. In my account of tetrad A, I have shown that this
element may or may not form rings, so that this character could not
be used as a criterion for identification in the earlier postspireme
stages. But in spite of these exceptions, there does exist in many
cases a strong tendency for a chromosome to assume the same shape
at similar stages in all the cells of an animal, and the exceptions have
no significance in relation to the question of a variation in the funda-
mental organization of the tetrads.

110 bulletin: museum of comparative zoology.

In the case of unequal tetrads, however, variation in shape does
have some meaning with reference to chromosomal organization. In
the specimens of Phrynotettix which I have studied, the shape of
tetrad B in two individuals is fundamentally different from that in
the other eleven, because, in the latter, a definite part of one member
of the pair is lacking. Similarly, in the case of Ci and C2, the difference
concerns a definite part of the members of the pairs. But the im-
portant thing to be kept in mind is that the organization of each of
these tetrads is constant for any individual animal, and such differ-
ences as exist between individuals can be readily accounted for.

b. Persistent Organization of Chromosomes.

1. The selected chromosomes. — One of the most important con-
clusions arrived at in the present study relates to the constancy in
the finer organization of the chromosomes, both from stage to stage
in the same individual, and from one individual to another. This
is shown in two ways : — first, by the existence in chromosome-pair
B of an architecture that is constant both for any one individual in
the various stages in which any architectural condition could be
recognized, and likewise for all the individuals studied; secondly, by
that of a particular pair of chromosomes (A) recognizable through all
the stages from spermatogonia to spermatids, the recognition being
made possible by the fact that the chromosomes in question possess
properties which are characteristic and constant for all stages.

Both of these selected ^chromosomes, A and B, tend to stain more
deeply than the other autosomes, but this tendency is much more
marked in A than in B. If chromosomes possessing similar peculiari-
ties be found in related species, may they not be regarded as homolo-
gous to the selected chromosomes A and B of Phrynotettix? I think
such homologies could be established. Miss Carothers ('13) shows
that the small unequal tetrad in Brachystola is usually associated
with the accessory chromosome, and is more intensely stained than
the other autosomes. Might it not be possible to analyse this un-
equal element in Brachystola and determine its relation to the unequal
tetrads of Phrynotettix? Since these two genera are closely related,
I believe this would be possible. Furthermore, the other unequal
tetrads described by Miss Carothers for Arphia and Dissosteira were
among the small chromosomes and, on account of the similarity in
behavior, might be found homologous to 5 or C of Phrynotettix.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. Ill

Miss Nowlin ('08) describes for Melanoplus bivittatus a precocious
tetrad (no. 11), which always appears in the metaphase as a rod
extended parallel to the spindle-axis. Such is also the behavior and
form of chromosome A. Furthermore, I have examined slides of
Melanoplus material and find that it also has a spireme loop that stains
more deeply than the others. May not this precocious tetrad of
Melanoplus be related to chromosome A of Phrynotettix?

Early in the course of my investigation I had the opportunity of
looking over some of Dr. McClung's collection of slides of acridian
material, and, though a thorough study was not made, I could easily
recognize in the pachytene stage of a number of species a spireme loop
which stained more deeply than the others. Such loops were found,
for example, in species of Aeoloplus, Amphitornis, Arphia, Brachy-
stola, Hadrotettix, Hesperotettix, Hippiscus, Melanoplus, Phaetaliotes,
and Stenobothrus. One characteristic of such tlireads, which, how-
ever, is not so marked in Phrynotettix, is a tendency to become asso-
ciated with the accessory chromosome. This is particularly true of
Melanoplus and Stenobothrus, the forms in which Davis ('08) was led
by this close association to describe a "double monosome." There
can be no question, I think, that these "double monosomes" were
merely the accessory chromosome plus one of these deeply stained
spireme segments. In view of these facts, the suggestion offers itself,
that similarity in the properties and behavior of certain chromosomes
in different species may be correlated with their taxonomic relation-
ships. Such correlation was, indeed, seen and discussed some time
ago by McClung ('08a). Meek ('12) has already made a comparative
study of the sizes of the chromosomes in several species of Steno-
bothrus, and has reached the conclusion that the five smaller pairs
of chromosomes are of the same size in all species, but that the large
(V-shaped) pairs differ from one species to another. It still remains
to be seen whether or not the chromosomes of different species can be
compared on the basis of their details of organization and behavior,
as well as size.

2. The heterochromosomes. — I believe most observers agree that
the heterochromosomes maintain their individuality tlii'ough the
growth-stages of the male germ-cell cycle. On another page (p. 87)
I have called attention to the similarity in behavior between the
autosomes and the accessory, this has also been noted by many others,
so that there is no very good ground for setting up a claim to funda-
mental distinction between the two kinds. It seems to me, therefore,
that if we admit a persistent individuality for the heterochromosomes.

112 bulletin: museum of comparative zoology.

we must at the same time admit a high degree of individuahty for the
autosomes.

3. PJasmosovies and nucleoli. — One of the most puzzHng problems
that cytologists have to deal with is the behavior and function of the
so-called ' plasmosomes ' or 'nucleoli.' They apparently exhibit such
a variety of reactions to methods of technique, and exhibit such
varying relationships to other structures in the cell, that it is almost
hopeless even to attempt to classify them. That they play some
important role in the physiology of the cell, there is not the slightest
doubt, but what that role is, or what relation they bear to the question
of chromosome-individuality, are problems that are far from a solu-
tion at the present time.

In my description of the tetrads A and B, I called attention to a
peculiar modification of one of the terminal granules of each. I
emphasized the fact that, in the case of tetrad A, this modified granule
furnished a means of identification for this element. Just what the
nature of this modification is, I cannot state definitely, but in the
pachytene stage it has the appearance of an expansion of a previously
condensed granule, and I have so treated it in my description. The
similar condition in B appears to arise in the same way, but in this
case there seems to be a more definite boundary to the modified gran-
ule, which thus resembles the plasmosomes, or "vesicles," described
by Carothers ('13). The expanded granule of A is usually not homo-
geneous, some areas within it appearing more dense than others. This
condition probably foreshadows that seen in the postspireme stages,
where it appears more like a collection of small granules, typically
tlu-ee in number. Miss Carothers described the 'vesicles' that she
found as being attached to spireme threads, and in some cases to
specific threads. Furthermore, she found that the occurrence of the
vesicles extended to several species, and, in some species, through
several generations of cells. I am indebted to Dr. McClung and to
Miss Carothers for the privilege of looking over some of the material
studied by the latter, as well as for the opportunity of studying slides
of other species ; I can confirm Miss Carothers's observations, and can
add that these so-called ' vesicles ' are present in nearly every species
of grasshopper that I have studied with this object in view.

I believe that the modified granules in Phrynotettix can be homol-
ogised with the 'plasmosomes' of other species. I would especially
call attention to the fact that these structures are always attached to
chromosomes, and that, in Phrynotettix, at least, they always involve
a certain part of the chromosome to which they are attached. I

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 113

believe we may therefore regard them as being related to the organiza-
tion of the chromosomes, just as much as the polar granules are. A
glance at the literature will show how constantly these structures are
found; but no one, except Miss Carothers, so far as I am aware, has
suggested that they are always attached to some definite region of a
chromosome. It would seem to be worth while for some one to make
a study of these structures from this point of view.

The changes in staining capacity which the plasmosomes undergo
at different stages raises an interesting question as to what may be
their relation to the chromatin of the thread with which they are
associated. Do they take up chromatin from the chromatin-thread,
thus increasing their own stainability, and give it back again as they
lose their power to hold the stain? Do they elaborate chromatin
from raw materials in the surrounding cell substance and give it up to
the chromatin-thread? Is their chromatic substance different from
other chromatin? Or, do they have some other way of becoming for
a time chromatic and later non-chromatic? May it not be possible
to answer some of these questions by carefully resolving into its
chromomeres the chromatin-thread with which they are associated,
and comparing the constitution at different stages? I believe this
could be done on favorable material. In Phrynotettix, these struc-
tures are definitely related to polar granules. Are the polar granules
to be classed in the same category as the plasmosomes? Is it possible
for a polar granule to become transformed into a plasmosome, and then
back into a polar granule again? The last question seems to be an-
swered in the affirmative by the conditions in Phrynotettix. In the
case of B, for example, one of the proximal granules becomes "ex-
panded" in only about 16% of the cases counted. In becoming
expanded it has become like a plasmosome. When it is not expanded,
it remains a polar granule. Is it any wonder, then, that the plasmo-
somes have been called 'variable' and 'uncertain' elements of the
nucleus?

Plasmosomes are associated with heterochromosomes, as well as
with autosomes. Davis ('08), for example, noticed one on the mono-
some of Stenobothrus, and I have confirmed the observation from
slides of my own. Morse ('09) found in cockroaches a plasmosome
constantly associated with the "chromatin nucleolus" (accessory),
and in addition another body in the cytoplasm, which he called a
plasmosome. A similar cytoplasmic body, which stains like chroma-
tin, is found in a number of Acrididae. Dederer ('07) found a plasmo-
some associated with the pair of idiochromosomes in Philosamia, and

114 bulletin: museum of comparative zoology.

Blackman ('05) found a plasmosome attached to the accessory chro-
mosome of Scolopendra. A long Hst might be added to show that
plasmosomes have been found associated with particular chromosomes.
Many attempts, not altogether successful, have been made to explain
the baffling relations to the other cell-structures of such bodies as have
been called plasmosomes, nucleoli, chromoplasts, karyospheres, etc.,
but any future attempts to elucidate these relations must, I believe,
be accompanied by a recognition of the relations that these structures
bear to the organization of individual chromosomes.

4. Persistence of chromosomes between mitoses. It still remains to
discuss what may be the nature of the "organization" of the chromo-
somes in the stages through which the nuclear substance passes from
one metakinesis to the next. I shall consider briefly (1) the origin of
the nucleus from the chromosomes, and (2) theories of continuity.

(1) Origin of the nucleus. In my description of the spermato-
gonial divisions of Phr\Tiotettix (p. 87-91), I pointed out that each
chi-omosome becomes surrounded, as early as the anaphase, by a
hyaline region, that this region expands in the telophase; that the
chromatin of each chromosome becomes diffused to a certain extent
within its own region; that a membrane becomes formed at the
boundary between the hyaline region and the cytoplasm, producing
the chromosomic "vesicle"; and that the nuclear membrane consists
of the outer walls of the vesicles at the periphery of the nuclear group.
I drew the conclusion that the hyaline region was formed at the ex-
pense of the cytoplasm and that the material of each chromosome
tended to remain within the space of its own vesicle, a core of chroma-
tin being particularly noticeable in the center of this region, and that
the prophase chromosome subsequently formed was developed out of
the substance of one, and only one, of the previously existing telophase
chromosomes. Sutton ('00) was the first to describe the vesicles of
the spermatogonia of a grasshopper. Since then, Otte ('07) has seen
similar structures in Locusta, and Davis ('08) in several Acrididae;
Pinney ('08) has described them for Plirynotettix. Sutton stated
that in the earlier stages of nuclear formation, each chromosome
produced a separate vesicle, just as I have found for Plirynotettix,
but that in later stages, the proximal ends fused together, giving a
common nuclear cavity, from which the distal ends of the vesicles,
particularly the longer ones, projected out like the fingers of a glove.
Sutton interpreted these conditions as lending strong support to the
theory of individuality. Otte believed that the individual vesicles
remain distinct throughout the whole of the interkinetic phases, and

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 115

Pinney reached a similar conclusion. Davis, on the other hand,
could recognize only an irregular outline for the nucleus, and did not
identify the vesicle of even the monosome with certainty. Gerard
('09) saw the hyaline regions about the telophase chromosomes of
Stenobothrus, and stated that the nuclear membrane was formed in
connection with them. Since similar conditions have been reported
by so many observers, it would seem that these vesicular structures
are the result of normal processes and not, as claimed by Vejdovsky
('11-12), artifacts.

If we turn to accounts other than those on orthopteran spermato-
genesis, we find that the formation of chromosornic vesicles out of
individual chromosomes in the telophase is by no means of rare
occurrence. Van Beneden ('83) noted in his work on Ascaris that
each of the two chromosomes of the female pronucleus often formed
a separate 'half -nucleus.' Hacker ('95b) observed that the chromo-
somes of the early cleavages of Cyclops brevicornis formed two groups
of "Blaschen," one group from the maternal and another from the
paternal pronuclei. Conklin ('02) calls attention to the occurrence
of such chromosomic vesicles, and gives the history of the nuclear
changes during the cycle of division in Crepidula as follows (p. 45) : —
"(1) The cliromosomes, consisting of chromatin enclosed in a linin
sheath, divide and move to the poles of the spindle, where they par-
tially surround the spheres. (2) Here they become vesicular, the
interior of the vesicle becoming achromatic, though frequently con-
taining a nucleolus-like body, while the wall remains chromatic. (3)
These vesicles continue to enlarge and then unite into the "resting
nucleus." The nuclear membrane is composed of the outermost walls
of the vesicles, while the inner walls stretch through the nucleus as
achromatic partitions. The chromosomal vesicles for the egg and
sperm nuclei remain distinct longer than those from the same nucleus
.... Such vesicles are found generally, if not universally, in the early
division of ova, though they are not usually found in other mitoses."
Small wood ('05) describes similar chromosomic vesicles in the eggs of
nudibranchs. He found that during the "rest-pause" between the
first and second maturation divisions the chromosomes frequently
have distinct vesicles. There may be a single vesicle for all the
chromosomes, or a single vesicle for each chromosome; all conditions
between these two extremes occur. Medes ('05) in her work on
Scutigera found in the second spermatocytes (p. 174) that: — "There
is no immediate formation of a nuclear membrane, but each separate
chromosome, as it disintegrates, becomes enclosed in a membrane of its

116 bulletin: museum of comparative zoology.

own, thus forming a structure similar to a nucleus but containing only
a single chromosome." Kornhauser ('15, p. 408) says concerning the
spermatogonia of Hersilia: — "The telophase chromosomes become
gradually fainter in outline, and a clear area in the cytoplasm begins
to form about them. It is, I believe, the boundary between this clear
area and the more reticular cytoplasm which forms the new nuclear
membrane." Thus it will be seen that it is quite usual for telophase
chromosomes to form individual nuclei, which later fuse to form the
whole nucleus, and with Smalhvood ('05) we may accept this tendency
as an argument for chromosomal individuality.

(2) Theories of continuity. Among those who support a theory of
continuity, there is not always agreement as to what structures are
carried from one cell-generation to the next. It is generally agreed
that the chromosomes are composed of at least two substances; the
chromatin and the ground substance (linin, plastin). Hacker ('04)
formulated the " Successionshypothese," stating that the persisting
structures of the chromosomes consisted of the "Grundsubstanz,"
or achromatic part. Bonnevie ('08a) and others, on the contrary,
regard the chromatic substance as the persistent portion and the
achromatin as the temporary part of the clu'omosome.

Vejdovsky ('07, '11-12) has evolved a most elaborate theory touch-
ing this problem. In his monograph of 1907, he based his conclusions
on a study of the ovogenesis and maturation of some annelids. He
concludes that the nucleus is derived from the chromosomes and from
them alone. He divides the interkinetic stages into two periods; the
one during which the nucleus is formed out of the chromosomes he
calls " katachromasis," and the one during which the chromosomes are
formed out of the nucleus he calls "anachromasis." In his later
monograph ('11-12) he analyses these processes still further and
attempts to describe in detail the events in the two periods. His
conclusions may be briefly stated as follows : — A chromosome is com-
posed of two substances, one a less deeply staining substratum, on
the surface of which is the other, the more deeply stainable chromatin.
In the early stages of katachromasis, the chromatin differentiates into
a spiral thread, or "chromonema," which is coiled about the surface
of the substratum. The substratum then dissolves, forming the
nuclear sap, or " enchylema." The chromonema further differentiates
into a finely coiled chromatic portion, inside of which is a linin core.
In this condition, he recognizes the anlage of the chromosome of the
succeeding generation. The linin substance of the chromonema is to
become the substratum of the future chromosome, and the finely

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 117

coiled chromatic portion will become its chromonema. In successive
generations, therefore, there is a changing composition of the chromo-
somes. During each katachromasis, the ground substance of the
chromosome dissolves, leaving the chromonema, which becomes
differentiated into the two kinds of substance found in the chromo-
some of the next generation. This is an ingenious theory, to say the
least, and carries with it some measure of support for the theory of
individuality, inasmuch as each new chromosome is formed out of the
substance of a preceding one.

I have found nothing in my studies to support any one of these
theories to the exclusion of the others. It is rather surprising, how-
ever, that Vejdovsky found no indication of the chromosomic vesicles
in the spermatogonia of the Orthoptera that he studied and, that he
regards those seen by others as artifacts. I find little evidence of a
chromonema in the telophase of the spermatogonia, and what evidence
there is would indicate that the chromatin becomes distributed on the
inner surface of the vesicular walls, not on the outer surface of an
achromatic core. In the telophase of the last spermatogonia, I find
a spiral thread forming, but it develops out of the chromatin at the
middle of the area occupied by a vesicle. But whether we accept
any one of these theories, or reject all of them, there still remain the
strongest grounds for believing, as they all indicate, that there is
some underlying organization which is in some way perpetuated
for each individual chromosome. I am inclined to the belief that
this organization involves both chromatic and achromatic substance.

In plant material evidence which indicates a continuity of the
chromosomes has not been wanting. Gregoire ('07, '10) believes
that the results of his own investigations and those of others on plants
furnish strong support for the individuality theory. Stout ('12) has
recently added evidence for this belief in his work on Carex aquatilis.
He says ('12, p. 36): — "The chromosomes are present in all resting
nuclei as visible units of a definite number. These individual chromor
somes can be traced as such through all stages of both somatic and
germ-cell divisions, with the exception of the various stages of synapsis
(synizesis)." Lee ('13) also finds continuity of the chromosomes
in plants through the " rest-stage." He beheves that the chromosomes
of even the metaphase become vacuolated, that this vacuolization
increases in the telophase, where, later, a spiral thread is formed out
of each cliromosome. This spiral thread becomes the prophase
chromosome of the succeeding division. He introduces the term
"spirophase" to designate the so-called "rest-stage."

118 bulletin: museum of comparative zoology.

There* seems to be a great amount of disagreement as to just what
constitutes individuahty, but I beheve that we may class as instances
of individuahty all cases where it can be shown that the substance of
any telophase chromosome gives rise to one and only one prophase
chromosome. In that event, any one of the three theories mentioned
above would support the theory of individuality. I believe that I
have demonstrated individuality for chromosome-pair A, and have
shown good evidence for it among the other chromosomes of Phryno-
tettix. Besides, it seems to me much more logical to regard the con-
stant reappearance of the same architectural conditions of a given
chromosome as a result of continuity of that architecture in some
form or other through all the cell-divisions, than to assume that the
organization is entirely destroyed and reestablished between successive
mitoses.

C. Chromosomes and Heredity.

Any discussion of the relation of the chromosomes to heredity must
deal to a considerable extent with theory and speculation. Yet there
are many facts which tend to the belief that the chromosomes are,
after all, directly concerned with the transmission of hereditary
qualities. A few facts and some theory will be considered in the
following paragraphs under the two heads: — (a) Mendelism and
maturation, and (b) some experimental evidence.

a. Mendelism and Maturation.

Wilhelm Roux was apparently the first to formulate, in the early
eighties, a theory in which an attempt was made to localize the
carriers of hereditary qualities in the chromosomes; this was later
elaborated by others, especially by Weismann, who postulated a
reduction division which has since been identified with one or the
other of the maturation divisions. Montgomery ('01) pointed out
that the chromosomes of the diploid series occur in pairs, the members
of each pair being of the same shape and size. There are thus two
similar series of chromosomes. He concluded that one series was
derived from the maternal, the other from the paternal ancestor. He
concluded further that the members of each pair unite to form the
bivalent chromosomes of the first spermatocytes. Boveri ('02) de-
cided from the results of his experiments on dispermic echinoderm

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 119

eggs, that the chromosomes were quahtatively different. Sutton
('03) in the following year, explained how the behavior of the chromo-
somes in matm*ation could be correlated with the behavior of Men-
delian characters. He showed: — (1) that the union of clu-omosomes
of diverse origin into pairs and their subsequent separation in one of
the maturation divisions would insure to every gamete one of every
kind of chromosome in the series: (2) that if the law of chance were
operative in the orientation of the pairs on the maturation spindles,
every possible combination of male and female chromosome could
result; and (3) that such a recombination according to the law of
chance would account for the transmission of Mendelian characters,
if the chromosomes retained their individuality and really were the
carriers of the qualities.

This work of Sutton has been generally accepted as proving the
correlation assumed, but it remained for Carothers ('13) to demon-
strate that the law of chance actually does operate in the distribution
of the chromosomes in the maturation spindles. In the case of the
unequal tetrads described by her, it was shown that either the large or
small member of the pair may go to the same pole as the accessory
chromosome, which, as usual, was found to go to one pole undivided.
Moreover, it was found that the ratio between the two results of
distribution was approximately one to one. Robertson ('15) has very
recently published some of his work on the Tettigidae, where he has
found the same rule to hold for the unequal pairs that were present
in his material. The behavior of tetrad Ci in Phrynotettix agrees
with that described by Carothers and Robertson. These cases
establish the fact that there really is a distribution of chi'omosomes
in the maturation divisions according to the law of chance.

A further consideration of the cases of unequal tetrads in Orthop-
tera will show in how far the theoretical possibilities as to chance
distribution have been realized. Baumgartner ('11) in reporting his
results on Gryllotalpa borealis before the American Society of Zoolo-
gists, stated that he found in the first maturation mitosis an unequal
pair of chromosomes, of which the larger dyad always went to the
same pole as the accessory. Payne ('12) found the same conditions
in this species of Gryllotalpa. He regards the large member as
possibly associated with the accessory to form a sex-group, similar to
the groups in Conorhinus and Fitchia (Payne, '09), or in Thyanta
(Wilson, '10), with the exception that in Gryllotalpa the grouping
occurs in the first spermatocyte metaphase instead of the second.
Payne suggests that the chromosomes instead of following a haphazard

120 bulletin: museum of comparative zoology.

method of distribution in the maturation divisions, may always move
the same way, i. e., all the chromosomes brought into the egg may
pass into the female-producing sperm. It is extremely doubtful if
the last suggestion will prove applicable as a general rule, but the
conditions in Gryllotalpa are interesting exceptions to what has been
found in the Acrididae and Tettigidae.^

Hartmann ('13) describes small chromosomes as dividing unequally
in some male germ-cells of Schistocerca. In one first-spermatocyte
cell he found two such chromosomes (tetrads) dividing unequally,
and he found some cases of unequal division in the secondary sperma-
tocytes. These observations, if correct, would lead one to suspect
that he might" have been dealing with a condition similar to that in
Phrynotettix, except that in the first division, either both the small
chromosomes divided sometimes reductionally and sometimes equa-
tionally, or, while one of them followed this method, the other always
divided reductionally.

Bringing together the results of Baumgartner and Payne for Gryllo-
talpa, those of Carothers for Acrididae, Robertson for Tettigidae, and
my own for Plirynotettix, we may arrange a graded series of condi-
tions beginning with (1) tetrad B, in Phrynotettix, which is unequal,
but divides equationally in the first division; passing (2) to Ci, which
divides with equal frequency either reductionally or equationally in
the first division, and when dividing reductionally shows chance
distribution with reference to the accessory; thence (3) to the unequal
types found by Carothers and Robertson, which always divide re-
ductionally in the first division but show chance distribution, and
finally (4) to Gryllotalpa, where division is always unequal in the first
spermatocytes, but the larger dyad always accompanies the accessory.
Whether this series offers any possible explanation as to the origin of
these unequal elements, and their different kinds of behavior, is prob-
lematical.

Robertson's work deserves further consideration, because he has
found two of the tkree possible combinations which would be expected
out of a random recombination of two unequal elements which con-
jugate. In the case of Tettigidea, he found the unequal tetrad in

' i^oslscrij)!. — Unfortunately Jif had [overlooked the results reported for Gryllotalpa vul-
garis by Voinov ('14), who found in the first spermatocyte metaphase an unequal pair of dyads,
which separate so that sometimes the larger dyad and sometimes the smaller one goes to the
same pole as the accessory chromosome. These results are in accord with those mentioned
above for the Acrididae and the Tettigidae and it may be surmised that similar conditions
perhaps obtain for Gryllotalpa borealis but have so far been overlooked.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 121

only two individuals, all the others showing a pair both members of
which were equal to the larger member of the unequal pair. The third
possibility, an equal pair, homologous to the smaller of the dyads,
was not found. This case is analogous to that of B in Phiynotettix.^
In Acridium, Robertson found two individuals, one a male, the other
a female, possessing an unequal pair of chromosomes, whereas all the
other individuals studied showed the homologous pair to be equal,
both members being equivalent in size to the smaller of the two
members of the unequal pair. This case is analogous to those of
tetrads Ci and C2 in Phrynotettix, where, also, only two combinations,
the same two, out of a possible three have been found.

Robertson calls attention to the obvious possibility of a loss of
chromatin from the unequal pair in Tettigidea, and suggests that the
loss of Mendelian factors could be accounted for in this way. He also
suggests that the loss of the distal ends of both the chromosomes,
resulting in a pair of small dyads each equivalent to the smaller mem-
ber of the unequal pair, might result in lethal conditions, or might
mean the loss of factors necessary for development. In the case of
the unequal pair in Acridium, he assumes that there has been an addi-
tion to one member of the smaller pair. If this element is similar to Ci
of Plu*ynotettix, as it seems to be, then the simpler explanation would
be that a part had been lost, just as in the case of the one in Tettigidea.
It is curious that in both Ci and the unequal pair in Acridium, the
same combination, i. e., a pair both members of which would be equal
to the larger member of the unequal pair, is lacking. I am inclined
to believe, if sufficient material were available, that the remaining
possible combinations would be found. The matter could, at least,
be tested by experiment. It is the hope of the writer to conduct
breeding experiments with this object in view.

One further point remains to be considered in relation to chromo-
some-pair C. I have described these tetrads in detail elsewhere
(p. 85), but a reference to figure 107 (Plate 9) will recall that there
are three types, which I have designated as Ci, C2, and C3. If similar
types exist in the female, — Robertson ('15) found an unequal pair in
a female of Acridium, — and random mating be assumed for the ani-
mals possessing the three different types, then one ought to obtain in

* Since writing this I have had an opportunity to examine slides frcn some new Phrynotettix
material collected during the summer of 1915 by Miss Garothers of the University of Pennsyl-
vania. In some of the individuals of the new material I have found the expected third type of
chromosome-pair B composed of two elements both equivalent to the shorter member of the
unequal type.

122 bulletin: museum of comparative zoology.

the offspring all possible combinations of the three kinds of chromo-
somes. Or, if mating involving any two of the types could be made,
there should result all possible combinations between them. On
account of the small number of animals available for my study, no
conclusions as to whether these conditions are realized in nature could
be drawn. The presence of the three types in these few animals,
however, strongly suggests the possibility of realization, especially
since two of the three possible combinations are realized for the two
kinds of chromosome in type Ci. The presence of a third type also
suggests that there may exist in this case the mechanism for the
transmission of triple allelomorphs.

b. Some experimental Evidence.

The most extensive breeding experiments the results of which tend
to show that the chromosomes are concerned in the transmission of
hereditary characters are those on Drosophila by Prof. T. H. Morgan
and his students. In the course of this work they have dealt with
over a hundred unit-characters which show Mendelian inheritance,
either in a typical or modified form. In Drosophila, there are four
pairs of chromosomes, of which one pair is very small, and one is a
pair of heterochromosomes, or " sex-chromosomes." In their behavior
in inheritance, the hundred and more characters fall into four groups,
each group tending to behave as a unit, just as it would be expected
to do in case it were carried by a single pair of chromosomes. Of
these groups of characters, one is very small, the others much larger,
the largest one being the group of " sex-linked " characters. Naturally
the small group of characters has been correlated with the small pair
of chromosomes and the group of sex-linked characters with the sex-
chromosomes.

But there have been exceptions in the case of many pairs of allelo-
morphs, especially those that are sex-linked, i. e., cases where factors
belonging to a certain group have gone into a mating together, but
have not always reappeared together, as they would be expected to do
if they were all carried by a single chromosome and that cliromosome
maintained its individuality. These phenomena have been explained
by the so-called "cross-over" hypothesis. In this connection Morgan
('11) developed what has been termed the "linear arrangement"
hypothesis, which was further elaborated by Sturtevant ('13). These
authors assume that the factors, or "genes," which represent the

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 123

characters, are distributed in a linear series along the length of the
chromosomes. Then, invoking the aid of the " chiasmatype " theory
of Jannsens ('09), they attempt to explain the "cross-overs" by as-
suming, first, that when two chromosomes conjugate side by side,
they may become twisted around each other, and, secondly, that the
later separation is along a plane, which cuts across the threads once
for every complete twist. Considering the matter in relation to the
tetrad stages, it might be imagined that the two tlireads cross each
other, and that at the point of crossing, a weakness of the strands
causes them to break and then recombine, forming threads each of
which is composed of a part of both the original conjugants.

Judging from his figures, Janssens founded his theory on conditions
similar to those shown in my figure 38, a-d (Plate 3). I am quite
sure that the evidence in Phrynotettix does not support the idea that
the chromatids break and recombine in any of the postspireme stages.
On the contrary, I believe that the chromatids maintain as strict
an individuality as I have claimed for the chromosomes themselves.
And since these tetrad figures are repeated in so many animals, and
even in plants, there would seem to be ground for supposing the be-
havior to be similar in all.

On the basis of this hypothesis, however, Morgan and his pupils
have been able to explain the anomalous behavior of the genes which
they call the "cross-over" in a very satisfactory way; furthermore,
they have been able to use it in connection with the linear-arrangement
hypothesis to predict the behavior of any given character, with refer-
ence to any other character in the group to which it belongs, provided
its behavior in relation to one or two of the characters of the group is
known. But there is one point yet to be noted. I have based my
criticism of the chiasmatype theory on the conditions as found in
spermatogenesis. One of the peculiar facts found in the work on
Drosophila is that there is no "crossing-over" in the male. But why
should such a phenomenon occur in the female and not in the male?
Is it not possible that in the "great growth" period of the oocyte, —
where the tetrads become so much more expanded and diffused than
in the male, even seeming to disappear entirely in some cases, — the
tetrads might suffer some such changes as those suggested by the
experimental results? There is also to be considered the often re-
peated condition of parasynapsis in Drosophila, as shown by Metz
(*14), which might offer greater opportunities for such "cross-overs"
to occur than are found in other animals.

Whatever else may be said of the results of the experiments on

124 bulletin: museum of comparative zoology.

Drosophila, it must be admitted that they go very far towards estab-
lishing a direct relationship between the elii'omosomes and the trans-
mission of Mendelian characters. Perhaps the most convincing evi-
dence of this kind is that obtained b}^ Bridges. He has found that in
certain strains involving sex-linked inheritance, some exceptional
females appeared which were like their mothers in every respect, and
showed no transmission of sex-linked characters from the father,
although such transmission would be expected, since the male sex
formula is XY and that of the female is XX. Furthermore, he found
that such exceptionally produced females inherit directly from their
mother the power of producing like exceptions (about 5%). The
explanation advanced by Bridges ('14) was that "the sex-linked genes
were borne by the X-chromosomes and that 10% of the eggs of the
exceptional females retained both of the Z^-chromosomes, or conversely
lost both to the polar body." This phenomenon was called "non-
disjunction." Breeding experiments showed that an X-chromosome
gene could not be the cause of the phenomenon, and the prediction
was accordingly made that half the daughters of a non-disjunctional
female would be found to contain in addition to the two X-chromo-
somes a supernumerary chromosome which would be a Y. Cytologi-
cal investigations have shown that approximately one-half of the
daughters of a non-disjunctional female do, in fact, contain a super-
numerary F-chromosome, while the remaining half contain only the
two X'-chromosomes. I may add that thi'ough the kindness of Dr.
Bridges, I have been able to examine some of his slides and convince
myself of the presence of the extra chromosome. This brilliant piece
of work makes it very hard to disagree with Bridges's conclusion ('14,
p. 109) that, "there can be no doubt that the complete parallelism
between the unique behavior of the chromosomes and the behavior of
the sex-linked genes and sex in this case means that the sex-linked
genes are located in and borne by the sex-chromosomes."

Returning now to a consideration of the linear-arrangement hypo-
thesis, it must be admitted that the theory has attractive possibilities,
and up to the present time has stood the test of experimental breeding
in Drosophila. It may not be out of place, therefore, to call attention
in this connection to the constancy of the granular, or chromomeral
organization of the chromosomes of Phrynotettix, particularly in
chromosome-pair B. May not this constancy of architecture of the
chromosomes have a meaning correlated with that assumed in the
linear-arrangement hypothesis? This possibility seems to me to be
worthy of further investigation.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 125

D. Summary of Conclusions.

It is believed that the present study of the spermatogenesis of
Phrynotettix magnus has demonstrated : —

1. That conjugation of the chromosome-pairs is by parasynapsis.

2. That the majority of the bivalent chromosomes divide equa-
tionally in the first maturation division.

3. That the chromosomes retain their individuality through the
spermatogenic cell-generations.

4. That the so-called ' plasmosomes ' take their origin from some
definite region (granule) of particular chromosomes, but that they
may be variable in occurrence and in extent of development.

5. That in the maturation divisions {e. g. chromosome-pair Ci)
the law of chance is followed in the distribution of the chromosomes.

6. That each chromosome possesses a definite organization, which
is expressed in the constancy of the relative sizes and positions of its
chromomeres (as seen, e. g., in chromosome-pair B).

In addition, the possibilities are suggested that: — (1) the matter
of the behavior of unequal pairs of chromosomes in regard to distribu-
tion and recombination may be tested by breeding experiments, (2)
the constancy in the arrangement of clu-omomeres along the length
of the chromosome-tlu-eads, as described for chromosome-pair B,
may have a meaning related to that suggested by Morgan's " hnear-
arrangement" hypothesis, and (3) that in the varying types of chromo-
some-pair C there may exist a mechanism for the transmission of
multiple allelomorphs.

126 bulletin: museum of comparative zoology.

IV. BIBLIOGRAPHY.
Baltzer, F.

'09. Die chromosomen von Strongylocentrotus lividus und Echinus
microtuberculatus. Arch. f. zellf., 2, p. 549-632, taf. 37, 38, 25 text
fig.
Baumgartner, W. J.

'04. Some new evidence for the individuality of the chromosomes.

Biol, bull., 8, p. 1-23, pi. 1-3.
'11. Spermatogenesis in the mole crickets. [Abstract]. Science, new
ser., 33, p. 275.
Beneden, E. Van,

'83. Recherches sur la maturation de I'oeuf, et la fecondation. Ascaris
megalocephaJa. Arch. d. biol., 4, p. 265-640, pi. 10-19.
Blackmail, M. W.

'05. The spermatogenesis of the myriapods. III. The spermato-
genesis of Scolopendra heros. Bull. M. C. Z., 48, p. 1-138, 9 pis., 9
text fig.
Bonnevie, K.

'08^. Chromosomenstudien. I. Chromosomen von Ascaris, Allium und
Amphiuma, u. s. w. Arch. f. zellf., 1, p. 450-514, taf. 11-15, 2 text fig.
'08^. Chi-omosomenstudien. II. Heterotypische mitose als reifungs-
character. Nach untersuchungen an Nereis limbata Ehlers, Thalas-
sema mellita Conn, und Cerebratulus lacteus Hubr. Arch. f. zellf., 2,
p. 201-278, taf. 13-19, 23 text fig.
Boveri, T.

'88. Zellen-studien. Jena, zeitschi-., 22, p. 685-882, taf. 19-23.

Also Separate, Zellen-studien. Heft 2. Die befruchtung und teilung
des eies von Ascaris megalocephala. 198 p., 5 taf.
'02. tJber mehrpoUge mitosen als mittel zur analyse des zeUkerns.
Verh. phys.-med. gesell. Wurzburg, n. f., 35, p. 67-90.
Bridges, C. B.

'14. Direct proof through non-disjunction that the sex-linked genes of
Drosophila are borne by the X-chromosome. Science, new ser., 40,
p. 107-109.
Browne, Ethel N.

'13. A study of the male germ cells in Notonecta. Journ. exp. zool.,
14, p. 61-121, 10 pis.
Brunelli, G.

'09. La spermatogenesi del Gryllus desertus Pall. (Division! spermato-
goniaU e maturative). Mem. Accad. dei Lincei, ser. 5, 7, p. 623.-653,
2 tav.
'10. La spermatogenesi della Tryxalis, Parte I. Divisione spermato-
gonici. Mem. Soc. Ital. del sci., ser. 33, 16, p. 221-236.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 127

Buchner, P.

'09. Das accessorische chromosom in spermatogenese und ovogenese
der orthopteren, zugleich ein beitrag zur kenntnis der reduktion-
Arch, f . zellf ., 3, p. 335^30, taf. 16-21, 5 text fig.
Carothers, £. Eleanor.

'13. The Mendelian ratio in relation to certain orthopteran chromo-
somes. Jom-n. morph., 24, p. 487-511, 3 pis., 1 text diag.
Conklin, E. G.

'02. Karyokinesis and cytoldnesis in the maturation, fertilization and
cleavage of Crepidula and other Gasteropoda. Journ. Acad. nat.
sci. Phil., ser. 2, 12, p. 1-121, 6 pi., 33 text fig.
Davis, H. S.

'08. Spermatogenesis in Acrididae and Locustidae. Bull. M. C. Z., 53,
p. 57-158, 9 pis., 24 text fig.
Dederer, Pauline H.

'07. Spermatogenesis in Philosamia cynthia. Biol, bull., 13, p. 94—106,
42 text fig.
Delia Valle, P.

'09. L'organizzazione della cromatina studiata mediante il numero dei

cromosomi. Aixh. zool., 4, p. 1-77, tav. 1.
'12. La continuita delle forme di divisione nucleare ed il valore mor-
fologico dei cromosomi. Arch, zool., 5, p. 119-200, tav. 9, 10.
Eisen, G.

'00. The spermatogenesis of Batrachoseps. Journ. morph., 17, p. 1—
117, pi. 1-14.
Farmer, J. B,, and Moore, J. E. S.

'05. On the maiotic phase (reduction divisions) in animals and plants.
Quart, jom-n. micr. sci., 48, p. 489-557, pi. 34-41.
Fasten, N.

'14. Spermatogenesis of the American crayfish Cambarus virihs and
Cambarus immunis (?), with special reference to synapsis and the
chromatoid bodies. Journ. morph., 25, p. 587-649, 10 pis., 1 text fig.
Foot, K., and Strobell, E. C.

'05. Prophases and metaphase of the first maturation spindle of Allolo-
bophora foetida. Amer. journ. anat., 4, p. 199-243, pi. 1-9.
Gerard, P.

'09. Recherches sur la spermatogenese chez Stenobothrus biguttulus
(Linn.). Arch, biol., 24, p. 543-625, pi. 19-21, 11 text fig.
Granata, L.

'10. Le cinesi spermatogenetiche di Pamphagus marmoratus (Burm.).
Arch. f. zellf., 5, p. 182-214, taf. 14^16, 1 text fig.
GrSgoire, V.

'05. Les resultats acquis sur les cindses de maturation dans les deux

regnes. Mem. I. La cellule, 22, p. 221-376, 147 text fig.
'07. La formation des gemini heterotypiques dans les v6g6taux. La
cellule, 24, p. 367^20, 2 pi., 4 text fig.

128 bulletin: museum of comparative zoology.

Gregoire, V.

'10. Les cineses de maturation dans les deux rSgnes. M6m. II. L'unit6

essentielle du processus m6iotique. La cellule, 26, p. 223-422, 145

text fig.
Hacker, V.

'92. Die eibildung bei Cyclops und Canthocamptus. Zool. jahrb.

Anat., 5, p. 211-248, taf. 19.
'95*. Die vorstadien der eireifung. Arch, f . mikr. anat., 45, p. 200-273,

taf. 14-17.
'95b. tjber die selbstandigkeit der vaterlichen und miitterlichen kem-

bestandtheile wahrend der embryonalentwicklung von Cyclops. Arch.

f. mikr. anat., 46, p. 579-618, taf. 28-30.
'04. Bastardirung und geschlechtszellenbUdung. Zool. jahrb. Suppl.,

7, p. 161-256, taf. 12, 13 text fig.
Haxtman, F. A.

'13. Variations in the size of chromosomes. Biol, bull., 24, p. 226-238,

4 pis.
Janssens, F. A.

'01. La spermatog^nese chez les tritons. La cellule, 19, p. 5-116, 3 pis.
'05. Evolution des auxocytes mdles du Batrachoseps attenuatus. La

cellule, 22, p. 379-425, 7 pis.
'09. La theorie de la chiasmatypie. NouveUe interpretation des cineses

de maturation. La cellule, 25, p. 389-^11, 2 pis., 27 text fig.
Jordan, H. E.

'08. The accessory chromosome in Aplopus mayeri. Anat. anz., 32,

p. 284-295, 48 text fig.
'11. The spermatogenesis of the opossum (Didelphys virginiana) with

special reference to the accessory chromosome and the chondriosomes.

Arch. f. zellf., 7, p. 41-86, pi. 1-3, 2 text fig.
Kornhauser, S. I.

'14. A comparative study of the chromosomes in the spermatogenesis

of Enchenopa binotata (Say) and Enchenopa (Campylenchia St&l)

curvata (Fabr.). Arch. f. zellf., 12, p. 241-298, pi. 18-22, 8 text

fig.
'15. A cjdiological study of the semi-parasitic copepod, Hersilia apodi-
formis (Phil.), with some general considerations of copepod chromo-
somes. Arch. f. zellf., 13, p. 399-445, pi. 27-29, 9 text fig.
Korschelt, E., und Heider, K.

'03. Lehrbuch der vergleichenden entwicklungsgeschichte der wirbel-
losen thiere. Allg. theil, lief. 2, aufl. 1 and 2, p. 539-750.
Lee, A. B.

'13. La structure des chromosomes et du noyau au repos chez Paris
quadrifoha. La ceUule, 28, p. 263-300, 2 pis.
Lerat, P.

'05. Les ph^nomenes de maturation dans l'ovog6nSse et la spermato-
g^nese du Cyclops strenuus. La ceUule, 22, p. 161-199, 4 pis.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 129

McClendon, J. F.

'10. Further studies on the gametogenesis of Pandarus sinuatus, Say.
Arch. f. zeUf., 5, p. 229-234, taf. 17, 5 text fig.
McClung, C. E.

'99. A peculiar nuclear element in male reproductive cells of insects.

Zool. bull., 2, p. 187-197, 13 text fig.
'00. The spermatocyte divisions of the Acrididae. Kans. univ. quarterly,

9; p. 73-100, pi. 15-17.
'02. Spermatocyte divisions of Locustidae. Kans. univ. sci. bull., 1,

p. 185-231, pi 7-10.
'05. The chromosome complex of orthopteran spermatocytes. Biol.

bull., 9, p. 304-340, 21 text fig.
'08a. Cytology and taxonomy. Kans. univ. sci. bull., 4, p 197-215.
'08l>. The spermatogenesis of Xiphidium fasciatum. Kans. univ. sci.

buU., 4, p. 253-262, pi. 15.
'14. A comparative study of the chromosomes in orthopteran spermato-
genesis. Journ. morph., 25, p. 651-749, 10 pis.
Matschek, H.

'09. Zur kenntnis der eireifung und eiablage bei den copepoden. Zool.
anz , 34, p. 42-54, 9 text fig.
Medes, Grace.

'05. The spermatogenesis of Scutigera forceps. Biol, bull., 9, p. 156-
178, pi. 2-6.
Meek, C. F. U.

'12 a. A metrical analysis of chromosome complexes, showing correla-
tion of evolutionary development and chromatin thread-width through-
out the animal kingdom. Philos. trans. Roy. soc. Lond., ser. B., 203,
p. 1-74, pi. 1-5.
'12b. The correlation of somatic characters and chromatin rod-lengths,
being a further study of chromosome dimensions. Journ. Liimean soc,
Lond., 32, Zoology, p. 107-119, 6 text fig.
Metz, C. W.

'14. Chromosome studies in the Diptera. I. A preliminary survey of
five different t>T)es of chromosome groups in the genus Drosophila.
Journ. exp. zool., 17, p. 45-59, 1 pi.
MoenkhauS; W. J.

'04. The development of the hybrids between Fundulus heterochtus and
Menidia notata with especial reference to the behavior of the maternal
and paternal chromatin. Amer. journ. anat., 3, p. 29-65, 4 pis.
Montgomery, T. H.

'01. A study of the chromosomes of the germ cells of Metazoa. Trans.

Amer. phH. soc, 20, p. 154-236, pi. 4-8.
'05. The spermatogenesis of Syrbula and Lycosa, with general considera-
tions upon chromosome reduction and the heterochromosomes. Proc.
Acad. nat. sci. Phil., 57, p. 162-205, pi. 9, 10.

130 bulletin: museum of comparative zoology.

Montgomery, H. T.

'06. The terminology of aberrant chromosomes and their behavior in

certain Hemiptera. Science, new ser., 23, p. 36-38.
'10. On the dimegalous sperm and the chromosomal variation of Eus-

chistus, with reference to chromosomal continuity. Arch. f. zellf.,

5, p. 120-145, taf. 9, 10, 1 text fig.
'11. The spermatogenesis of an hemipteron, Euschistus. Journ.

morph., 22, p. 731-817, 5 pis.
Moore, J. E. S.

'95. On the structural changes in the reproductive ceUs during the

spermatogenesis of elasmobranchs. Quart, journ. micr. sci., 38, p.

275-313, pi. 13-16, 4 text fig.
Morgan, T. H.

'11*. Random segregation versus coupling in Mendelian inheritance.

Science, new ser., 34, p. 384.
'lib. An attempt to analyze the constitution of the chromosomes on

the basis of sex-limited inheritance in Drosophila. Journ. exp. zool.,

11, p. 365-413, 1 pi.
Morse, M.

'09. The nuclear components of the sex cells of four species of cock-
roaches. Arch. f. zellf., 3, p. 482-520, taf. 26-28, 1 text fig.
Nowlin, Nadine.

'08. The chromosome complex of Melanoplus bivitattus Say. Kans.

univ. sci. bull., 4, p. 265-271, pi. 16, 17.
Otte, H.

'07. Samenreifung und samenbildung bei Locusta viridissima. Zool.

jahrb. Anat., 24, p. 431-520, taf. 35-37, 2 text fig.
Payne, F.

'09. Some new types of chromosome distribution and their relation to

sex. Biol, bull., 16, p. 119-166, 1 pi., 11 text fig.
'12. The chromosomes of Gryllotalpa borealis Burm. Arch. f. zellf., 9,

p. 141-148, 2 text fig.
'14. Chromosomal variations and the formation of the first spermato-
cyte chromosomes in the Eiu-opean earwig, Forficula sp. Journ.

morph., 25, p. 559-585, 64 text fig.
Pinney, Edith.

'08. Organization of the chromosomes in Phrynotettix magnus. Kans.

univ. sci. bull., 4, p. 307-316, pi. 23, 24.
Rabl, C.

'85. tlber zellteilung. Morph. jahrb., 10, p. 214-330, taf. 7-13, 5

text fig.
Rath, O. vom.

'92. Zur kenntniss der spermatogenese von Gryllotalpa vulgaris Latr.

Arch. f. mikr. anat., 40, p. 102-132, taf. 5.
'95. Neue beitrage zur frage der chromatinreduction in der samen- und

eireife. Arch, f . mikr. anat., 46, p. 168-238, taf. 6-8.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 131

Robertson, W. R, B.

'08. The chromosome complex of Syrbula admirabilis. Kans. univ.

sci. bull, 4, p. 273-305, pi. 18-22.
'15. Chromosome Studies. III. Inequalities and deficiencies in
homologous chromosomes: their bearing upon sjoiapsis and the loss
of unit characters. Journ. morph., 26, p. 109-141, 3 pis.
Saedeleer, A. de.

'13. Contribution a I'etude de I'ovogenese dans I'Ascaris megalocephala
bivalens. La cellule, 28, p. 301-362, 6 pis.
Schreiner, A., und K. E.

'07. Neue studien iiber die chromatinreifung der geschlechtszellen. II.
Die reifimg der mannlichen geschlechtszellen von Salamandra maculosa
(Lawr.), Spinax niger (Bonap.) und Myxine glutinosa (L). Arch. d.
bioL, 22, p. 419-492, taf. 23-26.
Sin§ty, R. de.

'01. Recherches sur la biologie et I'anatomie des phasmes. La cellule,
19, p. 117-278, 5 pis.
Smallwood, W. M.

'05. Some observations on the chromosome vesicles in the maturation
of nudibranchs. Morph. jahrb., 33, p. 87-105, pi. 2.
Snook, H. J., and Long, J. A.

'14. Parasynaptic stages in the testes of Aneides lugubris (Hallowell).
Univ. CaUf. publ. Zool., 11, p. 511-528, pi. 25, 26.
Stevens, N. M,

'05. Studies in spermatogenesis with especial reference to the " acces-
sory chromosome." Carnegie instn. Publ. 36, 320 pp., 7 pis.
'08. A study of the germ cells of certain Diptera, with reference to the
heterochromosomes and the phenomena of synapsis. Journ. exp.
zool., 5, p. 359-374, 4 pis.
'10a. The chromosomes in the germ-cells of Culex. Journ. exp. zool.,

8, p. 207-225.
'10^. An unequal pair of heterochromosomes in Forficula. Journ. exp

zool, 8, p. 227-241, 3 pis.
'12a. Supernumerary chromosomes, and synapsis in Ceuthophilus

(sp.?). Biol, bull., 22, p. 219-230, 35 text fig.
'12^. Further observations on supernumerary chromosomes, and sex
ratios in Diabrotica soror. Biol, bull., 22, p. 231-238, 13 text fig.
Stout, A. B.

'12. The individuaUty of the chromosomes and their serial arrangement
in Carex aquatilis. Arch. f. zellf., 9, p. 114-140, taf. 11, 12.
Sturtevant, A. H.

'13. The linear arrangement of six sex-Unked factors in Drosophila, as
shown by their mode of association. Journ. exp. zool., 14, p. 43-59.
Sutton, W. S.

'00. Spermatogonial divisions of Brachystola magna. Kans. univ.
quart., 9, p. 135-160, pi. 32-35.

132 bulletin: museum of comparative zoology.

Sutton, W. S

'02. On the morphology of the chromosome group in Brachystola magna.
Biol, bull., 4, p. 24-39, 11 text fig.

'03. The chromosomes in heredity. Biol, bull., 4, p. 231-251.
Taylor, Monica.

'14. The chromosome complex of Culex pipiens. Quart, journ. micr.

sci., 60, p. 377-398, pi. 27, 28, 3 text fig.
Tennent, D. H.

'08. The chromosomes in cross-fertihzed echinoid eggs. Biol, bull.,

15, p. 127-134, 1 pi., 2 text fig.
Vejdovsky, F.

'07. Neue untersuchungen uber die reifung und befruchtung. Bohm.

gesell. wiss., Prag., 103 pp., 9 taf., 5 text fig.
'11-12. Zum problem der vererbungstrager. Bohm. gesell. wiss.,

Prag., 184 pp., 12 taf., 16 text fig.
Voinov, D.

'14. Sur un nouveau mecanisme determinant le dimorphisme des 616-

ments sexuels; chromosome a polarity variable. Compt. rend, stances

Soc. biologie (S6ances de la reunion biologique de Bucarest du 19

Fevrier, 1914), 76, p. 509-511.
Wassilieff, A.

'07. Die spermatogenese von Blatta germanica. Arch. f. mikr. anat.,

70, p. 1-42, taf. 1-3.
Weismann, A.

'87. Ueber die zahl der richtungskorper und tiber ihre bedeutimg fur

die vererbung. Jena, viii + 75 pp., 3 text fig.
Wilcox, E. V.

'94. Spermatogenesis of Caloptenus femur-rubrum and Cicada tibicen.

Bull. M. C. Z., 27, p. 1-32, 5 pis.
'96. Further studies on Caloptenus femur-rubrum. Bull. M. C. Z., 29,

p. 191-203, 3 pis.
'97. Chromatic tetrads. Anat. anz., 14, p. 194-198.
'01. Longitudinal and transverse divisions of chromosomes. Anat. anz.,

19, p. 332-335.
Wilson, E. B.

'09. Studies on chromosomes. IV. The " accessory " chromosome in

Syromastes and Pyrrochoris with a comparative review of the types of

sexual differences of the chromosome groups. Journ. exp. zool , 6,

p. 69-99, 2 pi., 2 text fig.
'10. Studies on chromosomes, VI. A new type of chromosome combina-
tion in Metapodius. Journ. exp. zool., 9, p. 53-78, 5 text fig.
'11. Studies on chromosomes. VII. A review of the chromosomes of

Nezara; with some more general conclusions. Journ. morph., 22,

p. 71-110, 1 pi., 9 text fig.

WENRICH: spermatogenesis of PHRYNOTETTIX MAGNUS. 133

Wilson, E. B.

'12. Studies on chromosomes. VIII. Observations on the matm-ation
phenomena in certain Hemiptera and other forms, with considerations
on synapsis and reduction. Journ. exp. zool., 13, p. 345-431, 9 pb.
Winiwarter, H. von.

'00. Recherches sur I'ovogfoese et I'organog^nese de I'ovaire des mam-
miferes (lapin et homme). Arch, biol., 17, p. 33-199, pi. 3-8.
Wodsedalek, J. E.

'13. Spermatogenesis of the pig with special reference to the accessory
chromosomes. Biol, bull., 25, p. 8-45, 6 pis.
Zweiger, H.

'06. Die spermatogenese von Forficula auricularia L. Jena, zeit., 42,
p. 143-172, taf. 11-14.

EXPLANATION OF PLATES.

The drawings in plates 1 to 9 were in all cases made with the aid of a camera
lucida, using a Zeiss No. 12 compens. ocular and a Spencer homogeneous
immersion 1.8 mm. objective, N. A. 1.30. Drawings no. 63-65 were made at a
magnification of 2,600 diameters, all the others at a magnification of 3,000
diameters. The drawings have been reduced one third in reproduction.

Plate 10 consists of photomicrographs made at a magnification of 1,150
diameters.

Symbols used:

A designates selected chromosome-pair "A."

D (( « « u li r> >>

"C".

X " accessory chromosome.

G " composite granule.

g ** polar granule.

PLATE 1.

Wenrich.— Spermatogenesis of Phrynotettix magnus.

PLATE 1.

All figures are of the early spermatogonia.

Figs. 1, 2. — In metaphase.

Figs. 3, 4. — In anaphase. In figure 4 the cytoplasm has been drawn in order

to show the hyaline region around the chromosomes.
Fig. 5. — Early telophase.
Figs. 6, 7. — Side view and optical section (polar view), respectively, of an

early telophase in which the chromosomes have begun to expand.

The cytoplasm has been represented in order to show the hyaline

region around the chromosomes.
Figs. 8, 9. — Side view and optical section, respectively, of later telophase.

Vesicles persistent.
Figs. 10, 11. — Side view and optical section, respectively, of late telophase.

Vesicles mostly disappeared.
Figs. 12, 13. — Side view and optical section, respectively, of stage showing

the greatest diffusion of chromatin that was found.

BULL. MUS. COMP. ZOOL. Wenrich- Spermatogenesis Phrynotettix, Plate 1

Wenrich. — Spermatogenesis of Phrynotettix magnus.

PLATE 2.

Figs. 14-20. — Prophases of early spermatogonia.

Figs. 14a, 14b. — Two successive sections of the same cell in earUest prophase.

Figs. 15a, 15b. — Two sections of a single cell in early prophase. Formation

of fine spiral thread.
Fig. 16. — Polar view of later prophase. Fine spiral threads, all distinct.

Composite granules prominent.
Fig. 17. — Side view of later prophase. Accessory in vesicle. Longitudinal

epUt beginning to appear in threads.
Fig. 18. — Section from distal pole of nucleus. Vesicular walls persisting.
Fig. 19. — Later prophase. Composite granules breaking up into polar

granules.
Fig. 20. — Later prophase. Accessory still in vesicle.
Figs. 21, 22. — Side view and optical section, respectively, of telophase of last

spermatogonial division.
Figs. 23, 24. — Oblique side view and optical section, respectively, of early

preleptotene stage.

BULL. MUS. COMP. ZOOL. vVenrich- Spermatogenesis Phrynotettix, Plate 2

PLATE 3.

Wenrich. — Spermatogenesis of Phrynotettix magaus.

PLATE 3.

Figs. 25-27. — Later preleptotene stage.

Fig. 28. — Early? leptotene stage.

Fig. 29.— Later leptotene stage.

Fig. 30. — Early zygotene stage. «•

Fig. 3L — Later zygotene stage.

Figs. 32, 33. — Early pachytene stages.

Fig. 34. — Later pachytene stage.

Figs. 35-37. — Stages of breaking up of composite granule of pachytene stage

into its component polar granules.
Fig. 37.— Diplotene stage.
Fig. 38, a-g. — Different types of tetrads from a single cyst of postspireme,

or tetrad, stage.

BULL. MUS. COMP. ZOOL. Wenrich.- Sp£Rmatogenesis Phrynotettix, Plate 3

Wenrich. — Spermatogenesis of PhrynotelUx magnus.

PLATE 3.

Figs. 25-27. — Later preleptotene stage.

Fig. 28. — Earlji leptotene stage.

Fig. 29. — Later leptotene stage.

Fig. 30. — Early zygotene stage.

Fig. 3L — Later zygotene stage.

Figs. 32, 33.— Early pachytene stages.

Fig. 34.— Later pachytene stage.

Figs. 35-37. — Stages of breaking up of composite granule of pachytene stage

into its component polar granules.
Fig. 37. — Diplotene stage.
Fig. 38, a-g. — Different types of tetrads from a single cyst of postspireme,

or tetrad, stage.

BULL. MUS. COiMP. ZOOL. WtNRiCH.- Spermatogenesis Phrynotettix, Plate 3

PLATE 4.

Wenrich. — Spermatogenesis of Phrynotettix magnus.

PLATE 4.

Figs. 39-44. — Stages in division of first spermatocytes.

Fig. 39. — Polar view of metaphase, showing twelve chromosomes.

Fig. 40. — Polar view of metaphase (eleven chromosomes).

Fig. 41. — Side view of anaphase, showing dyads.

Fig. 42. — Polar view of anaphase, showing twelve dyads.

Fig. 43. — Polar view of anaphase, showing eleven dyads.

Fig. 44. — Telophase.

Figs. 45-49. — Interkinesis stages.

Fig. 45. — Dyads undergoing diffusion.

Fig. 46. — Dyads (except the accessory) still more diffused.

Fig. 47. — Stage of greatest diffusion. OutUne of dyads uncertain.

Fig. 48. — A prophase of the second spermatocyte division. Outline of dyads

reappearing in the middle of the area in which they diffused.
Fig. 49. — Late prophase of second spermatocyte. Dyads fully reformed.

BULL. MUS. COMP. ZOOL. Wenrich- Spermatogenesis Phrynotettix, Plate 4

PLATE 5.

Wenmch. — Spermatogenesis of Phrynotettix magnus.

PLATE 5.

Figs. 50-55. — Stages in division of the secondary spermatocytes.
Figs. 50, 51. — Polar views of metaphase, showing eleven and twelve chromo-
somes, respectively.
Fig. 52. — Side view of metaphase.
Fig. 53. — Side view of early anaphase.
Fig. 54. — Side view of late anaphase.
Fig. 55. — Polar view of telophase.
Figs. 56-61. — Pachytene stages of first spermatocyte.

Figs. 56, 57 showing spireme loop of selected chromosome-pair

"A,"
Figs. 58, 59 showing spireme loops of selected chromosome-pair

"B," and
Figs. 60, 61 showing spireme of selected chromosome-pair "C."
Figure 61 shows in addition the "B" spireme.

BULL. MUS. COMP. ZOOL. WENRICH- SPERMATOGENESIS PHRYNOTETTiX, PLATE 5

PLATE 6.

Wenbich. — Spermatogenesis of Phrynotettix magnus.

PLATE 6.

Fig. 62. — Stages in the transformation of chromosome-pair "A," from
pachytene spireme to metaphase of the first spermatocyte.

Fig. 63. — Similar stages for the chromosome-pair "5," for the equal-type.

Fig. 64. — Same stages for the pair "5," unequal type.

Fig. 65. — Similar transformation stages for chromosome-pair "C"

Fig. 66. — Telophase of an early spermatogonimn, showing members of chro-
mosome-pair "A."

Fig. 67. — Telophase of one of the last spermatogonia, showing members of
chromosome-pair "A."

Figs. 68, 69. — Preleptotene stages, showing persistence of chromosome-pair
"A."

Fig. 70. — Early leptotene. Conjugation of pair "A."

Fig. 71. — Later leptotene. Conjugation of pair "A." Spireme of the ac-
cessory.

Fig. 72. — Spireme of the accessory.

Fig. 73. — Zygotene stage, showing conjugation of pair "A."

Fig, 74. — Zygotene stage, showing conjugation of pair "A" completed.

BULL. MUS. COMP. ZOOL. WeNRicH.- Spermatogenesis Phrynotettix, Plate 6

'." ۥ f*'

^. / F " .* ' ' ■■

64 «_ m^ ^

««<^J

6 6 'A

70 <

PLATE 7.

Wenhich. — Spermatogenesis of Phrynotettix magnus.'

PLATE 7.

Figs. 75-78. — Zygotene stages showing incomplete conjugation of chromo-
some-pair "A."
Fig. 79. — Side view of first spermatocyte metaphase, showing tetrad "A,"

as an extended rod with a constriction in the middle.
Fig. 80. — Polar view of a telophase of the first spermatocyte, showing dyads

"A" and "B."
Fig. 81. Side view of a telophase of the first spermatocyte, showing dyads of

"A" and the accessory.
Fig. 82. — Side view of first spermatocyte telophase, showing the "A" dyads

in both daughter cells.
Fig. 83. — Interkinesis stage, showing the accessory dyad and the "A" dyad

(more condensed).
Fig. 84. — Interkinesis stage, showing the accessory dyad and the "A" dyad

(less condensed).
Fig. 85. — Telophase of the second spermatocj^te, showing the "A," "B,"

and accessory monads.
Fig. 86. — Telophase of the secondary spermatocyte, showing the "A" and

"B" monads.

BU

LL. MUS. COMP. ZOOL. WeNRICH.- SPERMATOGENESIS Phrynotettix, Plate 7

PLATE 8.

W ENRICH. — Spermatogenesis of Phrynotettix maenns

PLATE 8.

Figs. 87-96. — Telophases of spermatogonia, showing members of chromo-
some-pair "5" (all from one animal, no. 772).

Fig. 97, a-m. — One example of chromosome-pair "B," in the pachytene stage
from each of the 13 animals studied.

Fig. 98, a-p. — Examples of chromosome-pair "B" from the pachytene stage
of a single animal (no. 772).

BULL. MUS. COMP. ZOOL. Wenrich.- SPERMATOGENESIS Phrynotettix, Plate 8

I ^ rn^iJ

PLATE 9.

Wenrich. — Spermatogenesis of Phrynotettix maguus.

PLATE 9.

Fig. 99. — Side view of first spermatocyte metaphase, showing roughened
condition of the accessory, of chromosome-pair "5," and of the
polar tips of autosomes.

Fig. 100. — Side view of first spermatocyte metaphase, showing roughened
condition of chromosome-pairs "B" and " C," and of the accessory.

Figs. 101-105.— Telophases of spermatogonia, showing the larger members of
the chromosome-pairs " B" (unequal type) and "C" (type Ci).

Fig. 106. — An example of tetrad "B" from each of the 13 animals studied.

Fig. 107. — An example of tetrad "C" from each of the 13 animals studied.
(The corresponding letters in figures 106 and 107 refer to the same
animal) .

Figs. 108-110. — Telophases of connective-tissue nuclei, showing polar gran-
ules and chromosome "B."

Figs. Ill, 112. — Nuclei from the foUicular envelope.

BU

LL. MUS. COMP. ZOOL. Wenrich- Spermatogenesis Phrynotettix, Pu

A *'« ffll «i

^^ /- J e

106

f 9 h

107

5.-^

W ENRICH.' — SpermatxjeeneHis of Phrynotettii maRDUs.

PLATE 10.

Photomicrographs .

Fig. 113. — Spireme of "A" in the pachytene stage.

Fig. 114.— Tetrad of "A."

Fig. 115. — Pachytene spireme loop of "B."

Fig. 116.— Tetrad of "B."

Fig. 117.— Tetrad of "C" (type Ci).

Fig. 118. — Both tetrads, "B" and "C," dividing equationally in the same

spindle.
Fig. 119. — Tetrad "B" dividing equationally and tetrad "C" dividing re-

ductionally in the same spindle.
Fig. 120.— Unequal division of "C." Smaller dyad going to same pole as

the accessory.
Fig. 121. — Unequal division of "C." Larger dyad going to same pole as the

accessory.

BULL. MUS. COMP. ZOOL. vVenrich- Spermatogenesis Phrynotettix, Plate 10

113

*>

H

114

^-Ji

115

^~ B

116

117

/■

X ♦

/

'< *■

I!

118

X X

\A

I

119

y

4tH

V

120

121

D. H. W. PHOTO.

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.
Vol. LX. No. 4.

A REVISION OF THE LIZARDS OF THE GENUS CYCLURA.

By Thomas Barboub and G. K. Noble.

With Fifteen Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

February, 1916.

No. 4. — A Revision of the Lizards of the Genus Cyclura.
By Thomas Barbour and G. K. Noble.

INTRODUCTION.

Some years ago while working upon West Indian reptiles the
senior author became interested in Cyclura. Every opportunity
has been grasped which offered the slightest probability of securin'g
specimens, so that now the Museum of Comparative Zoology con-
tains more species of that genus than any other museum. That the
series is by no means large, will appear at once. The preparing of
this revision would have been difficult but for the friendly interest
of Mr. H. W. Fowler of the Academy of Natural Sciences of Phila-
delphia; the unique type of our C. nuchalis is in the Museum of the
Academy. We take great pleasure in dedicating C. stejnegeri from
Mona Island to Dr. Stejneger, through whose kindness a paratype
from the small series in the U. S. N. M. has been retained for the
M. C. Z. From the Carnegie Museum in Pittsburgh we have speci-
mens of C. rileyi and of C. macleayi, from the Isle of Pines, presented
in return for the identification, by the senior author, of the Carnegie
Museum series of West Indian reptiles. These he was allowed to
study through the kindness of Prof. L. E. Griffin. A number of
Rhinoceros Iguanas have been received from time to time at the New
York Zoological Park, have died and probably most of them have
found a resting place in the American Museum in New York. These
cannot now be found; one of the examples, however, now mounted
in the Museum of Comparative Zoology, a gift of the N. Y. Zoolog-
ical Society, was said to be from Navassa Island and seems to repre-
sent the species confined to that island. In general, zoological park
specimens, while very valuable for anatomical study, are often with-
out locality, although this is sometimes supplied from the fertile
imagination of an animal dealer. The fine series of examples of C.
carinata in the New York Zoological Park, was, however, a conspicu-
ous exception, since they were known to have come from Turks
Island. Unfortunately this entire, valuable series seems to have
been lost sight of, and a careful search at the American Museum of
Natural History failed to reveal a single one.

140 bulletin: museum of comparative zoology.

GENERAL CONSIDERATIONS.

The members of the genus Cyelura form a small compact group of
species confined to the Greater Antillean district of the West Indian
region. Related to the Rock Iguanas (Ctenosaura) of the Central
American mainland they are nevertheless well set off from the latter
by the possession of the peculiar corneous combs or pectinations on
the hind toes. Except for this character common to all the West
Indian forms, some of these would appear more closely related to
some race of Ctenosaura than to another of the island species. On the
whole, it does not seem advisable to recognize the genus Metapoceras
for the so-called Rhinoceros Iguanas of Navassa, Haiti, and Mona,
since they are ob\aously but slightly advanced modifications of such
a type as the Jamaican Iguana, which is a true, and probably an-
cestral, Cyelura in every respect. The species in the Cayman
Islands is nearly related to the Cuban, and the number of forms
known from the Bahamas represent two groups of species, one showing
affinities with the Cuban, Cyelura macleayi. In the Bahamas, haeo-
lopha of Andros Island seems most like macleayi, with its neighbor,
inornata, hardly less similar; while nuehalis of Fortune, rileyi of
Watlings, and carinata from Turks Island form another well differ-
entiated group of races. The latter species has head-scales of a
simple and scarcely modified, one might, at first sight, say obviously
primitive nature. We imagine, however, that this condition has
been reached secondarily, the transition back through some of the
other species being clearly traceable. So that while the scales of the
head of earinata are of a very simple and undifferentiated character,
it is nevertheless extremely improbable, especially in view of its habi-
tat, that the species can be considered ancestral or anything more
than a reversion to the probably, or possibly, primitive condition for
Cyelura, and Ctenosaura, or their progenitors. It does not seem
wise to lay much stress upon the distribution of the species of Cyelura
as a basis for any zoographic deduction or sm-mise. We know but
little of the habits of the species, the whole group is fast disappearing
and will soon be wholly extinct, and even now we are able to character-
ize but eleven species, probably a comparatively small part of those
in existence even two hundred years ago.

Early writers often mention Iguanas in the West Indies, and of
these some referred to the genus Iguana and some to Cyelura ; among
the latter was Catesby. This authority writing upon the Natural

BARBOUR AND NOBLE: THE GENUS CYCLURA. 141

History of Carolina and the Bahamas, states in 1743, that Iguanas or
Guanas were abundant upon many islands throughout the Bahamas,
so common in fact that schooners were cargoed with them and that
they were carried to Carolina for food. The name Guana is even now
used among the "Conchs" of the Bahamas, who still speak a peculiar
archaic English. A vague idea of how wide-spread these great lizards
were in early colonial times may be gained from the Bahaman place-
names. Thus, there is a Great Guana Cay, off the Abaco coast not
far from Green Turtle Cay, a settlement which once had some impor-
tance. This islet was visited by the senior author in 1904 but no
Guanas were found, and none had persisted to within the memory of
the elder folk living in the tiny hamlet. There is also a Guana Cay
near Little Harbor about half way up the chain of the Berry Islands,
and not far from one of the Bahaman Whale Cays, for this also is a
common place-name. Then we find another Great Guana Cay in the
Exuma chain of Cays. In all of these islands Guanas are now un-
known. On Bitter Guana Cay, however, but a few hundred yards
from the Great Guana Cay in Exuma, Mr. C. J. Maynard tells us
that up to 1915 a few Cycluras were still to be found. He believes
that these represent an undescribed race. As to the status of the other
Bahaman species : — baeolopha is still not uncommon, since its habitat,
Andros Island, is very large and contains much unsettled and indeed
even unexplored territory. Of nuchalis from Fortune Island we know
nothing. Stejneger's species, rileyi, is confined to two tiny islets in
the saline lagoon of Watling's Island; here Riley obtained the types
in 1903 and W. W. Worthington procured a few specimens in March,
1909. Our new species, inornata, is, or was, found upon a little island
called U Cay, in Allen's Harbor, north of Highborn Cay and situated
in the Ai'chipelago between Exuma and New Providence. Here in
1892 Maynard found the Iguanas not uncommon. He revisited the
islet in 1915, was storm bound there and hence had ample opportun-
ity to cover it very completely. He found but two Iguanas still
living upon U Cay. Both of these he shot ; one, our type, he secured,
the other escaped, wounded. Thus the species inornata, which
once doubtless existed on several islands about Allen's Harbor, is now
beyond doubt extinct. Since these creatures are excellent for food,
they are constantly hunted by the native negroes, often with dogs
trained for the purpose. These negroes during the course of their
sponging and turtling voyages cover the entire Bahaman Archipelago,
visiting even the most remote, inaccessible, and infertile cays. There
is a constant search for animal food, which unfortunately is by no

142 bulletin: museum of comparative zoology.

means abundant or easily obtained by the poor inhabitants of one
of Great Britain's most deUghtful but poverty-stricken colonies.
Iguanas are often brought from Andros to market in Nassau, upon
New Providence Island. One of the authors has seen the creatures
for sale there upon several occasions. We have concluded therefore
that a magnificent adult male baeolopha in the Academy of Natural
Sciences of Philadelphia (Reptile Coll. 8120) probably represents
such a specimen, although it is said to have been collected by a Mr.
Wilson on New Providence in 1861. Our belief is that the Iguanas
disappeared from New Providence long before this date. For the
benefit of the herpetologists we should also record that Salt Cay, near
Hog Island, opposite the town of Nassau, has been stocked with
Iguanas brought from Andros Island. Mr. Chamberlain, the owner
of Salt Cay is reported to have stated that they have thriven and
appear to have become well established in their new home.

Of some of the other species of the genus we know even less than of
these we have referred to. Cyclura collei has almost certainly com-
pletely disappeared upon the mainland of Jamaica and it was only by
the greatest stroke of good fortune that Mr. Arthur Perrin of Cam-
bridge who kindly volunteered to make a special excursion for the
purpose, was able to secure the specimen, which we describe, from
Goat Island not far from Old Harbor off southern Jamaica. Dr.
A. G. Mayer tells us that an Iguana was secured a few years ago on
one of the cays near Montego Bay, and that he believes a few still
exist there. Of the species on Navassa, Haiti, and Mona we know
practically nothing. Mr. W. M. Mann who spent some months in
Haiti, and who made an excellent collection of reptiles there in 1913
was unable to learn anything of Iguanas and secured none. Mr.
Halter from the American Museum, visited Santo Domingo in 1915
and could learn nothing of existing Iguanas. Of the Cyclura on
Mona Island we know only that Stejneger quotes Bowdish as say-
ing that he got his specimens in 1901, among the rocks.

The journeys which the senior author has made on a number of
occasions permit us to speak with more authority regarding Cuba.
Gundlach in 1880 wrote " Esta especie vive en varios cayos y en las
costas de la isla de Cuba y de la isla de Pinos; pero es hoy una
especie rara, aunque antiguamente fuese comun y llevada a los merca-
dos, siendo su came estimada como excel ente manjar." He goes on to
say that it usually lives in burrows in the sand dunes or in sandy places
about the coasts where it is easily dug out. Now the Cuban Iguana
has with increasing civilization become still more rare and restricted

BARBOUR AND NOBLE: THE GENUS CYCLURA. 143

in range. C. T. Ramsden, the accomplished naturahst of Guanta-
namo has found a few specimens on the coast cHffs of the extreme
eastern end of the island. Wirt Robinson sent one to the M. C. Z. from
Santiago in December, 1903. The only very young Cyclura we have
seen was one loaned by Ramsden, and obtained by Oskar Tollin during
a trip to Belig, near Cabo Cruz, in the summer of 1914. The species
is still abundant on the Cayos near Manzanillo and those off the south
coast near Santa Cruz del Sur. It is fairly abundant on the Isle of
Pines. In 1915 while Prof, de la Torre and his assistant Sr. V. J.
Rodriguez were collecting with W. S. Brooks and T. Barbour in the
region of Guane, we learned that Iguanas were still not uncommon
in the limestone mountains which encircle the glorious valley of Luis
Lazo. Here we got two fine adult specimens, one for the Museo
Poey in Havana and one for the M. C. Z. Prof, de la Torre says that
Iguanas are also fairly common on the Pan de Guajaibon and he
writes us that he has recently seen one near Baracoa. They persist
as well on many of the small and remote cayos of the north coast.
Curiously enough in spite of what Gundlach says — no one appears
to eat Iguanas in Cuba at the present time. Ramsden has also
observed this and writes me that he has been told that Cubans believe
the Iguana to be very poisonous. When hung up by the tail, they say,
a baba or burujo, as they call it, black drivel or vomit, runs from its
mouth. This is supposed to be deadly. The black vomit due to
blood in the stomach, which marks the final stage of a fatal case of
yellow fever, is also called in Cuba burujo, a name also used for the
grounds of coffee, and it may be that some imagined similarity between
these burujos, coupled with forgetfulness which increased as Iguanas
grew rare, has now spread the idea that Iguanas are unfit for food,
when once, in Cuba as elsewhere, they were eagerly sought after.

It will be noticed that some old, long standing specific names have
been dropped. The reason for this is as follows: — the name nuhila
was based upon a young specimen without locality. The descrip-
tion given by Gray is worthless. The type is mentioned in Boulenger's
Catalogue and hence doubtless is still in existence. When it is
examined it will probably be possible to determine whether the name
supercedes one used here or whether it represents another distinct
species. It is impossible to determine this. Cope attached the name
nuhila to a specimen which he said was from Cat Island and was U. S.
N. M. 14576; but Stejneger tells us that this number is borne by a
specimen of Leiocephalus and that there is now no Cat Island Iguana
in the U. S. N. M. and no evidence that there ever was one. Cuvier

144 bulletin: museum of comparative zoology.

first used cyclura as a specific name for what he called L'Iguane de la
Caroline. After what Catesby said we imagine that he had a Bahaman
specimen which had been carried to Carolina and had probably been
sent to Paris from there. Which Bahaman species he had it is impos-
sible to decide from his meagre descriptions. So unless the t;y^e is
still in existence and sufficiently well preserved, which is improbable,
it will not be possible to more than surmise that Cuvier probably
had a specimen of C. baeolopha. Stejneger and Barbour have both
used the name Cyclura cyclura Cuv. for the Cuban Iguana and this
might be considered as restricting the name. It is probably better
to drop the name altogether until someone studies this old type, and
also the type of nubila in the British Museum.

, KEY TO THE SPECIES.

Median frontal shield enlarged and tubercular.

fei Nasals broadly in contact with the rostral stejnegeri.

b^ Nasals separated from the rostral by small scales, or granules.

c^ Posterior prefrontals separated from the frontal shield by two rows of

scales nigerrima.

<? Posterior prefrontals separated from the frontal shield by a single row

of narrow scales cornuta.

'■ Median frontal shield very slightly enlarged, not tubercular.
h^ Nuchal section of dorsal crest formed of spines much longer than those
of the back section.

c^ Prefrontal and frontal scales small and irregular carinata.

<? Prefrontal and frontal scales enlarged and definitely arranged.

rfi Nuchal section of dorsal crest formed of 14 spines, some twice as

wide as the others nuchalis.

(P Nuchal section of dorsal crest formed of 20 spines, all of about the

same width rileyi.

b^ Nuchal section of dorsal crest formed of spines not distinctly larger than
those of the back section
c^ Scales covering the upper surface of the head flat, depressed, the
interstices forming a network; dorsal crest continuous on the

shoulders collei.

c^ Scales covering the upper surface of the head, especially the snout,

swollen; dorsal crest interrupted on the shoulders.

d^ Inf ralabials separated from the molar scales by a single discontinuous

row of scales; canthal scale very much larger than the precanthal.

e^ A single, large, flat median frontal; gular region the same color

.as ventral surface baelopha.

BARBOUR AND NOBLE: THE GENUS CYCLURA. ' 145

e^ Two large flat median f rentals; gular region white in strong

contrast to rest of ventral surface inornata.

d? Infralabials separated from the malar scales by several rows of small
scales; canthal scale about the same size as the precanthal.

e^ Enlarged frontal shield separated from the posterior prefrontals
by two rows of scales; 37 spines in the back section of dorsal
crest macleayi.

e^ Enlarged frontal shield separated from the posterior prefrontals
by five rows of small scales; 44 spines in the back section of
dorsal crest caymanensis.

DISCUSSION OF THE SPECIES.

Cyclura macleayi Gray.

Plate 1, 2; Plate 13, fig. 5, 6.

Gray Cat. lizards, British mus., 1845, p. 190.

Diagnosis: — Nasals broadly in contact with the rostral. Pre-
frontal region covered by a pair of elongate supranasals, immediately
followed by two pair of large prefrontals, the posterior pair several
times as large as the anterior pair; both pairs of prefrontals broadly
in contact on the median line of the snout, all these scutes covering
the snout slightly swollen and convex. Frontal region between pre-
frontals and the scarcely indicated supraocular semicircles covered
by two irregular rows of scales, the anterior row formed of scales
several times as large as those in the posterior one, immediately fol-
lowing the posterior row a large rounded median scale. Supraorbital
semicircles scarcely differentiated from the supraocular disc, separated
by two, partly by three, rows of scales. Occipital region covered
with enlarged and swollen scales, about two rows of scales between the
occipital and the semicircles. Canthus rostralis consisting of three
large scales, the posterior one elongate and in contact with two supra-
ciliaries which are also elongate, — all these scales on the top of the
head swollen, slightly keeled, and, with the exception of the small
supraocular scales, uniformly enlarged. Dorsal crest low, the largest
spine scarcely over a centimeter high, interrupted on the shoulders
and rump, 37 spines between these two points. Color above, brown-

146 bulletin: museum of comparative zoology.

ish gray sprinkled with pale, yellowish green, the spots very abundant
and partly confluent posteriorly; flanks marked by four broad, verti-
cal stripes of pale bluish gray, each stripe edged with dark slaty gray;
sides and upper surface of the head broadly blotched with pale, bluish
yellow.

Habitat:— Cuba, the Isle of Pines, and the neighboring Cays.

Description: — Adult male, M. C. Z., 11050 from the Valley of Luis
Lazo, western Cuba, April 1915, C. de la Torre and T. Barbour.

Rostral as wide as the mental, broadly in contact with nasals;
nasal large, somewhat pentagonal, perforated by a large, ovoid nos-
tril; each nasal in contact with a large, elongate supranasal and a
squarish postnasal; nasals and supranasals broadly in contact in
the middle of the snout; the pair of supranasals immediately fol-
lowed by two pair of large prefrontals, the posterior pair several times
as large as the anterior pair, both pairs of prefrontals broadly in
contact in the middle hue of the snout ; a few granules on the crossing
point of the two prefrontal sutures; all these scutes covering the upper
surface of the snout slightly swollen and convex; between prefrontals
and the scarcely indicated supraocular semicircular two irregular rows
of scales, the anterior row formed of scales several times as large as
those in the posterior one; immediately following the posterior row a
large rounded median scale; supraorbital semicircle differentiated
from the supraocular disc but the scales on the outer and anterior
portion of the supraocular region smaller than the others ; semicircles
separated by two, partly by three, rows of large scales; occipital
located with its posterior end on a line with the posterior end of
the semicircles; scales of the occipital region enlarged and swollen,
the outer ones largest; about two rows of scales between the occipital
and the semicircles; two or three rows of superciliary shields not
clearly differentiated; canthus rostralis consisting of three large scales,
the first elongate and in contact with two superciliary scales that are
also elongate; all of these scales on the top of the head swollen,
slightly keeled, and, with the exception of the small supraocular scales,
uniformly enlarged; a well-developed series of strongly keeled sub-
oculars continued backward as a supratympanic series; six supra-
labials to the middle of the eye; a series of three or four rows of small
scales separating the supralabials from the suboculars; above the
angle of the mouth and in front of the lower edge of the ear a large
tubercular shield ; above it about the middle of the front edge of the
ear two large shields, preceded by a third, all three tubercular;
below the angle of the mouth a few tubercular scales, irregularly

BARBOUR AND NOBLE: THE GENUS CYCLURA. 147

arranged; five infralabials to the middle of the eye; a single row of
very large, keeled malar scales, and two anterior ones in contact with
the infralabials, the rest separated from the infralabials by one or two
rows of small scales; dorsal and ventral scales small, about eleven
contained in the vertical diameter of the tympanum ; from the nuchal
fold along the median line of the neck and back a row of low, blunt
spines, the largest slightly over a centimeter high; this crest inter-
rupted on the shoulders and rump, thirty-seven spines between these
two points; upper surface of limbs with slightly imbricated, keeled,
posteriorly pointed scales considerably larger than the body-scales;
scales covering the upper surface of the fore arm and tibia much larger
than those covering the upper arm and femur; on the upper arm about
eight, on the lower about five of these scales to the vertical diameter
of the tympanum; a single series of twenty-two femoral pores;
inner side of second toe with one comb, of third toe with two combs,
each consisting of three lobes ; tail compressed, covered with obliquely
keeled scales in vertical rows, forming faintly indicated verticils;
tail surmounted by a serrated crest similar to the body-crest but
formed of slightly larger spines.

Coloration: — Ground tone of dorsal surface brownish gray; whole
dorsal surface sprinkled with pale, yellowish green, the spots very
abundant and partly confluent posteriorly; flanks marked by four
broad, vertical stripes of pale bluish gray; each stripe edged dark
slaty gray, sides and upper surface of the head broadly blotched with
pale bluish yellow; sides of the tail with a series of irregular vertical
stripes of bluish gray becoming regular and evenly spaced posteriorly,
ventral surface somewhat lighter than the upper surface.

Variation and remarks: — A very young specimen, a female meas-
uring only 115 millimeters from snout to vent, collected at Belig, Cabo
Cruz, Cuba, by O. Tollin and now in the collection of C. T. Ramsden
varies greatly in color from the adult, but the lepidosis of the speci-
men is very similar to that of the adult. In this example the ground
tone of the dorsal surface is grayish blue tinged with greenish; along
the middle line of the back there is a series of broad white crossbars
edged broadly before and behind with black; these black and white
crossbars are continued on the sides as a series of wavy stripes, each
stripe pointing obliquely backward; the ventral surface is paler than
the dorsal, and is covered by broken continuations of the lateral stripes.

148

bulletin: museum of comparative zoology.

Material examined.

No. of

Speci-

M.C.Z. mens

11050 1

8456
6915

10966
1096

?i

Age
Adult

hf. grown
adult

adult and
juv.

Sex
male

male
male

males

Locality Date

Valley of Luis 1915

Lazo, Cuba

Guantanamo 1913

Santiago de Cuba 1903

Los Indios
Isle of Pines

Collector
C. de la Torre

and
T. Barbour
C. T. Ramsden
Wirt Robinson

G. A. Link

Remarks
De-
scribed

Cyclura caymanensis, sp. nov.

Plate 3.

Type, an alcoholic skin, M. C. Z., 10534, Ca^onan Islands, probably
from Cayman Brae, 1911, W. W. Brown, Jr.

This species is so closely related to C. viacleayi from Cuba that a
detailed description is superfluous. The distinguishing characters of
the species are adequately presented in the following diagnosis.

Diagnosis: — Scales covering the upper surface of the snout some-
what similar to those of C. viacleayi; enlarged frontal shield separated
from the prefrontals by five irregular rows of small scales instead of
two rows of large ones as in C. macleayi. Scales of the frontoparietal
region much smaller and more numerous than those of C. macleayi.
Supraorbital semicircles not differentiated from the frontoparietal
scales. Canthus rostralis consisting of three short, rather oblong
scales, all about the same size. Dorsal crest not interrupted on either
shoulders or rump, not reduced at all on the shoulders, considerably
reduced on the rump; forty-four scales in the dorsal crest from
shoulders to rump, all spines of this section very low, scarcely over
2 mm. high; spines of the neck and tail-crest about twice as high as
these on an average. Limiting row of each verticil formed of scales
several times as large as the other verticil scales, the row preceding
the limiting row somewhat wider than the other rows. Coloration
like C. macleayi but very much paler and grayer; an indication of
regular pale yellow blotches along the median line of the back; no
stripes or markings on the sides.

Habitat: — Cayman Brae and Little Cayman,

Remarks: — Garman (Bull. Essex inst., 1888, 20, p. 105; author's

BARBOUR AND NOBLE: THE GENUS CYCLURA. 149

separate p. 5) quotes Maynard to the effect that " The Iguana occurs
commonly in the chffs of both tliis island and Little Cayman." He
was discussing the island of Cayman Brae. During the autumn of
1915 this island was visited by the most terrific hurricane of historic
times. The whole terrestrial fauna of the island is said to have suffered
very seriously.

Material examined.
We have only seen a single specimen of this species, the type.

Cyclura baeolopha Cope.

Plate 4, 5, 6; Plate 13, fig. 1, 2.

Cope, Proc. Acad. nat. sci. Phil., 1887, p. 123. Barbom-, Mem.
M. C. Z., 1914, 44, p. 298.

Diagnosis: — Nasals broadly in contact with the rostral. Pre-
frontal region covered by a pair of rectangular supranasals broadly
in contact in the middle line of the snout; each supranasal in contact
with a pair of narrow prefrontals which are followed by a very large
posterior prefrontal; the anterior and posterior prefrontals form a
median suture continuous with the nasal and supranasal suture, —
all of these scutes covering the upper surface of the snout strongly
convex, even tubercular. Frontal region between the prefrontals
and the supraorbital semicircles covered by several rows of large
irregular scales; the row in contact with the prefrontals consisting
of very large scales, the largest being about a third as large as the
posterior prefrontal; between the semicircles on a line with their
anterior end a single large flat scale. Supraorbital semicircles formed
of large tubercular scales clearly differentiated from the slightly
swollen scale of the supraorbital and frontal regions; semicircles
separated by two partly by four rows of scales. Occipital region
covered with scales slightly larger than the frontoparietals, the outer
rows much larger than the others; two rows of scales between the
occipital and the semicircles. Canthus rostralis consisting of a single
large canthal scale and a short squarish precanthal, both swollen and
slightly keeled; the canthal scale in contact with two elongate supra-
ciliaries. Dorsal crest formed of low blunt spines, the largest about
half a centimeter high; this crest broadly interrupted on the shoulders

150 bulletin: museum of comparative zoology.

and rump; 56 spines between these two points. Color above brown-
ish green, tinged on the head, shoulders and along the mid line of the
back with pale yellowish green.

Habitat: — Andros Island, Bahamas. •

The most abundant species still existing in the Bahamas.

Description: — x\dult male, M. C. Z., 6979, Andros Island, Bahamas,
1904, Harvard Bahama Expedition of 1904.

Rostral as wide as the mental, broadly in contact with the nasals,
nasal large, ovoid, and perforated in the posterior half by a somewhat
semicircular nostril; each nasal in contact with a rectangular supra-
nasal and a slightly larger, triangular postnasal; nasals and supra-
nasals broadly in contact in the middle of the snout; each supranasal
in contact with a pair of narrow prefrontals which are followed by a
very large posterior prefrontal; the anterior and posterior prefrontals
form a median suture continuous with the nasal and supranasal suture;
all of these scutes covering the upper surface of the snout strongly
convex, even tubercular; between the prefrontals and the supraorbital
semicircles several rows of large irregular scales; the row in contact
with the prefrontals consisting of several very large scales, the largest
being about a third as large as the posterior prefrontal; between the
semicircles on a line with the anterior end a single large flat scale; the
semicircles formed of large tubercular scales clearly differentiated from
the slightly swollen scales of the supraorbital or frontal regions;
supraorbitals roughly hexagonal and uniform in size; supraorbital
semicircles separated by two, partly by four rows of scales, occipital
located with its posterior end on a line with the posterior end of the
semicircles; scales of the occipital region slightly larger than the
frontals, the outer row of occipitals much larger than the others;
two rows of scales between the occipital and the semicircles; two or
three rows of superciliaries, a single large canthal scale and a short
squarish precantlial on each side; canthal scale in contact with two
elongate superciliaries, the whole series swollen and slightly keeled;
a well-developed series of strongly keeled suboculars continued back-
ward as a supra tympanic series; eight supralabials to the middle of the
eye, a series of three or four rows of small scales separating the supra-
labials from the suboculars; on the anterior edge of the ear three
enlarged tubercular scales, preceded by a group of smaller ones, the
larger one of which is located above the angle of the mouth near the
ear; below the angle of the mouth a regular series of tubercular scales
decreasing in size anteriorly; seven infralabials to the middle of the
eye; a single row of very large, swollen malar scales; the two anterior
ones in contact with the supralabials, the rest separated by a single

BARBOUR AND NOBLE: THE GENUS CYCLURA.

151

row of small scales ; dorsal and ventral scales small, about eleven con-
tained in the vertical diameter of the tympanum; from the nuchal
fold along the median line of the neck and back, a row of very low blunt
spines, the largest about half a centimeter high; this crest interrupted
broadly on the shoulders and rump; fifty-six spines in the dorsal
crest between these two points; upper surface of limbs with slightly
imbricated, keeled, posteriorly pointed scales; scales covering the
upper surface of the fore arm and tibia much longer than those cover-
ing the upper arm and femur; on the upper arm about seven, on the
lower about six of these scales to the vertical diameter of the tym-
panum; a single series of twenty femoral pores, inner side of second
toe with one comb, of third toe with two combs, each consisting of
three lobes; tail compressed, covered with obliquely keeled scales in
vertical rows, tail surmounted by a serrated crest, similar to the body
crest but formed of larger spines.

Coloration: — Ground color brownish green washed on the head and
arms along the mid region of the back and tail, as well as on the
ventral surface with pale yellowish green; the pale yellowish green
of the head somewhat suffused with blue, especially on the dorsal
surface.

Material examined.

No. of

M.CZ.

speci-
mens

Age

Sex

Locality

Date

Collector

Remarks

5960

1

adult

female

6979

1

adult

male

Mangrove Cay,
Andros Island

1904

Bahama Exp.
1904

6947

1

adult

male

Mangrove Cay,
Andros Island

1904

Bahama Exp.
1904

6975

3

Head of

both

Mangrove Cay,

1904

Bahama Exp.

De-

adult male

Andros Island

1904

scribed

2 young

Also four mounted specimens, two adult males, a female and young,
Mangrove Cay, Andros Island, Bahama Exp. of 1904.

Cyclura inornata, sp. nov.
Plate 14.

Type, an adult female, M. C. Z. 11062, U. Cay in Allen's Harbor,
near Highborn Cay, Bahamas, March 2, 1915, C. J. Maynard.

152 bulletin: museum of comparative zoology.

Diagnosis. Nasals broadly in contact with the rostral. Prefrontal
region covered by a pair of elongate supranasals, broadly in contact
with the middle line of the snout, immediately followed by two pairs
of prefrontals and a fifth scale wliich is intercalated at the crossing
point of the sutures, the posterior pair several times as large as the
anterior pair; both pairs of prefrontals broadly in contact in the middle
line of the snout. Frontal region covered by a transverse row of four
large scales in contact with the prefrontals and by two more large
scales mesially arranged and separated from the transverse row by
a single row of small scales. Supraorbital semicircles not apparent,
but the scales of the supraocular region much smaller than those of
the frontoparietal regions. Canthus rostralis consisting of a very
elongate can thai scale preceded by a small precanthal ; all these scales
on the top of head very slightly swollen, some scarcely keeled. Dorsal
crest consisting of very low blunt spines, the largest scarcely three
millimeters high, this crest greatly diminished but not interrupted
on the shoulders, widely interrupted, however, on the rump, sixty
spines from shoulder to rump. Color above grayish brown sprinkled
very slightly with yellowish gray; spinal region tinged with straw
color; sides of the snout blackish; gular region chalky white in strong
contrast to the rest of the ventral surface.

Habitat. — U. Cay in Allen's Harbor, Highborn Cay, Bahamas.
Probably now extinct.

Description of Type. — Rostral as wide as the mental, broadly in
contact with the nasals ; nasal large, somewhat ovoid, perforated by a
large ovoid nostril; each nasal in contact with an elongate supra-
nasal and a triangular postnasal; nasals barely, supranasals broadly
in contact in the middle line of the snout; supranasals immediately
followed by two pairs of prefrontals and a fifth scale incalated at the
crossing point of the sutures; the posterior pair several times as large
as the anterior pair; both pairs of prefrontals broadly in contact in
the middle line; no definite supraorbital semicircles; scales of the
supraocular region much smaller than those of the frontal region; in
contact with the prefrontals a transverse row of 4 large scales; sepa-
rated from this row by a single row of small scales two more large scales
mesially arranged, occipital located well forward and flanked on
either side by a group of scales larger than those of the frontal region;
frontal region covered by scales somewhat larger than those of the
occipital region; two or three rows of superciliary shields not clearly
differentiated; canthal scale very elongated, preceded by a small
prefrontal, all these scales of the top of head very slightly swollen,

BARBOUR AND NOBLE: THE GENUS CYCLURA. 153

some scarcely keeled; a well-developed series of keeled suboculars
continued backward as a supratympanic series, eight supralabials to
the middle of the eye; a series of four or five rows of small scales
separating the supralabials from the suboculars ; on the anterior edge
of the ear a single row of large, strongly tubercular scales preceded by
a group of smaller, tubercular scales grading off in size anteriorly;
below the angle of the mouth a group of large tubercular scales, in
close contact with each other, grading off in size anteriorly; eight
infralabials to the middle of the eye; a single row of large slightly
swollen malar scales; a disconnected single row of small scales be-
tween some of the malars and infralabials ; dorsal scales slightly larger
than the ventrals, about twelve contained in the vertical diameter of
the tympanum; from the nuchal fold along the median line of the
neck and back a row of very low blunt spines, the largest scarcely
three millimeters high; the crest greatly diminished but not inter-
rupted, however, on the rump; 60 spines in the dorsal crest from
shoulder to rump; upper surface of limbs with slightly imbricated,
keeled posteriorly pointed scales, considerably larger than the body-
scales; scales covering the upper surface of the fore arm and tibia
much larger than those covering the upper arm and femur; on the
upper arm about nine on the lower about seven of these scales to the
vertical diameter of the tympanum; a single series of twenty-one
femoral pores; inner side of the second toe with one comb of third
with two combs, each consisting of three lobes. Tail compressed,
covered with obliquely keeled scales in vertical rows, forming distinct
verticils; tail surmounted by a serrated crest, similar to the body
crest, but formed of slightly longer spines anteriorly.

Coloration. Ground tone of dorsal surface grayish brown, sprinkled
very slightly with yellowish gray; the dorsal crest and spinal region
tinged with straw color; upper surface of the head tinged with bluish;
muzzle and sides of head brownish black, gular region chalky white in
strong contrast to the rest of the ventral surface; whole ventral
surface somewhat lighter than dorsal surface; limiting row of the
verticils bluish gray.

Material examined.
We have only seen a single specimen of this species, the type.

154 bulletin: museum of comparative zoology.

Cyclura rileyi Stejneger.
Plate 7; Plate 15, fig. 3, 4.
Stejneger, Proc. Biol. see. Wash., 1903, 16, p. 129.

Diagnosis: — Nasals broadly in contact with the rostral. Scales
of the top of head flat or only slightly swollen. Prefrontal region
covered by a pair of elongate supranasals in contact with the nasals
and also in contact with each other, each supranasal followed by two
large prefrontals, the posterior the larger, the prefrontals of each
side in contact with each other but separated from the series of the
opposite side by two rows of large scales. Top of head behind pre-
frontals covered with numerous small scales; the scales of the supra-
ocular region much smaller than the others; except for a grouping
of a few large scales on each side of the occipital, and a semirosette
of enlarged scales in the frontoparietal region, these scales without
a definite arrangement. Canthus rostralis consisting of a group of
three scales, the canthal and precanthal about the same size and larger
than the third scale. Dorsal crest interrupted on both shoulders
and rump, formed of scales of varying height; nuchal section formed
of about twenty spines about the same in width and varying from
one to ten millimeters in height, according to the proximity to the
extremities of the series, back section formed of seventy-six spines,
scarcely over a millimeter in height, except for the last fourteen spines
which average about 5 millimeters; caudal section formed of heavier
spines than those of the back, about 4 millimeters in height. Limit-
ing row of each verticle clearly differentiated. Ground color, bluish
gray, heavily blotched with confluent tawny yellow spots except on
the posterior ventral surface, which is uniform yellowish gray; head
very much paler, tail darker than the rest of the body.

Habitat: — Two small cays in the large salt water lagoon on Watlings
Island.

Description: — Adult male, M. C. Z. 10918, Watlings Island, Baha-
mas, April, 1915, \\. W. Worthington.

Rostral as wide as the mental broadly in contact with the nasals;
nasals vtry large, about the size of the posterior prefrontals, broadly
hexagonal, in broad contact with each other; each nasal perforated
posteriorly by an elongate nostril; each nasal in contact with an
elongate supranasal and two postnasals; nasals and supranasals

BARBOUR AND NOBLE: THE GENUS CYCLURA. 155

broadly in contact in the middle line of the snout; immediately
following the supranasals and separated by a double row of scales,
two large smooth prefrontals on each side, the posterior pair the
larger; these scales between the prefrontals as large as the scales
covering the frontal region; top of head behind the prefrontal region
covered by small, irregular polygonal shields, those on the supraocular
region much smaller than the others; three poorly indicated rosettes
of larger scales, one on the frontoparietal region and one on each side
of the occipital, each rosette consisting of a somewhat rounded scale
surrounded by a circle of subrectangular scales ; all shields on the top
of head smooth or very slightly swollen, occipital somewhat larger
than the adjacent scutes ; superciliary shields slightly larger than the
supraorbitals; the first two and the last two scales of this series some-
what elongate, canthal scale preceded by a somewhat squarish pre-
canthal of about the same size, a small subcanthal in contact with
the canthal scale anteriorly; 2 loreal scales between precanthal and
postnasal, squarish, not much smaller than precanthal; the rest of
the loreal scales small and elongate, a series of strongly keeled subocu-
lars not reaching the tympanum ; temporal shields small ; on the ante-
rior edge of the tympanum just above the angle of the mouth a group
of three or four large tubercular scales; below the angle of the mouth
five regular rows of enlarged round scales, the series running obliquely
forward ; supralabials low, six to below the center of the eye. Lower
labials larger than the supralabials, six to the center of the eye; two
rows of large molar scales separated posteriorly from the infralabials
by a single row of smaller scales; on each side of the jaws below the
angle of the mouth five or six regular rows of rounded scales; dorsal
crest interrupted on both shoulders and rump; nuchal section of the
crest consisting of twenty spines all of about the same width, varying
in height from one millimeter at the ends of the series to ten milli-
meters in the center; back section formed of seventy-six spines scarcely
over a millimeter in height, the last fourteen spines, however, over five
millimeters in height; caudal section formed of heavy spines about
four millimeters in height; scales on the upper side of the arms larger
than the dorsal scales, those on the fore arm a trifle larger than those
on the upper arm; a single series of twenty-three femoral pores;
inner side of second toe with one "comb," of third toe with two
"combs" each consisting of three lobes; tail covered with faintly
indicated verticils.

Coloration: — Ground color bluish gray, heavily blotched with
confluent tawny yellow markings; upper and lateral surfaces of head
very pale yellow; tail and hind limbs darker than the rest of the body.

156 bulletin: museum of comparative zoology.

Material Examined.

No. of
epeci-
M.C.Z. mens Age Sex Locality Date Collector Remarks

10918 1 adult male Watling's Island 1909 W. W. Worthington Descrip.
9272 1 adult male Watling's Island 1903 J. H. Riley Para-

type

Cyclura nuchalis, nov. sp.
Plate 8, fig. 1, 2.

Type, an adult, Acad. Nat. Sci. Phil. 11985, Fortune Island, Baha-
mas. Collection of Arthur Erwin Brown.

Since this species is in general similar to C. rileyi no detailed descrip-
tion is necessary, the differences being expressed in the following
diagnosis.

Diagnosis: — Nasals broadly in contact with the rostral. Scales
of the prefrontal and frontoparietal regions similar to those of C.
rileyi; enlarged frontal scale proportionately larger, surrounding scales
proportionally smaller than those of C. rileyi; supraorbital semicircles
distinct posteriorly, formed of slightly tuberculate scales. Occipital
scales more tuberculate than those of C. rileyi. Prefrontal scale
more elongate than that of C. rileyi. Dorsal crest very different from
that of any other Cyclura. This crest interrupted on the shoulders,
widely interrupted on the rump; nuchal section formed of fourteen
wide spines, the three central ones largest, about a centimeter high,
the rest grading off in size toward the extremities; of these three
central spines the most anterior one about half a centimeter wide at
the base, in other words about twice as broad as any of the others;
the crest running from shoulders to rump scarcely over a millimeter
high, composed of sixty-two spines; caudal crest slightly higher than
that of the back. Scales of the limiting row of each verticil only
slightly larger than other scales of the verticil; about five scales in the
bulge of each limiting row while C. rileyi averages four or less. The
type of coloration similar to that of C. rileyi but the ground color
brownish gray instead of blue-gray; general tonality of the head gray-
ish brown instead of straw color.

Habitat: — Fortune Island, just to the south of Long Island,
Bahamas.

Material examined.
The type only is known.

BARBOUR AND NOBLE: THE GENUS CYCLURA. 157

Cyclura carinata Harlan.

Plate 8, fig. 3, 4; plate 13, fig. 3, 4.

Harlan, Journ. Acad. nat. sci. Phil., 1824, 4, p. 242, 250, pi. 15.
Barbour, Mem., M. C. Z., 1914, 44, p. 299.

Diagnosis: — Nasals broadly in contact with the rostral. Pre-
frontal region covered by a pair of irregular supranasals; nasals and
supranasals of each side separated from each other by a single large
scale. Frontal, frontoparietal, and occipital regions covered by uni-
formly small scales, irregular in shape and strongly keeled. Supra-
orbital semicircles not differentiated but the scales of the supraocular
region smaller than the other supracephalic scutes. Two large,
vertically arranged canthal scales on each side. Dorsal crest broadly
interrupted on the shoulders and rump; the neck-crest half a centi-
meter high, the body -crest only three millimeters high; color above
brownish gray, with numerous but faint reticulations; head tinged
with blue, chest with smoky.

Habitat: — Turks Island, Southern Bahamas.

Description: — Adult male, M. C. Z. 1252 Turk's Island, Southern
Bahama Islands, 1862, A. S. Bickmore.

Rostral as wide as the mental, broadly in contact with the nasals;
nasals of medium size, somewhat pentagonal perforated posteriorly
by a semicircular nostril; each nasal in contact with a large pent-
angular postnasal and a pair of irregular supranasals; nasals and
supranasals of each side separated from each other by a single, large
triangular scale, all the rest of the scales of top of head small and irre-
gular, no enlarged prefrontal, frontal or parietal scales; a very slight
indication of a supraocular disk; scales of the supraocular and supra-
ciliary region as well as the outer parietal region somewhat smaller
than the rest of the supracephalic scales; scales of the prefrontal,
frontal, and occipital region irregular and all about the same size,
while the scales of the supraocular and outer parietal regions are
uniformly smaller; occipital rather large and located well forward;
all scales of the top of head strongly keeled but hardly tubercular;
two large, vertically arranged canthal scales on each side; a well-
developed series of slightly keeled suboculars carried back a trifle
beyond the orbit, ten supralabials to the middle of the eye; a series
of three or four rows of small scales separating the supralabials from

158 bulletin: museum of comparative zoology.

the suboculars; no tubercular or swollen scales in the temporal
region, only a few enlarged scales below the angle of the mouth,
eleven infralabials to the middle of the eye; a double row of small
slightly keeled malar scales separated from the infralabials by one
or two rows of scales of the small size; dorsal and ventral scales
small; from the nuchal fold along the median line of the neck and back
a row of blunt spines ; on the neck the spines about half a centimeter
high; the crest broadly interrupted on the shoulders and rump;
forty-seven spines in the dorsal crest between these two points, the
first four and the last four spines of this series very much reduced,
the largest spine about three millimeters high; upper surface of limbs
with slightly imbricated, keeled, posteriorly pointed scales consider-
ably larger than the body-scales; on the upper arm about twenty on
the lower arm about twelve of these scales to the vertical diameter
of the tympanum; twenty-three femoral pores; inner side of second
toe with one comb, of third toe with two combs each consisting of
three prominent and two small lobes; tail compressed, covered with
obliquely keeled scales in vertical rows, forming distinct verticils;
the limiting row of each verticil formed of strongly keeled scales;
tail surmounted by a serrated crest, similar to the body-crest but
formed of larger and sharper spines.

Coloration: — Ground tone of dorsal surface brownish gray; mimer-
ous but very faint yellow-brown reticulations extending from head to
tail; these reticulations forming faint yellowish blotches on the head;
tail uniformly yellowish gray; sides of head and gular region tinged
with blue; chest smoky; rest of ventral surface the same color as tail.

Material e:c amine d.
The specimen described.

Cyclura collei Gray.

Plate 9; Plate 15, fig. 5, (i.

Gray, Cat. lizards British mus., 1845, p. 190. Barbour, Mem.
M. C. Z., 1914, 44, p. 298.

Diagnosis: — Nasals separated from the rostral by several rows of
fine granules. Prefrontal region covered by a series of three large
shields on each side, each shield slightly swollen and convex; the

I

BARBOUR AND NOBLE: THE GENUS CYCLURA. 159

series separated by a double row of rather large irregular scales.
Frontal and frontoparietal regions with small and irregular scales
very slightly keeled, each scale depressed so as to make the interstices
stand out like a network. Supraorbital semicircles differentiated
only posteriorly where they are formed of broad, slightly keeled scales.
Occipital region with a huge swelling on each side, each covered with
flat scales. Canthus rostralis formed of a group of three, medium
sized keeled scales. Dorsal crest not interrupted on either shoulder
or rump; largest spine about a centimeter in length; fifty spines in
the crest from shoulders to rump. Color above mud-gray washed
with green anteriorly; a series of straw color stipplings covering the
dorsal and lateral surfaces, these stipplings uniting into blotches
posteriorly.

Habitat: — Jamaica, where it is now exceedingly rare, being only
found on a few islets off the coast where the mongoose has not been
introduced. The mongoose eats the eggs and the very young.

Description: — Adult male, M. C. Z. 9397 Goat Island, near Old
Harbour, Jamaica, 1914, Arthur Perrin.

Rostral wider than mental, separated from the nasals by several
rows of very fine granules; nasal large, ovoid and perforated on the
posterior half by a large semicircular nostril; immediately behind and
adjoining the nasal, a series of three large shields, slightly swollen
and a trifle convex; the series separated by a double row of rather
large irregular scales; the last pair of scales in the series about twice
as large as the anterior pair; the scales of each series broadly in contact
with each other without any intervening scales; a pair of large tri-
angular postnasals; scales of the frontoparietal region all small and
irregular, each scale depressed so that the interstices stand out like
network; supraorbital semicircles only evident posteriorly, formed of
broad, slightly keeled scales; scales covering the supraocular region
smaller than those of the frontal region, each scale very slightly
keeled; occipital smaller than nasals, located well forward between
the semicircles from which it is separated by three row^s of scales;
occipital region swollen out into a pair of huge humps, each covered
with a group of rather large, flat, slightly keeled scales; two rows of
strongly keeled supraciliaries; canthus rostralis consisting of a group
of three, medium sized keeled scales; a well-developed series of
strongly keeled suboculars continued backward as a supratympanlc
series to above the middle of the ear; six supralabials to the middle
of the eye; a series of very small scales separating the suboculars and
the supralabials; above the angle of the mouth and in front of the

160 bulletin: museum of comparative zoology.

lower edge of the ear three large tubercular shields in a horizontal
row; above them along the edge of the ear three or four rows of smaller
scales, strongly tubercular and grading off in size anteriorly; below
the angle of the mouth a few scattered enlarged scales, each surrounded
by a circle of granules; seven lower labials to the center of the eye;
a row of enlarged malar scales, the posterior ones strongly keeled and
separated from the infralabials by a single row of fairly large scales
of which the posterior ones are also keeled; dorsal and ventral scales
small, about thirteen contained in the vertical diameter of the t^on-
panum; from the nuchal fold along the median line of the neck and
back a series of medium sized spines, not interrupted or greatly re-
duced on the shoulders, and only decreased in size on the rump; the
longest spines about a centimeter in length; fifty spines in the crest
from the shoulders to the rump; upper surface of the limbs with
slightly imbricated, keeled, posteriorly pointed scales somewhat larger
than the dorsals; on the lower arm about eleven, on the upper about
fifteen of these scales to the vertical diameter of the tympanum; a
single series of eighteen femoral pores; inner side of second toe with
one comb, of the third toe with two combs, each consisting of three
large and two small lobes; tail compressed, covered with obliquely
keeled scales in vertical rows forming distinct verticils; tail sur-
mounted by a row of spines slightly larger than, but continuous with,
those of the body-crest.

Coloration: — Ground tone of dorsal surface brownish gray, almost
a mud color; top of snout, sides of head washed with green; lower
labials yellowish green; dorsal and lateral surfaces of the body faintly
blotched with straw color, the blotches often breaking up into groups
of small spots; upper surface of the thighs, sides of tail profusely
blotched with the same color; ventral surface muddy gray; the legs
tinged slightly with green.

Material examined.
The specimen described.

Cyclura cornuta (Bonnaterre),

Plate 10.

Bonnaterre, Tabl. encyc. erpet., 17S9, p. 40, pi. 4, f. 4. Stejneger,
Kept. U. S. N. M. for 1902, 1904, p. 670, f. 122-126.

BARBOUR AND NOBLE: THE GENUS CYCLURA. 161

Diagnosis: — Nasals separated from the rostral by a single row of
scales; nasals separated from each other by two rows of scales. Pre-
frontals in a double series of three large shields, strongly convex, the
posterior pair particularly so, the two rows separated from each other
by several rows of small scales, the posterior pair of prefrontals sepa-
rated from the median frontal tubercle by a single row of very nar-
row scales. Supraorbital semicircles scarcely differentiated from the
supraocular scales but somewhat larger than the frontoparietal scales.
A single large canthal scale preceded by a small, hexagonal precanthaL
Dorsal crest low, not over four millimeters high, reduced on the
shoulders, nearly interrupted on the rump but not a distinct break in
the whole. Verticils faintly indicated, the limiting row only a trifle
larger than the row preceding it. Color very faded in the specimens
examined, but probably uniform olive-gray in life, slightly more
yellowish on the head and under surface.

Habitat: — Haiti.

Description: — Two specimens, a young one, and the head of a half
grown individual M. C. Z. 3597, Jeremie, Haiti, 1859, D. F. Weinland.

Rostral wide, as wide as mental, separated from the nasals by a
single row of scales; nasals large, ovoid, perforated by large nostrils
posteriorly, separated from each other by two rows of scales ; on each
side of the top of the snout, immediately following and adjoining the
nasals two rows of three large shields, strongly convex, the posterior
pair tubercular, the rows separated from each other by two or three
rows of small scales; of these two rows of large scales the posterior
pair is nearly as long as the two others together; a large median
frontal tubercle separated from this posterior pair of prefrontals by a
single row of narrow scales; supraocular semicircles scarcely differ-
entiated, but slightly larger than the supraorbital scales and dis-
tinctly larger than the frontoparietals; occipital located well forward
between the semicircles from which it is separated by two or three
rows of small scales, situated on a line between the posterior
borders of the orbits; a single large canthal scale preceded by a
small, hexagonal precanthal; a well-developed series of strongly
keeled suboculars continued backward as a supratympanic series to
above the ear; seven supralabials to below the middle of the eye; two
or three rows of granules separating the suboculars from the supra-
labials; above the angle of the mouth and in front of the lower edge
of the ear a large tubercular shield, above it about the middle of the
edge of the ear another tubercle almost as large; nine lower labials
to the center of the eye; a series of enlarged malar scales, the posterior

162 bulletin: museum of comparative zoology.

ones strongly keeled and separated from the lower labials by four rows
of small scales; from the nuchal fold along the median line of the
body a series of low spines which is much reduced between the shoul-
ders, nearly interrupted on the rump; fifty spines on the back between
these two points; upper surface of the limbs with slightly imbricated,
keeled, posteriorly pointed scales, somewhat larger than the dorsals;
on the lower arm about seven, on the tibia about five to the vertical
diameter of the tympanum ; a single series of about seventeen femoral
pores; inner side of second toe with one " comb" of third toe with two
combs each consisting of three main and two smaller lobes; tail
covered with faintly indicated verticils; tail surmounted by a crest of
heavy spines, a trifle larger than the back spine.

Coloration: — The color of both specimens is very faded, but it
was probably uniform olive-gray in life, perhaps slightly more yellow-
ish on the head and under surface.

Material examined.
The specimens described.

Cyclura nigerrima Cope.
Plate 11; Plate 15, fig. 1, 2.
Cope, Amer. nat., 1885, 19, p. 1006.

Diagyiosis: — Very similar to C. cornuta from which it may be dis-
tinguished by the following characters: —

Nasals separated from the rostral by two rows of scales, one of the
rows of very large scales. Nasals separated from each other by three,
and in part by four rows of scales. Prefrontals separated from the
large median frontal scales by two rows of scales. Supraorbital semi-
circles not apparent but this may be due to the old age of the speci-
men. Precanthal scale as large or a trifle larger than canthal scale,
both subrectangular or squarish. Dorsal crest low, interrupted on
both shoulders and rump, forty-nine spines in the crest between these
two points. Verticils similar to those of C. cornuta but very indis-
tinct, the bulge not very prominent. Color indeterminable through
fading, but probably not unlike that of C. cornuta.

Habitat: — Navassa Island.

BARBOUR AND NOBLE: THE GENUS CYCLURA. 163

Description:— The diagnosis given above is sufficient to characterize
the species quite adequately. It is so similar to C. comuta that no
detailed remarks are necessarv.

Material examined.

The diagnosis was taken from M. C. Z. 4717, Navassa Island,
received from the Smithsonian Institution. Cope states (Proc. Amer.
philos. soc, 1886, 23, p. 264) that his description of nigerrima was
taken from a specimen partially skeletonized. He then proceeds
to diagnose C. onchiopsis,'SL. synonymous form, also from Navassa
Island and based upon three specimens in the U. S. N. M. We
strongly suspect that our specimen is one of the types of onchiopsis.
It was received in Cambridge before the appearance of Cope's paper,
but it is known that Cope often drew up notes and descriptions of
species and frequently subjected them to long delays before they
actually appeared in print. Our specimen agrees remarkably with
Cope's description and Dr. Stejneger writes me that he has not found
the specimens which Cope refers to in the National Museum.

There is also a fine mounted adult male in the M. C. Z. said to
come from Navassa Island and representing this species. It was
presented by the N. Y. Zoological Society.

Cyclura stejnegeri, nov. sp.
Plate 12.

Type, a young specimen U. S. National Museum 29367; Mona
Island, August, 1901, B. S. Bowditch. Paratype M. C. Z. 11145,
formerly U. S. N. M. 29365, an adult male having the same data.

Diagnosis: — Very similar to C. comuta from which it may be dis-
tinguished by the following characters : —

Nasals in contact with the rostral; two, and in part three rows of
scales between the nasals. Prefrontals separated from the enlarged
median frontal scale by two rows of scales. A single large, elongate
canthal scale preceded by three small precanthals. Dorsal crest much
reduced between the shoulders, absolutely interrupted on the rump,
fifty-one scales in the crest from shoulder to rump. Limiting row
of each verticil not much wider than the other rows of the verticils.
Color somewhat faded, uniform dark olive-green.

Habitat: — Mona Island.

164 bulletin: museum of comparative zoology.

Remarks: — No further discussion of this species is necessary in
view of its similarity to C. cormita. It may be added, however, that
Stejneger when writing his Herpetology of Porto Rico, suspected
the distinction between this species. He had, however, no Haitian
material for comparison and was further deterred by some notes which
Gunther (Trans. Zool. soc. Lond., 1882, 11, p. 218, pi. 44) published
regarding a specimen with no locality which died in the London Zoo
and to Stejneger it seemed unlikely that a specimen from Mona
would find its way alive to London. We believe that this specimen
really came from Mona Island, a possibility by no means so remote
as Stejneger seemed to think, especially when the rarity of the species
in Haiti is taken into account. It is not unlikely that the small
Haitian sailing vessels may even visit Mona purposely, to take
Iguanas for food. It is also not improbable that such individuals
may be carried alive to Haiti and thence one may have found its way
to London. {Cf. Stejneger's discussion of this specimen, Ann. rept.
U. S. N. M. for 1902, 1904, p. 671).

Material examined.

We have only seen Bowdish's specimens listed by Stejneger (loc^
cit.). One of these is now in the M. C. Z., 11145, a paratype, figured.

PLATE 1.

Barbour and Noble. — The Genus Cyclura.

PLATE 1.

Cyclura madeayi Gray."

Fig. 1.— Side view of head. M. C. Z. 8456. Guantanamo, Cuba.
Fig. 2. — Upper view of head of the same specimen.

Fig. 3. — Dorsal view of very young specimen. Belig, Cuba. Collection of
C. T. Ramsden.

BULL. MUS. COMP. ZOOL.

Barbour and Noble— Cyclura, Plate 1

GEORGE NELSON. PHOTO

PLATE 2

Barbour and Noble. — The Genua Gyclura.

PLATE 2.

Cyclura madeayi Gray.

Fig. 1. — Side view of head of adult male. M. C. Z. 11050. Luis Lazo, Cuba.
Pig. 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. 200L.

Barbour and Noble.-Cyclura, Plate 2

GEORGE NELSON. PHOTO

y

PLATE 3.

Barbour and Noble. — The Genus Cyclura.

PLATE 3.

Cyclura caymanensis Barbour & Noble.

Fig. 1.— Side view of head. Type M. C. Z. 10534. Cayman Islands.
Fig. 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble- Cyclura, Plate 3

GEORGE MELSON. PHOTO.

PLATE 4.

B&BBOUB AND NoBLB. — The Genus Gyclura.

PLATE 4.

Cyclura baeolopha Cope.

Dorsal view of young specimen. M. C. Z. 6975A. Mangrove Cay, Andros
Island, Bahamas.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.-Cyclura, Plate 4

GEORGE NELSON. PHOTO.

PLATE 5.

Bahbour and Noble. — The Genus Cyclura.

PLATE 5.

Cyclura baeolopha Cope.

Fig. 1. — Side view of head of adult male. M. C. Z. 6975B. Mangrove Cay,

Andros Island, Bahamas.
Fig. 2 — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble— Cyclura, Plate 5

->>%:

.»>:, '.-^V

GEORGE NELSON. PHOTO.

PLATE^e.

Barbour and Noble. — The Genus Gyclura.

PLATE 6.

Cyclura baeolopha Cope.

Fig. 1.— Side view of head of adult female. M. C. Z. 5960.
Fig. 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.— Cyclura, Plate 6

GEORGE NELSON, PlIQTO.

PLATE 7.

Barbour and Nobi«. — The Genus Cyclura.

PLATE 7.

Cyclura rileyi Stejneger.

Fig. 1.— Side view of head of adult male. M. C. Z. 10918. Watling's

Island, Bahamas.
Fig- 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.— Cyclura, Plate 7

:^^^,.

GEORGE NELSON, PHOTO.

PLATE 8.

Harbour and Noble. — The Genua Cyclura.

PLATE 8.

Cyclura nuchalis Barbour & Noble.

Fig. 1. — Side view of head. Type, Collection Acad. nat. sci. Phil., 11985,

Fortune Island, Bahamas.
Fig. 2. — Upper view of head of the same specimen.

Cyclura carinata Harlan.

Fig. 3.— Side view of head of adult male. M. C. Z. 1252. Turk's Island,

Southern Bahamas.
Fig. 4. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.— Cyclura, Plate 8

GEORGE NELSON, PHOTO.

PLATE 9.

Barbour and Noble. — The Genus Cychira.

PLATE 9.

Cyclura collei Gray.

Fig. 1. — Side view of head of adult male. M. C. Z. 9397. Goat Island,

near Old Harbour, Jamaica.
Fig. 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.— Cyclura, Plate 9

GEORGE NELSON, PHOTO.

PLATE 10.

Barbour and Noble. — The Genus Cyclura.

PLATE 10.

Cyclura cornuta (Bonnaterre) .

Fig. 1. — Side view of head of half grown specimen. M. C. Z. 3597A. Jere-

mie, Haiti.
Fig. 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour a>jd Noble— Cyclura, Plate 10

^-^ --'*^'

GEORGE NELSON, PHOTO.

PLATE 11.

Barbour and I\oble. — The Genus Gyclura

PLATE 11.

Cydura nigerrima Cope.

Fig. 1.— Side view of head of adult male. Cotype? M. C. Z. 4717. Na-

vassa Island.
Fig. 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble— Cyclura, Plate 11

V'

\

GEORGE NELSON. PHOTO

Babboub and Noble. — The Genus (lycliira.

PLATE 12.

Cydura stejnegeri Barbour & Noble.

Fig. 1. — Side view of head of adult male. Paratype M. C. Z. 11145. Mona

Island.
Fig. 2. — Upper view of head of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.-Cyclura, Plate 12

-*>:

GEORGE NELSON. PHOTO

PLATE 13.

Barboub and Noble. — The Genus Gyclura.

PLATE 13.

Cyclura baeolopha Cope.

Fig. 1.— Segment of taH. M. C. Z. 5960. Cf. Plate 6, fig. 1.
Fig. 2. — Foot of the same specimen.

Cyclura carinata Harlan.

Fig. 3.— Segment of tail. M. C. Z. 1252. C/. Plate 8, fig. 3.
Fig. 4. — Foot of the same specimen.

Cyclura viacleayi Gray.

Fig. 5.— Segment of taU. M. C. Z. 11050. Cf. Plate 2, fig. 1.
Fig. 6. — Foot of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.— Cyclura, Plate 13

>

KSS

mfi

J'v

4Uttr

E. N. FISCHER DEL.

PLATE 14.

Barbour and Nobla. — The Genus Gyclura.

PLATE 14.

Cyclura inornata Barbour & Noble.

Fig. 1.— Upper view of head. Type M. C. Z. 11062. U Cay in Allen's

Harbor, near Highborn Cay, Bahamas.
Fig. 2. — Side view of head of the same specimen.
Fig. 3. — Segment of tail of the same specimen.
Fig. 4. — - Foot of the same specimen.

BULL. MUS. COMP. 200L.

Barbour and Noble —Cyclura, Plate 14

E. N. FISCHER DEL.

PLATE 15.

Barbour and Noble. — The Genus Cyclura.

PLATE 15.

Cyclura nigerrima Cope.

Fig. 1.— Segment of tail. M. C. Z. 4717. Navassa Island.
Fig. 2. — Foot of the same specimen.

Cyclura rileyi Stejneger.

Fig. 3.— Segment of tail. M. C. Z. 1091S. Watlings Island.
Fig. 4. — Foot of the same specimen.

Cyclura collet Gray.

Fig. 5.— Segment of tail. M. C. Z. 9397. Goat Island, near Old Harbour

Jamaica.
Fig. 6. — Foot of the same specimen.

BULL. MUS. COMP. ZOOL.

Barbour and Noble.-Cyclura, Plate 15

.^ "T

^

/Xw-^Iil^^*^^

2

J?^^

E. N. FISCHER DEL.

fLD ^

i-i.

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LX. No. 5.

THE ANTS OF THE PHILLIPS EXPEDITION TO
PALESTINE DURING 1914.

By W. M. Wheeler and W. M. Mann.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM,

February, 1916.

No. 5. — The Ants of the Phillips Expedition to Palestine during 1914-

CONTRIBUTIONS FROM THE ENTOMOLOGICAL LABORATORY OF THE
BUSSEY INSTITUTION, HARVARD UNIVERSITY, NO. 107.

By W. M. Wheeler and W. M. Mann.

The junior author, while -accompanying Dr. John C. Phillips on a
recent zoological expedition to Palestine and the adjacent countries
for the Museum of Comparative Zoology, succeeded in amassing quite
a collection of ants. As many collections of these insects have been
made from time to time in Egypt and Asia Minor and have been care-
fully described in numerous papers by Ern. Andre, Emery, Forel,
Mayr, and Ruzsky, it seemed improbable that another collection
would contain anything new. After the specimens were mounted
and examined, however, we were surprised to find among them a new
and peculiar species of Deromyrma and a few undescribed varieties
and subspecies of well-known Mediterranean species. We decided,
therefore, to publish a list of all the forms collected, together with
such field-notes as seemed interesting.

FORMICIDiE.

1. Ponera eduardi Forel. ^ 9 (ergatoid).— Baniyas, Syria.

2. Sima bifovcolata Mayr var. syriaca, var. nov.

Worker. Agreeing very closely with the typical form described
from Delagoa Bay and Zanzibar, except in the following characters: —
the tibiee have no suberect hairs, the mandibles have only three instead
of four or five teeth, and the petiolar node is semicircular in profile.
The tip of the gaster is not brown. The peculiar paired granular pits
on the occiput seem to be quite as distinctly developed as in the type.

Several workers found running on plants at Wady Gazelle, Sinai
Peninsula. The typical form has been recorded by Mayr only as far
north as the White Nile where it was taken by Tragardh.

3. Aphaenogaster splendida ^oger. y — Rasheya, Syria; in damp,
shady places.

168

bulletin: museum of comparative zoology.

4. Aphaenogaster {Deromyrma) -phillipsi, sp. nov. (Fig. 1).

Worker. Length 6-6.8 mm.

Very slender. Head twice as long as broad, with the rather small
but convex eyes well in front of the middle of its sides; cheeks sub-
parallel, distinctly convex, postocular borders of head straight and
gradually converging to the occipital border, which has a strongly
reflexed margin and is only about half as broad as the distance between
the eyes. In profile the upper surface of the head is convex anteriorly,
the gular surface flat. Mandibles with slightly concave external
borders, three stout apical teeth and several basal denticles. Clypeus
flat and subcarinate, the middle of its anterior border broadly and

Fig. 1. Aphaenogaster {Deromyrma) phillipsi, sp. nov. Body of ijporker in profile;
head of sEtme from above.

sinuately excised. Frontal area impressed, distinct, with a median
cannula. Frontal carinae approximated and parallel behind. An-
tennae very slender, the scapes surpassing the occipitaF margin by
about § their length; funiculus without a club, its joints very slender,
the second, third, and terminal longest. Thorax slender, pro- and
mesonotum subequal, the former in profile feebly convex, the latter
sloping, slightly concave in the middle and with a small but distinct
convexity in front just behind the promesonotal suture. Mesoepinotal
constriction abrupt, short and moderately deep. Epinotum in profile
about 1| times as long as high, its base convex in profile and twice as

WHEELER AND MANN: ANTS OF PALESTINE. 169

long as the concave declivity, armed with two small, rather acute
teeth, which are directed upward and slightly backward and outward.
Petiole more than twice as long as broad, in profile with a rather low,
rounded node. Postpetiole about 1| times as long as broad, less than
half again as broad as the petiole, in profile with a similar but some-
what larger node. Gaster elongate elliptical, narrowed in front.
Legs very long and slender; spurs of the posterior tibiae short but
distinct.

Gaster very smooth and shining, remainder of body more feebly
shining. Mandibles subopaque, finely and densely striate; head,
thorax, petiole, and postpetiole shagreened, the meso- and epinotum
opaque, rugulose-punctate; the anterior portion of the head above,
including the antennal foveae and excluding the clypeus, longitudi-
nally rugose, becoming reticulately rugulose and punctate posteriorly ;
pronotum and upper surfaces of petiolar and postpetiolar nodes
smoother and more shining. Epinotal declivity shining, feebly and
transversely rugose.

Hairs yellow, very short, blunt, sparse on the body, entirely lacking
on the legs; very short, but distinct and oblique on the antennal
scapes, especially towards their tips. Pubescence absent.

Head, thorax, petiole, postpetiole, and antennae pale ferruginous;
legs brownish yellow; gaster clear, pale yellow, with the posterior
I of the first segment dark brown.

Described from eleven workers from Petra, Palestine. These were
found in the early morning eating portions of the bait with which
small mammal traps had been baited.

This species differs considerably from either of the two previously
described Palaearctic species of Deromyrma, cecconii Emery from
Crete and rhaphidiiceps Mayr from Turkestan. The Cretan species
is smaller (5.7 mm.), has the body black, the tibiae with oblique hairs,
the petiolar node is angular in profile, the epinotal teeth are longer,
the head is shorter and of a different shape behind and the antennal
funiculi have an indistinct, 4-jointed clava. In coloration and the
shape of the head phillipsi resembles rhaphidiiceps, but the latter is
smaller, the occipital margin of the head has no reflected margin, the
clypeal border is entire, the first funicular joint is longer than the
second and the tibiae are hairy.

5. Messor rufotestaceus Foerster. U . — Wady Gharandel, Sinai
Peninsula and Petra, Palestine; in the former locality living in
crater nests, in the latter under stones and more abundant.

170 bulletin: museum of comparative zoology.

6. Mcssor barbanis Linne sviljsp. strudor Latr. var. oricnialis Emery.

y . — Rashe\'a, Animik, and Baruk, Syria; Ain Gleidat and Wady
Hisa/ Palestine.

7. Mcssor harharus Linne subsp. semirufus Ern. Andre. ^ . —
t ' Ammik, Hasbeiya, Zahleh, and Baniyas, Syria; Wady Hisa,
^ Palestine. One of the commonest ants in Syria, in crater nests.

8. Mcssor harharus Linne subsp. semirufus Ern. Andre var. concolor
Emery. ^ . — Shiba, Syria; Fuweila, Arabia; Wady Feran,
Sinai Peninsula; Wady Hisa, Palestine.

9. Messor harharus Linne subsp. mcridionalis Ern. Andre. ^ . —
Petra and Wady Mojeb, Palestine; Shiba and Wady El Katana,
Syria.

10. Messor harharus Linne subsp. acgyptiacus Emery. ^ 9 cf • —
W'ady Feran, Wady Gazelle and Mt. Sinai, Sinai Peninsula;
Cairo (Mann) and Fayum (Wm. Granger), Egypt.

1 1 . Phcidolc paUidula Nyl. Ql y . — Petra and Wady Kerak, Palestine ;
Ammik, Syria; Wady Feran and Wady Gazelle, Sinai Peninsula.

12. Phcidolc incgacephala Fabr. 2i y . — Baruk, Syria.

13. Phcidolc sinaitica Mayr. % y . — Cairo, Egypt.

14. Crematogastcr scutcllaris Oliv. subsp. schmitti Mayr var. ionia
Forel. y . — Rasheya, Syria; Petra, Palestine. We refer these
specimens to Forel's variety on account of the distinct infuscation
of the head and thorax.

15. Crematogaster auherti Emery subsp. jehovae Forel. ^ . — W^ady
Kerak and Ain Gleidat, Palestine; Shiba, Syria.

16. Crematogastcr auherti Emery subsp. antaris Emery. ^ . — Mt.
Sinai, Sinai Peninsula; Fuweila, Arabia.

17. Crematogastcr incrmis Mayr. y. — - Mt. Sinai, Wady Gazelle,
and W'ady Feran, Sinai Peninsula; Wady Mojeb, Palestine.
This is the commonest species of the genus in the Sinaitic Penin-
sula.

IS. Crematogaster lortcti Forel. y .— Ain Gleidat, Palestine. Many
colonies, nesting under stones in moist localities.

19. Monomorium vcnustum F. Smith subsp. niloticum Emery, y . —
Wady Gazelle, Sinai Peninsula.

20. Monomorium solomonis Linne. y . — Fuweila, Arabia; Wady
Gharandel, Sinai Peninsula; Wady Mojeb, Palestine.

21. Monomorium solomonis Linne subsp. suhopacum F. Smith var.
Phoenicia ^jinery. y 9 cf. — Akaba, Arabia; Petra, Palestine.

22. Monomorium hicolor Emery subsp. nitidiventre Emery, y 9 . —
Cairo, Egypt.

WHEELER AND MANN: ANTS OF PALESTINE. 171

The female (dealated) measures 3.8-4 mm. and resembles the
worker in color and sculpture, except that the base of the gaster above
is yellowish red like the head, thorax, and pedicel and more opaque.
Head longer than broad, with straight, subparallel sides, distinctly
excised posterior border and rather angular posterior corners. Thorax
elongate elliptical, fully 2| times as long as broad ; in profile the dorsal
surface of the mesonotum, praescutellum and scutellum form a
straight line; epinotum with a pronounced median longitudinal
impression. Postpetiole not broader than the petiole, distinctly
broader than long.

23. Monoviorium abeUlei Ern. Andre. ^ . — Wady Feran and
Wady Gazelle, Sinai Peninsula; Petra, Palestine.

24. Monomorium {Holcomyrmex) dentigerum Roger. ^ . — Petra,
Palestine.

25. Monomorium {Holcomyrmex) dentigerum Roger var. haal, var. nov.

Worker. Differing from the typical form in its decidedly darker
color, the body and antennal scapes being castaneous or blackish
brown; the discs of the mandibles, the clypeus, front, cheeks, mesono-
tum, pleurae, and lower portions of the petiole and postpetiole deep
red; the legs and tips of antennal scapes yellowish brown.

Numerous specimens from Shiba, Syria and Wady Kerak, Palestine.

26. Leptothorax tuberum Fabr. var. luteus Forel.

Three workers from Ain Gleidat, Palestine agree closely with the
description of this form except that they have the posterior half of
the first gastric segment and the whole of the remaining segments pale
brown instead of yellow, like the remainder of the body. We deem it
inadvisable to describe this form as a new variety on the basis of so
little material.

27. Tetramorium striativentre Mayr. ^ . — Wady El Katana, Syria;
Wady Mojeb, Palestine.

28. Tetramorium caespitum Linne. ^ . — Petra, Palestine.

29. Tetramorium caespitum Linne var. forte Forel. ^ . — Baruk,
Syria.

30. Tetramorium caespitum Linne var. schmitti Forel. S . — Baruk
and Ammik, Syria; Wady Mojeb, Palestine; Mt. Sinai, Sinai
Peninsula.

172 bulletin: museum of compakative zoology.

31. Tetramormm caespitum -Linne subsp. punicum F. Smith var.
lucidulum 'Emery. ^. — Petra, Palestine; Ammik and Baniyas,
Syria; Mt. Sinai, Sinai Peninsula. The colonies at Petra were
very populous and were nesting under stones.

32. Tetramorium caespitum Linne subsp. judas, subsp. nov.

Worker. Length 2.3-3.5 mm.

Allied to the subspecies semilaeve Ern. Andre but the whole body,
except the mandibles and clypeus, shining and the sculpture very
feeble. Head much as in semilaeve, with the sides, posterior corners
and a streak between the front and the posterior corners smooth and
shining, the rugae on the front delicate and numerous, continued
nearly or quite to the occiput. Pro- and mesonotum smooth and
shining, with only traces of rugae at the sides, epinotum subopaque
and rugose; pleurae more or less rugose as are also the sides of the
petiole and postpetiole, the summits of the nodes of the latter shining,
nearly smooth, or merely indistinctly punctate-rugulose. Gaster
smooth and shining throughout. Color dark brown; mandibles, cly-
peus, cheeks, antennae, and legs testaceous.

Nine specimens from Wady Mojeb, Palestine.

This form seems to be near the var. splendens Ruzsky of the subsp.
semilaeve, but we infer that the thorax is more strongly sculptured
than the head in this form.

33. Bothriomyrmex meridionalis Roger var. syria Forel. ^ . — Ain
Gleidat, Palestine; Rasheya and Wady El Katana, Syria; Wady
Gazelle, Sinai Peninsula.

34. Tapinoma erraticum Latr. ^. — Petra, Palestine; Baruk,
Syria.

35. Tapinoma erraticum Latr. subsp. nigerrimum Nyl. y . — Petra,
Palestine; Fuweila, Arabia.

37. Acantholepis frauenfeldi Mayr. ^ 9 . — Wady Hisa, Palestine.

38. Acantholepis frauenfeldi M.a,yT Ya.r. bipartita F. Smith. ^ 9. —
Rasheya and Wady El Katana, Syria.

39. Acantholepis carbonaria Emery. ^ . — Wady Gazelle, Sinai
Peninsula.

40. Acantholepis capensis Mayr var. canescens Emery. ^ . — Two
workers from Fuweila, Arabia agree very closely with a single
specimen of this ant from Erythraea in the senior author's col-
lection. This variety is also recorded from Kaka on the White
Nile and from Bogosland and Somaliland.

WHEELEfl vVND . MANN : ANTS OF PALESTINE. 173

41. Plagiolepis pygmaea Latr. ^ 9 . — Ain Gleidat and Petra,
Palestine; Ammik, Syria; Wady Feran, Sinai Peninsula.

42. Prenolepis {Nylanderia) jacgerskjoeldi Mayr. y . — Cairo,
Egypt; Wady Kerak, Palestine.

43. Formica rufibarhis Fabr. var. clarorufibarbis Ruzsky. ^ . —
Baruk, Syria. The specimens have the base of the gaster red
as in the var. clara Forel, but the top of the head is infuscated as
in the typical rufibarhis.

44. Cataglyphis bombycina Roger. "^ g . — Lake Fayum, Egypt
(Wm. Granger).

45. Cat(iglyphis bombycina Roger var. sinaitica, var. nov.

Soldier and Worker. Differing from the typical form in the much
darker coloration, the occiput, thorax, petiole, gaster, and femora
being deep castaneous brown or even blackish, the knees, tibiae, head,
antennal scapes, and first funicular joint and in some specimens also
the thoracic dorsum, paler brown; the mandibles in the worker
yellowish red. Antennal funiculi beyond the first joint and the teeth
of the mandibles black. The hairs on the tibiae are distinctly longer
than in the typical form and the silver pubescence, especially on the
gaster, is even denser.

A single soldier and numerous workers from Wady Gazelle, Sinai
Peninsula.

46. Cataglyphis albicans Roger subsp. livida Ern. Andre. S 9 . —
Petra, Palestine; Wady Gharandel, Sinai.

47. Cataglyphis viatica Fabr. subsp. bicolor Ern. Andre. ^ . — Baruk
and Ammik, Syria; Wady Mojeb, Palestine.

48. Cataglyphis viatica Fabr. subsp. bicolor Ern. Andre var. nigra
Ern. Andre, y .— Cairo, Egypt; Lake Fayum, Egypt (Wm.
Granger).,

49. Cataglyphis viatica Fabr. subsp. bicolor Emery var. orientalis
Forel. y .— Wady Mojeb and Wady Hisa, Palestine; El Katana,
Syria; Wady Feran and Wady Gharandel, Sinai Peninsula.

50. Cataglyphis cursor Fonsc. subsp. aenescens Nyl. S . — Shiba,
Syria.

51. Camponotus (Myrmoturba) macidatus Fabr. subsp. thoracicus
Fabr. var. oasium Forel. y. — Ammik, Syria; Wady Feran,
Sinai Peninsula.

52. Camponotus {Myrmoturba) maculatus Fabr. subsp. thoracicus
Fabr. var. cypriacus Forel. ^ 9 . — Wady Gharandel, Sinai
Peninsula.

174 bulletin: museum of comparative zoology.

53. Camponotus {Myrmoturha) maculatus Fabr. subsp. thoracicus
Fabr. var. xerxes Forel. y . — Zahleh, Syria.

54. Camponotus {Myrmoturha) viaculatm Fabr. subsp. thoracicus
Fabr. var. sanctoides Forel. ^ — Wady Feran and Mt. Sinai,
Sinai Peninsula.

55. Camponotus (Myrmoturha) maculatus Fabr. subsp. thoracicus
Fabr. var. mortis Forel. y . — Wady Feran, Sinai Peninsula.

56. Camponotus (Myrmoturha) maculatus Fabr. subsp. sanctus Forel.
S. — Shiba, Bakeyas, El Katana, and Rasheya, Syria; Petra,

Palestine.

57. Camponotus (Myrmoturha) maculatus Fabr. subsp. turkestanicus
Ern. Andre. ^ . — Wady Hisa, Palestine.

58. Camponotus (Myrmoturha) maculatus Fabr. subsp. haldaccii
Emery. ^ . — Baniyas, Syria.

59. Camponotus (Myrmoturha) maculatus Fabr. subsp. aethiops Fabr.
var. concavus Forel. g . — Shiba, Syria; Mt. Herraon, Palestine.

60. Camponotus (Orthonotomyrmex) lateralis Oliv. var. atricolor Nyl.
^ . — Rasheya and Ammik, Syria.

61. Camponotus (Orthonotomyrmex) interjectus Wsi\T. ^. — -Wady
Kerak, Palestine.

62. Polyrhachis (Myrmhopla) simplex Mayr. y 9 cf . — Wady
Kerak, Palestine; Wady Feran, Sinai Peninsula.

This species was very abundant in both of these localities, always
in damp places and always associated with a certain tree. In Wady
Feran several nests were seen on plants. These were made of portions
of leaves and twigs fastened together with films of silk. In Wady
El Katana, near the Dead Sea, the only nest found was beneath a stone
at the base of a tree. It contained many larvae, some of which were
lying on the ground, and others on a sheet of silk. On nearly every
tree in the vicinity of this nest there were many Membracidae which
were constantly attended by the Polyrhachis workers and in most
cases even sheltered in sheds' constructed by the ants. The workers
were observed while carrying the larvae up the trees and using them
to spin the silk of the sheds.

MAR 23 ]m

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.
Vol. LX. No. 6.

RESULTS OF THE YALE PERUVIAN EXPEDITION OF IQU.

THE ARACHNIDA.

By Ralph V. Ch.\mberlin.

With Twentt-pive Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

March, 1916.

No. 6. — Results of the Yale Peruvian Expedition of 1911. — The
, Arachnida.

By Ralph V. Chamberlin.

The arachnids, upon a study of which this paper is based, were
collected for the most part by Prof. H. W. Foote while a member of
the Yale Peruvian Expedition of 1911. The collection has proved to
be extraordinarily interesting, presenting a very large proportion of
previously undescribed forms among which are eighty -two new species
and twelve new genera. A few specimens were secured at Panama;
but aside from these, the material all comes from localities in a little
known section of southeastern Peru at elevations ranging from 3,000
to 11,500 feet above sea level. Comparatively few of the species
seem to be identical with forms recorded from more northerly locali-
ties of Peru by Taczanowski in his Les Araneides du Peru (Bull. Soc.
nat. Mosc. 1878, 53, p. 278-374; Horae Soc. entom. Ross., 1878, 14.
p. 140-175; 1879, 15, p. 102-136). Some of the localities mentioned
are not on published maps.

This collection of arachnids was turned over to the author for study
through the kindness of Professor Foote, who has also consented that
the entire lot remain the property of the Museum of Comparative
Zoology. Prof. A. Petrunkevitch of Yale had planned to report
upon the collection but the press of other duties prevented. He had
made notes and drawings of several species which were kindly sent
to me.

The principal Peruvian localities from which the arachnids of the
collection come with elevations and months during which specimens
were secured are as follows : —

Arequipa,

7,600 feet

June

Cuzco,

11,500 feet

July

Huadquina,

5,000 feet

July

Ollantaytanibo,

9,000 feet

July

Torontoy,

8,000 feet

July

San Miguel,

6,000 feet

July

Urubamba,

9,500 feet

July

Lucma,

7,000 feet

August

Paltaybamba,

5,000 feet

August

178 bulletin: museum of comparative zoology.

Santa Ana, 3,000 feet August

Tineoehaca, 7,000 feet August

Vilcabamba, 10,000 feet August

Conservidayo River, —— August

Sorontoy, 7,000 feet September

The specimens from Panama were collected in June.

SCORPIONIDA.

BUTHIDAE.

TiTYUs FOOTEi, sp. nov.
Plate 1, fig. 1-4.

Carapace and dorsum of preabdomen from testaceous to dark
reddish with blackish markings near eyes and along the anterior border
of the carapace, and especially across the caudal borders of the abdo-
men, the dark bands widest at middle. Legs and pedipalps yellowish.
Cauda proximally from yellow to testaceous of reddish cast becoming
decidedly darker distad, the last one or two segments usually black
or nearly so ; sting, in individuals in full color, more or less chestnut.

Carapace and all tergites of preabdomen strongly granular. Pre-
abdominal sternites each roughened with minute granules. All seg-
ments of Cauda granular, the fifth most strongly so. All tergites of
preabdomen with median dorsal keel well developed, that of the fifth
incomplete behind, and with two well-developed dorsal keels each
side of it; sternite of fourth and fifth segments with five keels of which
the median is very weak. First segment of cauda with ten keels, a
complete median lateral ridge being present between the usual upper
and lower lateral ridges ; second segment with keels the same except-
ing that the median lateral is developed only caudad; third and
fourth segments w^ith the eight ridges, two median dorsal and median
ventral and the two upper and two lower laterals; fifth segment with
but five distinct keels, the odd one being a complete median ventral;
all keels granular.

Basal portion of sting finely granular. A well-developed acute
spine beneath, (Plate 1, fig. 1).

Ca^MBERLIN: THE ARACHNIDA. 179

Legs and pedipalps with strongly developed longitudinal granular
ridges.

Hand of pedipalp broader than tibia, its ridges strongly developed,
these continuing upon the immovable finger where they become
smooth. Finger with fifteen or sixteen oblique lines of granules,
(Plate 1, fig. 2) ; not fully twice as long as proximal hand (cir. 7:4).

Comb with 20-22 teeth. Basal middle lamella enlarged, Plate 1,
fig. 4.

Length of types 38 to 52 mm.

Cauda of male proportionately longer and more slender than in the
female. (Fourth segment of cauda in female 2.5 mm. wide X 5.0 mm.
long, in male 2.5 mm. wide X 5.8 mm. long).

Localities. — Huadquina, 5,000 feet, July 26. (Type, M. C. Z., 121;
paratypes, M. C. Z. 122, five specimens.) Ollantaytambo 9,000 feet,
July 20. (M. C. Z. 123, one young specimen).

Named for Prof. H. W. Foote of Yale University.

This species is apparently nearest to T. bahiensis (Perty). In
coloration it differs in having the tibia of the pedipalp light and uni-
form, not conspicuously darkened in the way characteristic of bahien-
sis, and in having the fifth segment of the cauda conspicuously dark.
The hand of the pedipalp is conspicuously more slender and its ridges
are much more strongly developed and those along the immovable
finger continue without break to the proximal end of the hand. The
basal middle lamella of the comb is relatively much larger, being
more inflated, and with its inner edge semicircular instead of straight.
The spine on under side of sting is more slender and acute with at
most only a single and relatively small granule on its upper edge.

BOTHRIURIDAE.

Brachistosternus andinus, sp. nov.
Plate 1, fig. 5-8.

General color yellow, the carapace and tergites of the preabdomen
darkened with brown and blackish markings; carapace free from the
darker color in a triangular area in front of eyes and in a narrower
area behind them ; tergites of preabdomen lighter along caudal border
and in a narrow median longitudinal line; postabdomen above pale

180 bulletin: mitsefm of comparative zoology.

yellow, below darker, being mottled with brown which is most abun-
dant on the fifth segment and is partially absent from the first.

Carapace and first six segments of the abdomen smooth and shin-
ing; seventh abdominal segment finely granular, (Plate 1, fig. o).

First segment of postabdomen equal in length and breadth, the
second a little longer than wide. Dorsal and upper lateral keels of
first segment of cauda finely granular, the granules on those of other
segments fewer and weaker or essentially absent. Ventral median
keel of Cauda absent from first four segments, weakly developed on
the fifth; lower lateral keels of fifth segment conspicuous, strongly
granular, the lower lateral keels of the other segments obsolete.
Ventral surface of the fifth segment conspicuously granular excepting
at anterior end, that of the fourth segment with scattered granules,
the venter elsewhere smooth.

Marginal keels of femur of pedipalp smooth; tibia along posterior
ventral edge with trichobothria eight in number, these arranged in
two rows of four each; hand of chela thicker than the tibia, smooth,
with a row of 8-10 trichobothria under the outer ridge. The lateral
granules on the mesal surface of the movable finger remaining apart
and distinct from the main row over entire length, eight in number in
each row, (Plate 1, fig. 6).

The distal lamella of comb abruptly much narrower than the proxi-
mal ones so that the anterior edge appears indented at its l)eginning
as in B. ehrenhcrqi. Teeth of each comb 28 in number, (Plate 1,
fig. 8).

Length 34 mm.

Locality— Ollantaytambo 9,000 feet, July 20. (T\pe, M. C. Z.
124, one gravid female).

SOLPUGIDA

SOLPUGIDAE.

MuMMUCiA VARiEGATA (Gervais).

Caleodes variegata Gervais, Gay Hist. Chile. ZooL, 1849, 4, p. 15, t. 1, f. 2.
Mummuda variegata Simon, Ann. Ent. soc. France, 1879, ser. 5, 9, p. 151,
t. 3, f. 29, .30.

Loco%.— Ollantaytambo, 9,000 feet, July 20. (M. C. Z. 125,
one female).

CHAMBERLIX: THE ARACHNIDA. 181

PHALANGIDA.
GOXOLEPTIDAE.

GONOLEPTES ENOPLUS/ Sp. nOV.

Plate 2, fig. 7-8; Plate 3, fig. 1-5.

Main portion of carapace very dilute ferruginous, its caudal and
caudolateral borders dusk\' but integument along edge of these parts
white; area in front of cervical furrow deeply mottled with black;
tergite of abdomen blackish with caudal border lighter. Legs black,
but the distal articles of tarsus III conspicuously light, whitish, and
the metatarsus and especially the tarsus of leg IV also somewhat
lighter. Coxa IV dilute ferruginous black across distal end both
below and above excepting the white line along the edge, its process
also black excepting the paler tips. Coxae of legs I to III densely
mottled with black excepting niesal ends which are paler. Sternites
of abdomen blackish excepting the paler caudal borders and the white
integument adjoining the caudal edges.

Body very wide toward caudal end of cephidothorax from where
very stronglj' and abruptly narrowed cephalad to opposite bases of
thirfl legs, then more gently narrowing to anterior end; also con-
spicuousl}^ narrowing caudad to end of abdomen; the greatest width
of carapace ecfual to body length, (Plate 2, fig. 7).

Carapace with only four complete transverse sulci, but the first of
these with two branches on each side; first sulcus lying in a deeper
cer\ical furrow, as usual angularly bent caudad at middle; second
sulcus moderately angularly bent forward at middle; third sulcus
with ends bent abruptly caudad and then running sul)parallel with
body axis; fourth sulcus angularly bent forward at middle; all sulcj
connected by a median longitudinal sulcus which is deepest between
the first two. Eye-tubercle sharply set off and elevated, wider than
long, the two cones near its caudal border not especially high. An-
terior border of carapace conspicuously elevated as usual, the elevated
rim much widest at middle. Lower frontal margin extended lietween
the notches for mandibles and laterad of each in an acute spine-like
process. Upper frontal margin at each lateral corner bearing several
low teeth. Lateral border of carapace behind conspicuously elevated

' efowXoi, armed.

182 bulletin: museum of comparative zoology.

as usual, the ridge largely composed of a continuous row of tubercles
which, large behind, decrease cephalad and disappear caudad of level
of first sulcus, the row of tubercles continued also across caudal
border. A row of two or three tubercles in a second low ridge occurs
just mesad of the outer rim a little caudad of widest part of carapace.
The cones, immediately in front of last sulcus and near middle line,
are rather low, distally rounded. Area between last two sulci with
numerous larger and smaller tubercles or granules, the area between
the second and third w4th fewer and mostly very small granules and
the area between the first and second with still fewer scattered and
inconspicuous granules, (Plate 2, fig. 7).

First three tergites of abdomen with a continuous transverse row of
conspicuous tubercles. Anal scutum without processes; rather ob-
scurely granular.

First three pairs of coxae proximally parallel, the first two bending
moderately cephalad from near middle of length, the third transverse
for whole length ; second coxae scarcely one fifth longer than the third ;
first coxae along midventral surface with a row of distinct granules,
the second with a few almost obsolete granules and the third smooth.
Fourth coxae very strongly enlarged and directed ectocaudad as usual;
terminating on dorsoectal side in a stout process which has a shorter
prong on the ventral side beyond middle of length; on mesal side of
distal end with a shorter stout process which is toothed on its ectal
side.

Spiracles moderate, distinct.

First article of mandibles strongly narrowed toward base; distal
half with a strongly and abruptly elevated hump on dorsal surface.
Distal article with fewer hairs in front and distad, (Plate 2, fig. 8).

Pedipalps when extended shorter than carapace; not crossed.
Trochanter roundly elevated and smooth above; ventrally with a
single conical tubercle bearing distally a bristle. Femur a little
complanate beneath and bearing few low granules from which hairs
arise; dorsally also with scattered low granule-like elevations at bases
of some of hairs. Patella smooth. Tibia not compressed; along
mesoventral edge with two similar but stouter spines with proximal
portions thicker as usual. Tarsus along mesoventral edge with two
subequal spines of usual type and along ectoventral edge with two
similar large spines and between these a much smaller spine and a
similar one between the distal large spine and the base of the claw.
Claw much shorter than tarsus, conspicuouslv curved, (Plate 3, fig.
1,2).

chamberlin: the arachnida. 183

Femora of first two pairs of legs finely and conspicuously granular;
femora of third legs more coarsely granular, the granules on ventral sur-
face toward distal end largest; patella with a few granules; tibia with
finer granules above but ventrall}' with larger distinct tubercles or
teeth. Fourth leg with metatarsi abruptly more slender than tibia;
trochanter IV on dorsoectal surface toward base with a stout, distally
truncate process and on opposite side with two larger, distally acumi-
nate processes; femur conspicuously curving dorsomesad, with nu-
merous stout processes of differing lengths of which one from the
dorsal surface, one opposite on ventral surface and one toward distal
end from mesoventral surface are much the largest (Plate 3, fig. 4, 5) ;
patella with numerous small seriate tubercles of uniform size; tibia
above and laterally with setigerous tubercles like those of patella but
ventrally with a median-longitudinal series of 7-9 long processes which
decrease in length distad; metatarsus with a series of low setigerous
tubercles.

Tarsus I with five joints of which three are in the distal division;
II with eight joints of which three are in the distal division; III with
five joints of which the most distal is greatly enlarged and bears a
setigerous process on its anterior end projecting between the claws,
the ventral surface of articles densely clothed with short fine hairs,
(Plate 3, fig. 3) ; IV with six joints.

Length 8 mm. ; greatest width of carapace also 8 mm.

Length of leg 111 mm.; of leg II 20 mm.; of leg III 16 mm.; of
leg IV cir. 21+ mm.

Locality — San Miguel, 6,000 feet, July 24. (Type, M. C. Z. 126,
one male).

GONOLEPTES HUADQUINAE, sp. nOV.

Plate 4, fig. 3-8.

General color above brownish grey, the head region darker and the
anal scutum blackish; a series of lighter circular spots across dorsum
of carapace and abdomen, one over and about each tubercle. Coxae
of first three pairs of legs beneath black, the coxae of the fourth pair
subtestaceous. Sternite of abdomen blackish with lighter transverse
borders and spots. Trochanters of legs above and below brownish
yellow; other joints of legs blackish but femora IV testaceous with
fewer dusky mottlings; femora I to III paler at distal end with an
obscure narrow light annulus near middle ; tibiae all with an obscure

184 bulletin: museum of comparative zoology.

median pale annulus; metatarsi also with faint median annulus and
metatarsi III and IV with a more distinct light ring at distal end;
tarsi III and IV with distal article pale and also with a siihmedian
pale annulus.

Carapace broadest at level of third sulcus from where the body
narrows conspicuously caudad, the comparati\ely long aljdomcn
being more abruptly narrowed or constricted at third segment with
the anal scutum narrowly rounded; cephalad the cephalothorax is
abruptl\' narrowed near \eve\ of third legs as usual and a little in-
dented on each side caudad of anterior end.

Carapace with the usual four sulci of which the first is bent l)uck
angularly at middle and the second is more moderately angulate in
the opposite direction; the third sulcus is also weakly angulate and
the fourth more strongly and acutely so between bases of the cones;
a distinct median longitudinal sulcus between the first and the second
sulci. Eye-tubercle distinctly limited, moderate in height, much
wider than long; paired interocular processes close together, distinct,
conically acimiinate, pale in color. Tubercles along lateral sub-
marginal elevated rims smaller and more obscure, these, as usual, be-
coming larger but of only moderate size caudad, the tubercles widely
separated, those across caudal border similar. A pair of wideh' sepa-
rated tubercles between eye-tuliercle and first sulcus. Area between
sulci I and II with a few low tubercles; the second area with an
irregular transverse row of more distinct setigerous tubercles and the
tubercles of the last area more nvunerous as usual. Conical processes
of carapace moderate in size, acutely acuminate with an acute curved
branch on subdorsal side, (Plate 4, fig. 6, 7). Anterior border of head
elevated as usual, being limited caudad by a transverse furrow as in
other species. Frontal margin not dentate.

First three tergites of abdomen each with a transverse row of tuber-
cles which decrease in size laterad and which are well separated.
Anal tergite with a transverse row of four tubercles. Sternites of
abdomen with corresponding rows of smaller tubercles.

Coxae I distally bent rather strongly forwards, the second more
moderately so, the third straight and a little shorter than the second;
coxae I to III inclusive each with a row of distinct tubercles along the
midventral line. Coxae IV of the usual general form; tuberculate;
the tubercles moderate, numerous but not dense; at distal end on
mesal side with a proximally stout but not long process which distad
is abruptly narrowed and terminates in an acicular point; on ectal
side of distal end a shorter conical process.

chamberlin: the arachnida. 185

Spiracles conspicuously exposed.

First article of mandible strongly constricted as usual with the
article abruptly expanded distad of the constriction and with the
dorsal hump conspicuous. Second article long, moderately curved
ventrad at middle, the dorsal (anterior) face being somewhat depressed
or excavated and bearing a moderate number of short stiff hairs.
Fingers of chela crossing at tips, the outer one stouter, longer and more
strongly bent at tip, (Plate 4, fig. 3).

Pedipalps in type about as long as body, but in a second specimen in
which the abdomen is strongly retracted they are considerably longer.
Coxa strongly elevated above, the ventral process long and subcylin-
drical, white in color. Trochanter constricted proximally; at distal
end on ventral surface a large conical tubercle. Femur subcylindrical,
a few inconspicuous tubercles along ventral surface. Patella unarmed.
Tibia along mesoventral line with two long curved spinous processes
and two very small ones between the two large and in front of the
distal one respectively; along ecto ventral line with two long less
curved processes of which the anterior is the larger and near the base
of the latter and distad of it a very much reduced spine. Tarsus in
mesoventral line with two long processes and along ectoventral line
with two large processes and a third smaller process on ectal side be-
tween the others and in a second specimen with a small spine in front
of others on ectal as well as mesal side. Claw much longer than the
tarsus, (Plate 4, fig. 4, 5).

Legs I to III with femora, patellae, tibiae, and metatarsi tubercu-
late, more strongly developed on femora than on more distal joints
and more conspicuously ventrally than dorsally. Fourth legs with
tubercles of femora largely replaced by more conspicuous conical
processes; tubercles of other joints also stronger than on other legs
Tarsus I of five segments, of which three are in the second division;
II with eight, also with three in the second division; III with four
(the most proximal with two false sutures thus giving appearance of
seven articles), these densely clothed ventrally with fine short hairs
as usual, none of the articles modified specialh'. Leg IV, (Plate 4,
fig. 8).

Length 5.6 mm.; greatest width of carapace 4 mm.

Length of leg I cir. 8 mm.; of leg II, 14 mm.; of leg III, 11 mm.;
of leg IV, 14 mm.

Locality.— Huadquina, 5,000 feet, July 24. (Type, M. C. Z. 127,
one female). San Miguel, 6,000 feet, JuJy. (Paratype, M. C. Z. 128,
one female).

186 Bl'LLETIX: MUSEUM OF COMPARATIVE ZOOLOGY.

GONOLEPTES SCOTIUS/ sp. IIOV.

Plate 3, fig. 6-8; Plate 4, fig. 1-2.

Body throughout black of brownish cast, lines of white along cara-
pace caudally and laterally and bordering tergites and sternites of
abdomen. Caudal edges of coxae IV also bordered with white.
Carapace under lens seen to be not densely mottled with small lighter
spots in head region and over border behind. The legs similarly
mottled but light spots more numerous, the trochanters of all legs
and the tarsi of third and fourth legs and in some also of first and
second lighter in color. Palpi paler than legs, the spots being larger
and in part confluent.

Body relatively narrower than in enoplus, the greatest width of
carapace being much less than the total length of body and about
equal to length of carapace. The general form of body shown in
Plate 3, fig. 6.

Carapace with four transverse sulci of which the first is deeper and
is angulate at middle in the usual way, bifurcate on each side toward
end; second sulcus angularly bent forwards at middle, the third
toward each end bending back caudoectad; a strongly impressed
longitudinal median sulcus between first and second transverse sulci
which is more weakly continued between second and third and third
and fourth. A transverse sulcus also indicated in line with caudal
edge of eye-tubercle. Eye-tubercle sharply defined, moderate in
height, wider than long, bearing a few small but distinct tubercles
with paired ones present in place of the cones of enoplus rounded or
boss-like, these two smaller than eyes but larger than other tubercles.
Median portion of head just in front of eye-tubercle elevated and the
anterior border each side of this part elevated as usual. Lower
frontal margin excavated for mandibles as usual ; upper frontal margin
without distinct teeth at each rounded lateral corner but with rounded
boss at end of ridge there. From cervical region caudad on each
lateral border of carapace the usual series of tubercles ; a tubercle-free
area opposite the third sulcus and behind this about three noncontigu-
ous larger tubercles and then along caudal margin a series of smaller
tubercles which are widely separated and irregularly spaced. The
conical processes at posterior end of carapace long, acuminate, close

' ffKoTios, dusky.

chamberlin: the arachnida. 187

together. Dorsal area of carapace between first and fourth trans-
verse sulci with scattered small granules.

The three anterior tergites of abdomen each with a transverse row of
well-separated conical tubercles which decrease in size from the middle
ones laterad. Anal tergite caudally rounded, with two rounded ele-
vations near base.

First two pairs of coxae curving a little forwards, the third straight
throughout its length; second coxae a little longer than the third;
first coxae with the usual row of setigerous tubercles which are very
small; the second and third coxae smooth or nearly so. Fourth
coxae greatly enlarged in the usual way; ending distally on the ecto-
dorsal side in a large curved process which is unbranched, and on
opposite side in a shorter one which presents a low rounded process
on ectal side below tip. Coxae all with numerous small granules
each bearing a hair. Spiracles distinctly exposed.

First joint of mandibles immediately distad of its constriction with
an abruptly elevated rounded hump. Second joint with a few hairs
in front toward distal end.

Trochanter of pedipalp with a rounded hump above and a single
setigerous conical tubercle below. Femur subcylindrical ; along ven-
tral median line between base and middle three large conical tubercles.
Patella unarmed. Tibia along meso ventral line with two spinous pro-
cesses of which the anterior is much more slender and along the ecto-
ventral line with two spinous processes of which the anterior is the
larger, and in front of the latter a third much smaller spine. Tarsus
along ectoventral line with five spines of which the first from caudal
end and the third and fourth are larger; three spines in meso ventral
line; claw large, as long or somewhat longer than the article (Plate 4,
fig. 2).

Femora of first three pairs of legs granulotubercular \entrally, the
tubercles of the third pair largest; tibiae of these legs also tubercular,
the tubercles small, those of the ventral surface and especially those
of the third pair largest; metatarsi more obscurely granular. Tro-
chanter of leg IV granular; on inner side with a rounded process and
on outer side with a much longer acuminate process which curves
back caudad above base ; femur tubercular and distad with a number
of larger processes of which two much exceeding the others in size
arise from the ventral surface toward the distal end and distally
curve mesad; patella uniformly seriately tubercular; tibia with
numerous tubercles which are larger, subconical, on ventral surface;
metatarsus over proximal half with numerous tubercles, the distal,

188 bulletin: Must:u.M of comparative zoology.

clavately enlarging lialf nearly smooth, (Plate 4, fig. 1). Tarsus I
with five segments; II with six articles of which three are in the second
division; III with six which are clothed ventrally with numerous fine
hairs and none of which is specially enlarged, but the metatarsus is
greatly enlarged toward its distal end which is abruptly narrowed
and resembles_a tarsal article in form and pubescence, while from the
ventral surface toward proximal end arises a very large, abruptly
ciu'ved branch or process, (Plate 3, fig. 8).

Length of type 7.8 mm. ; greatest width of carapace 6.1 mm.

Length of leg I cir. 9 mm.; of leg II, 16 mm.; of leg III, 12-1- mm.;
of leg IV cir. 17 mm.

ioco///(/.— Lucma, 7,000 feet, August 7. (Type, M. C. Z. 129,
one male; paratypes, 130, two males).

Pachylus orinus,^ sp. nov.
Plate 5, fig. 1-3.

Body strongly narrowed cephalad; conspicuously constricted both
laterally and dorsally along first furrow just back of eye-region and
nearly in line with coxae of third legs; widest a little back of con-
striction where the sides are convexlv rounded and from w^here the
body narrows cavidad to end of abdomen with no constriction be-
tween cephalothorax and abdomen; abdomen caudally semicircularly
rounded.

Carapace with five transverse sulci of which the most anterior lies
in the previously mentioned constriction and at its middle is bent
caudad at a distinct angle; the second one forms a slight angle at
middle with apex cephalad, while the other sulci are straight; no
connecting longitudinal median sulcus excepting a very weak one
between the first and second. Eye-tubercle sharply set off, about
equal in length and breadth; the cone between eyes high and acute;
eye-tubercle separated by a transverse furrow or depression from the
conspicuovisly elevated frontal border. Lower frontal margin deeply
notched or excavated for insertion of mandibles and between the two
notches extending ventrad in a spiniform process, otherwise unarmed ;
imarmed above. Carapace with lateral borders strongly elevated,
the marginal ridge over its middle, ectally curving portion conspicu-

1 '

opeicot, a mountaineer.

CHAMBERLIN : THE ARACHNIDA. 189

ously granular, the end portions smooth; caudal border with a trans-
verse row of tubercles; the area between fifth and fourth sulci with a
transverse row of fewer and less developed tubercles, the two areas
next cephalad with still fewer and more obscure tubercles similarly in
a transverse line, (Plate 5, fig. 1).

The first two abdominal segments each bearing above a transverse
row of numerous, well-developed granules. Dorsal anal scutum
bearing five conspicuous spinous processes of which the median is
much longest, and in front of the most anterior of these on each side
a series of much smaller teeth which decrease progressively cephalad;
the second spine from the median one on each side is elevated dorsally
along its whole length in a keel-like form, the elevated portion pro-
jecting cephalad over base in a rounded, tubercle-like process.

First three pairs of coxae proximally parallel, those of the first two
pairs distally curved considerably cephalad, the third ones straight
throughout. Second coxae but little more than one fifth longer than
the third (15:11). The coxae of the second and third pairs along
anteroventral surface with a low, sharply elevated edge which is
obscurely tuberculate; first coxae with no elevated edge but wnth a
row of very small tubercles. Fourth coxae very strongly enlarged as
usual and directed caudad; each one terminating at distal end on
ectal side above in a stout acute spine and on opposite mesal side of
end in a similar but smaller spine; elsewhere smooth, not tuberculate
or dentate.

Spiracles distinct, rather large.

]Mandil)les stout; first article strongly narrowed proximally, the
expanded distal portion with a strongly elevated, long, keel-like
hump above which is smooth ; second article in front sparsely clothed
with hairs and with fewer hairs behind toward distal end, (Plate 5,
fig. 2).

Pedipalps rather short, when extended scarcely as long as carapace ;
not crossed. Trochanter short, . subcylindrical, with a low hump
above on which are one or two slight tubercles; ventrally with a few
hairs springing from slightly tubercular bases. Femur cylindrical,
not compressed, moderately convexly elevated from end to end
above; wholly without spines; a few short hairs from slightly tuber-
cular bases below. Patella unarmed. Tibia not compressed; along
mesoventral edge with two rather slender, acutely acuminate spines
and between these a minute third one; along ecto ventral edge with
three long spines and between the caudal and middle one of these a
fourth minute one; the middle one of the larger spines much largest

190 bulletin: museum of comparative zoology.

and a little surpassing the distal end of tarsus when joints are flexed,
the other two spines proximally abruptly thicker than distad, anterior
spine cognate to base of large spine. Tarsus on mesal side ventrally
with a series of four slender, distally bristle-like spines; on ectal side
with a series of about twelve pale slender spines which are very short
excepting two which are much larger and of about same size as the
three of mesal side ; tarsal claw of about same length as tarsus, (Plate

5, fig. 3)- .

First three pairs of legs with femora, patellae, and tibiae especially
on ventral surface finely but rather sparsely tubercular, a hair arising
from each tubercle, distal joints wholly smooth. Fourth legs stout,
with tarsi abruptly much more slender than the metatarsi; tro-
chanter with two robust granules on ectal side and at distal end on
mesal side with a stout thorn; femur granular above, along ventroectal
edge with a series of mostly stout conical spines which, beginning at
about one fourth the length from the proximal end as low tubercles
increase regularly in length distad; along meso ventral surface a series
of fewer and lower conical tubercles and on the mesal surface with two
or three irregular series of conical tubercles and spines ; patella strongly
tubercular above and laterally, and with tubercles replaced by stout
conical spines of which three or four are comparable in length to the
larger ones of femur; tibia above and laterally strongly tubercular,
below with tubercles replaced by stouter conical tubercular elevations
and longer spines like those of the proximal joints ; metatarsus above
and laterally densely granular, below with stouter seriate cones or
teeth, much smaller and more uniform than the spines of the proximal
joints. Tarsus I with five joints, three of which are in the distal
division; tarsus II with eight joints; III and IV with six joints.

General color dilute ferruginous, the head region weakly dusky;
caudal border of carapace and the first two tergites of abdomen black;
patellae and distal ends of femora and tibiae of legs dusky or black,
the fourth legs a darker, more strictly ferruginous cast than the others.
Abdomen darker beneath than coxae of legs.

Length to base of median caudal spine 10.2 mm. ; to tip of caudal
spine 12.3 mm. Greatest width of carapace 6 mm.

Length of leg I, (exclusive of coxa) cir. 15 mm.; of leg II, 22 mm.;
of leg III, 7 mm.; of leg IV, 30 mm.

Lom/%.— Huadquina, 5,000 feet, July. (Type, M. C. Z. 131).
San Miguel, 6,000 feet, July. (Paratype, M. C. Z. 132, one specimen).

chamberlin: the arachnida. 191

COSMETIDAE.

Paravanones peruvianus, sp. nov.

Pl'ate 2, fig. 1-6.

General background of body a dilute ferruginous. Carapace
dusky from a close network of black lines, the color less dusky in a
deltoid area back of eye-tubercle with apex caudad and embracing
two short black stripes, one caudad of each eye, in a narrow mid-
longitudinal line, and along the transverse sutures and in a broader
area along each side of the carapace, tubercles pale. Ventral surface
of cephalothorax and coxae paler, a clearer light ferruginous; under
lens seen to be covered with a less dense network of dark lines. Ab-
dominal segments both above and below black or nearly so, a line
back of each segment white. Legs also with an inconspicuous net-
work of fine dark lines; femora, patellae, and tibiae dark at distal ends
and tarsus of leg I also blackish.

Carapace widest at level of third sulcus, convexly rounded, abruptly
narrowed a little in front of caudal end and at level of caudal edge of
third legs in front of which the sides converge but moderately, the
anterior corners more oblique.

Carapace crossed by five transverse furrows which are wide, not
suture like, and of which the first is bent back angularly at the middle.
Eye-tubercle much wider than long, concavely depressed from ends
toward middle; a series of four tubercles in a curved line on each
elevated end subparallel with edge of ocellus and several weaker
tubercles immediately mesad of these; the median portion of eye-
tubercle smooth. Anterior margin not elevated or only obscurely
so on each side, the median portion between eye-tubercle and front
margin broadly elevated. At each anterolateral corner a prominent
process, (Plate 2, fig. 3), the border near level of anterior edge of second
coxae with a process projecting caudoventrad and meeting a process
from second coxae. Edges of carapace in its widest part with a
number of rather low conical tubercles, the tubercles across caudal
border more numerous and with two much larger subcylindrical
processes near middle; the more anterior lateral margins essentially
smooth. On areas between furrows a limited number of tubercles
which are larger and somewhat more numerous in caudal than in
anterior region, and in each area a pair of tubercles which are higher

I

192 bulletin: museum of comparative zoology.

and more cylindric than the others; surface of carapace in general
granulate, (Plate 2, fig. 3).

A row of conical tubercles across each abdominal segment. Anal
scutum proximally with smaller granules and tubercles, distally with
a number of larger tubercles, about five of which appear as dentations
on the caudal margin, one at middle and two or three on each side
of it.

First coxae bent forwards at .ends, the second slightly so. Third
coxae but little shorter than the second. Coxa I in both male and
female on caudal side above middle with two conical processes and at
distal end on caudo ventral corner with a longer process which is bent
dorsad; along its antero ventral surface a row of tubercles the most
distal of which appears at the anteroventral corner as a freely pro-
jecting larger conical process; from anterior surface at distal end a
long cylindrical process which bends dorsad close to surface and over-
laps the end of a process springing from anterodorsal surface. A low
broad process from distal end of coxa II near anterodorsal corner; a
row of a few tubercles along anteroventral line and a similar one on
third coxae. Coxa IV at distal end on mesal side with a low rounded
process, (Plate 2, fig. 4-6).

Spiracles clearly exposed.

First joint of mandible strongly elevated above distad of the con-
striction; with a transverse row of cavidally projecting teeth or conical
tubercles across upper border of caudal surface of the dorsal hump;
dorsal surface of hump with fine teeth or granules; second article
rather short and broad with a few hairs at distal end below^ (caudad)
and more in front, (Plate 2, fig. 1).

Pedipalps short, closely flexed ventrad and caudad, not crossed.
Coxa subcylindric. Trochanter narrow at base, expanding distad,
with a broad conical projection beneath. Femur strongly flattened
from side to side, being abruptly very high immediately distad of the
narrow base; a row of teeth along the middorsal line and a series of
longer subcylindric, distall}^ rounded processes along ventral line.
Patella of usual form; bearing small setigerous tubercles above.
Tibia strongly flattened from side to side, especially on ventral half,
clavately increasing in height from base distad; with a number of
small tubercles on dorsal surface proximad. Tarsus subcylindric,
with a low thin keel from ventral surface at proximal end, (Plate 2,
fig.^2).

Femora, patellae, and tibiae of legs I to III finely tubercular or
granular, the metatarsi more abruptly so proximally and especially

chamberlin: the arachnida. 193

on legs III. Leg IV in female similar to leg III; in male with a series
of much stronger teeth along dorsal or mesodorsal line and on ventral
surface at distal end with a series of five conspicuous long processes of
which the second is longest and is bent at distal end, the others being
straight and decreasing in length distad; the patella with a stout
spine above at proximal end. Tarsus I with five segments; II with
thirteen segments; III with six segments, of which the last three are
abruptly more slender; IV with seven segments of which three are in
the second division; tarsi III and IV clothed with fine hair on ventral
surface.

Length of male type 6 mm.; greatest width of carapace, 4.5 mm.;
length of leg I cir. 1 mm.; of leg II, 2Lo mm.; of leg III, 13 mm.;
of leg IV, 8 mm.

Length of female paratype 6 mm.; greatest width of carapace
4.5 mm.; length of leg I, 10.5 mm.; of leg II, 20 mm.; of leg III,
14 mm.; of leg IV, 18 mm.

Localities. — Santa Ana, 3,000 feet, August 4. (Type, M. C. Z.
133, male; parat^'pe, M. C. Z. 34, female). San Miguel, 6,000 feet,
July. (M. C. Z. 135, one female).

PHALANGIIDAE.

LlOBUNUM MONTICOLA, Sp. nOV.

Plate 5, fig. 9; Plate 6, fig. 1-3.

Body above brown, a paler longitudinal median stripe which
narrows to a point at caudal end of abdomen, this stripe on anterior
portion of abdomen embracing a deep black, sharply defined median
stripe. Eye-tubercle black. Ventral surface of body yellowish.
Trochanters all deep black and coxae dusky distad; patellae also
solid black and tibiae black at distal ends.

Body blunth^ rounded behind, the abdomen broadest across caudal
end; carapace narrowed markedly cephalad. Entire surface of body
above and below densely granular, the granules conically acutely
pointed. Abdomen clearly set off from cephalo thorax by a suture;
the two segments at end distinct.

Eye-tubercle high, its anterior and posterior faces subvertical or a
little constricted at base; a curving line of acute spinous points

194 bulletin: museum of comparative zoology.

mesad of each ocellus and curving caudoectad and ventrad on the
caudal surface of tubercle, (Plate 6, fig. 1).

First joint of mandible nearly smooth, a little roughened over dorsal
surface; hairs above few and straight, more on mesal surface. Second
joint also smooth except for a light dorsal roughening; with a few
longer hairs at distal end above and laterally with a more numerous
patch on mesal side distad, (Plate 6, fig. 2, 3).

Pedipalps moderately long. Femur ventrally along ectal and
similarly along mesal side with a row of acute teeth ; also with a patch
of teeth above and laterally at distal end. Dorsoectal and dorso-
mesal surfaces of patella with fine teeth which are largest and most
numerous on the dorsomesal process, a narrow dorsal stripe free from
teeth or nearly so. Tibia with teeth ventrally and especially laterally
and dorsolaterally, a middorsal stripe free from teeth. The long
slender tarsus free irom teeth but with numerous short fine and a few
coarser short hairs.

Legs with numerous fine teeth which become smaller and less con-
spicuous in going distad.

Length of body 3.8 mm. Length of palp cir. 4.5 mm.; of leg I,
32.5 mm.; of leg II, cir. 80 mm.; of leg III, 41 mm.; of leg IV, 5L5 mm.

Localities.— Paltaybamba, 5,000 feet, August 27. (Type, M. C. Z.
136). Santa Ana, 3,000 feet, August 3. (Paratype, M. C. Z. 137,
one specimen).

LioPAGUS,^ gen. nov.

Eye-tubercle weakly longitudinally furrowed; wholly smooth,
bearing no teeth or spines or other processes.

Body essentially smooth.

Scutum unarmed.

First joint of mandible at base with the usual ventral process.

Patella of pedipalp with a moderate inner apophysis. Claw dis-
tinctly pectinate.

A one pointed process on dorsal margin of distal end of all coxae.
Legs thin; moderately long. Femur II with two false sutures; III
with none and IV with one.

Genotype. — Liopagus simplex, sp. nov.

Related to Prionostemma Pocock in the presence of false sutures
only on second and fourth femora. From that genus it is most e'asily
distinguished in having on femur II only two sutures or joints instead
of the three uniformly present in Prionostemma.

I Xcios. smooth, Trdyos, a peak.

chamberlin: the arachnida. 195

LiOPAGUS SIMPLEX, sp. nov.
Plate 5, fig. 4-8.

Body above brownish grey, darker caudad; similar below with
a darker brown or dusky stripe across each abdominal segment.
Femora toward distal end each with a conspicuous deep black annu-
lus. Legs brown; patellae and the distal ends of tibiae darker.

Body nearly parallel sided from level a little back of eye-tubercle
to one a little caudad of ends of coxae IV, then abruptly narrowing to
caudal end and semicircularly rounded in front. Obscurely granular
in a line across border of segments.

Eye-tubercle as wide as high. Viewed from behind appearing
conspicuously constricted at base, the front face subvertical, not so
convex as the caudal. Tubercle wholly smooth; slightly depressed
longitudinally along middle line, (Plate 5, fig. 4, 5).

All joints of legs free from tubercles but with scattered very fine
teeth on femora and a large one at distal end above on femora and
patellae. Coxae with a single slender process from distal end
above.

First joint of mandible smooth, with the process below as usual ; a
few short hairs above and a series of them along a meso ventral line.
Second joint small; with a few short hairs above, glabrous beneath,
(Plate 5, fig. 6).

Femur of pedipalp between two and three times as long as the
patella without its apophysis; with a series of short stout hairs be-
neath and smaller, less erect ones on dorsal surface but with no teeth.
Patella with apophysis moderately slender, cylindric, more densely
clothed with hair. Tibia also with more numerous short stiff hairs;
a slight rounded process at distomesal corner, this clothed with hairs
as elsewhere on joint. Tarsus with numerous fine short appressed
hairs and fewer shorter, stiff erect hairs. Claw pectinate, (Plate 5,
fig. 7, 8).

Length dr. 3.5 mm. Leg I missing. Length of leg II cir. 42 mm. ;
of leg III, 2L5 mm. ; leg IV broken off at tip.

Locality.— Huadquina, 5,000 feet, July 30. (Type, M. C. Z. 138).

196 bulletin: museum of comparative zoology.

ARANEIDA.

AVICULARIIDAE.

Hemirrhagus peruvianus, sp. nov.
Plate 6, fig. 4-10; Plate 7, fig. 1-2.

Integument of carapace, chelicerae, femora of legs and palpi,
sternum and labium dark chocolate-brown, the other parts of legs
and palpi lighter brown, in part of dilute reddish tinge; sternum and
femora of legs beneath sometimes lighter than above. Integument
of abdomen brown, paler beneath than al:)ove; above with a small
light area toward anterior end. Carapace covered with a woolly coat
of russet-brown hair of a more or less coppery lustre, and some longer
grey ones at caudal edge especially. The legs are thickly clothed
with short brownish hairs with deep black ones more sparsely inter-
mixed; the numerous longer bristles are dark proximally, becoming
grey, grey or white distad, these forming rather indistinct longitudi-
nal streaks on the femora and patellae. Abdomen densely clothed
with brown hair of a distinctly coppery lustre, that of anterior surface
black and forming a distinct black area; longer bristles of dorsal
surface grey distally; sometimes dark hairs form a larger element on
the ventral surface and this sm-face then appears dark grey or blackish
instead of copper-brown as more usual. Fringes of endites and cheli-
cerae orange proximally, lighter distad.

Eye-tubercle sharply limited; of moderate height. Eye-area less
than twice as wide as long in both male and female (67 : 37) ; rows
subequal or the posterior one slightly longer than the anterior (up to
ratio 67:65). Anterior row of eyes in dorsal view a little procurved;
in anterior view rather strongly procurved. Anterior median eyes
with diameter not much more than half that of the laterals {cir. 6:11),
a little more than their radius apart, but decidedly less than their
radius from the laterals. Posterior row of eyes conspicuously' re-
curved. Posterior laterals smaller than the anterior laterals from
which separated by less than their radius. Posterior medians three
fourths the diameter of the laterals ; mesal side more curved than the
ectal, more or less angulate anteroectally, (Plate 6, fig. 4).

Head low, only slightly rising in front of the thoracic groove;

chamberlin: the arachnida. 197

highest some distance eaudad of eyes. Thoracic fovea some distance
back of middle of carapace; straight, transverse.

Sternum subequal in length and breadth, being sometimes slightly
longer than wide (male) and sometimes slightly wider than long
(female). Moderately convex. Sigillae submarginal.

Labium a little wider than long. Spinules in a transverse band of
about four rows, irregularly and closely arranged.

All tarsi densely scopulate; the scopulae of the anterior tarsi
divided by a narrow setose line, those of the posterior tarsi by a broad
setose band which, however, is narrow^er than the joint. Anterior
metatarsi scopulate mostly more than half way to base, the posterior
metatarsi scopulate at distal ends only. Hair on anterior surface of
coxa I both above and below suture moderately long, in part prone,
with in addition a number of finer and shorter straight hairs which
tend to be somewhat clubbed at tip; no spinescent bristles. Paired
claws bearing from three to four moderate teeth, commonly three or
two being fully developed with one or two appearing as mere points,
(Plate 7, fig. 1, 2). In addition to the scales of the ordinary type
occurring on dorsal surface of tarsi, (Plate 6, fig. 5) occasionally one of
the second type, (Plate 6, fig. 6), is to be seen.

Metatarsus I ventrally with an apical and a subbasal spine, a long
one also on anterior surface. Tibia of male with three spines along
ventrocaudal line and a fourth a little more dorsad; or with spines
as many as seven to none, there being on caudoventral surface three
l)asal, two submedian and one apical, and on anteroventral a pair
toward apex and one lateral surface (male from Huadquina); the
number may differ on the right and left legs of same specimen.
Patella armed with a spine on caudal surface (male) . Metatarsus IV
with three pairs of ventral spines and two on anterior and also on
posterior surface (female) or with five or six on each lateral surface
(male).

Inferior spur of tibia I of male longer than the superior, moderately
curved and bearing a large stout black spine inserted on its dorso-
caudal surface. The superior spur bearing a similar stout spine on
mesal surface, (Plate 6, fig. 7).

Tibia of male palpus thicker proximally than patella or femur,
narrowing distad ; a narrow ridge on mesal side from middle distad,
elevated at distal end into a low tubercle. Tarsus short, bilobate
as usual. Palpal organ with spine in lateral view appearing at right
angles to the main axis of bulb, narrowing distad, with apex slender
and acute, below tip with a short, stout, subtriangular spur. In

198

bulletin: museum of comparative zoology.

anterior view the tip of process is seen to curve rather abruptly ectad,
the spine occurring at the bend on the convex side, (Plate 6, fig. 10).

Male (Tincochaca) . Length, 15 mm. Length of cephalothorax,
7.3 mm.; width, 6.0 mm.

Leg I
Leg II
Leg III
Leg IV

fem.
6.2 mm.
6.0
5.1

7.2

tib.+pat.
9.0mm.
7.0
6.2

8.5

met.
4.0 mm.
3.6
4.6
7.0

tar.
3.5 mm.
3.0
3.0
4.0

total

24.7 mm.
19.6
18.9
26.7

Female {Tincochaca). Length, 17 mm. Length of cephalothorax,
8.5 mm.; width, 7.0 mm.

Leg I
Leg II
Leg III
Leg IV

fem
5.8 mm.
5.1
4.6
6.0

tib. + pat.
7.8 mm.
6.2
5.7
8.0

met.
3 . 3 mm.
2.7
3.3
5.4

tar.
3 . 2 mm.
2.6
3.2
3.2

total

20.1 mm.
16.6
16.8
24.6

Localities. — Huadquina, 5,000 feet, July. (Type, M. C. Z. 139,
male; para type, immature female 324). Tincochaca, 7,000 feet,
August. (M. C. Z. 140, male and female). Santa Ana, 3,000 feet,
August. (M. C. Z. 141, one female). Urubamba, 9,500 feet, July
18. (M. C. Z. 325).

Hemirrhagus major, sp. nov.

Plate 7, fig. 3-8.

Integument of carapace, sternum, legs, and palpi dark brown or
chocolate to chestnut; that of labium and endites reddish brown or
chestnut. Hairs of carapace somewhat wavy, moderately thick,
light brown of more or less dark golden or bronze lustre. Hair of
chelicerae dense, light brown or bronze lustre. Sternum clothed
with a dense coat of short brown hair with black ones sparsely inter-
mixed; bristles dark proximally, rufous distally, mostly of moderate
length. Legs clothed with a coat of short brown hair with darker
dusky or blackish hairs intermixed and especially abundant on dorsal
surface of femora, the brown hair in part of shining brown lustre; a
narrow stripe of grey or white hairs across ends of joints above ; two
longitudinal stripes on patellae above formed by hair free and bristle

chamberlin: the arachnida. 19^

free areas over which the lighter hairs unmixed with dark extends from
each side. Bristles of legs shorter proximad, becoming longer and
more numerous especially on tibiae ventrally and on metatarsus above
as well as laterally and ventrally; bristles dark proximally, becoming
light rufous distally, rather coarse. Integument of abdomen above
dark, somewhat dusky brown, ventrally lighter brown, with several
small obscure white spots on each side; bristles of dorsum subdense,
forming a light brick-red area.

Eye-tubercle black, well defined, moderate in height, highest
between posterior median eyes, more strongly convex anteriorly than
posteriorly. Eye-area trapeziform, the posterior row being distinctly
longer than the anterior, the difference typically distinctly more
marked than in peruvianus (up to 85: 78); area from a little less than
half as long as greatest width to a little more thus also differing from
peruvianus. Anterior row of eyes from moderately procurved to
nearly straight, much more procurved in anterior view than in dorsal.
Anterior median eyes with diameter two thirds that of laterals ; about
two thirds their diameter apart, less than their radius from the laterals.
Posterior lateral eyes about three fourths the diameter of the anterior
laterals, less than their radius from the latter. Posterior medians
pyriform in outline, being narrowed to a point caudad and widely
rounded cephalad, (Plate 7, fig. 3).

Head moderately elevated, in outline slightly convex, highest a
little caudad of eye-tubercle. Fovea straight or vaguely procurved,
short and deep.

Sternum longer than wide. Posterior sigillae about their length
from margins. Median ones submarginal ; anterior marginal.

Mesal margin of furrow or chelicera bearing a row of fourteen
teeth.

Labium wider than long, truncate distad. Spinules in a narrow
band (about four rows deep) across distal end, the spinules not dense.

Spinules on proximal end of endite rather numerous.

Anterior surface of coxa I with longer in part semiprone hairs and
some of the short, slender, distally obtuse or clubbed hairs such as
are present in peruvianus but in addition below the suture with
numerous dark, strongly chitinized spiniform bristles, these more
numerous distally, some rather more slender ones also occurring
above suture.

Tarsal claws four or five (anterior and posterior of leg I respectively) ,
proportionately shorter, more slender and more uniform than in
peruvianus. On the posterior claws, which are longer and propor-

200 bulletin: museum of comparative zoology.

tionately more slender, one or two extra points may occur in addition
to the five ordinary teeth, (Plate 7, fig. 5-7).

Scopulae of anterior tarsi divided by a narrow setose line, the
posterior ones by a broader one, but this much narrower than the
joint. Anterior metatarsi scopulate well toward base; posterior
with scopular hairs only distally and there sparse or absent. Meta-
tarsus I (female) ventrally with a pair of small apical spines and a
single one proximad of middle. Tibia I (female) also ventrally with a
pair of spines at distal end and a single submedian one ; in the male
with seven spines on ventral and ventrocaudal surface and two on
anterior surface. Metatarsus IV with fifteen to eighteen spines
irregularly arranged; tibia IV with eight to ten.

In the male the metatarsus of leg I is abruptly much more slender
than the tibia and is strongly bowed dorsad. The spurs of the tibia
are elevated on a conspicuous common basal process standing at right
angles to the article; the inferior process is not much longer than the
superior but it is proximally stouter being narrowed distad, moderately
curved toward end, bearing on dorsal surface a dark, stout, acute
process attached near apex and corresponding to the larger one of
penivianus; superior process directed more cephalad, more uniform
in diameter, curved a little mesad toward tip, the spine rather slender,
closely applied to surface excepting at tip which is divergent, (Plate
7, fig.^4).

In the male palpus the tibia is thicker than the patella or femiu",
narrowing distad, a narrow ridge-like thickening mesoventral edge
from middle distad and elevated at distal end into a low tubercle,
ventrally with a dense growth of long bristles. Tarsus short, bilobate
as usual, the mesal lobe at distomesal corner adjacent to bulb extended
into a blackish, densely chitinized tubercle. Spine of bulb propor-
tionately much longer than in pcruvianus, basal part with axis corre-
sponding with long axis of bulb, curving semicircularly first ventrad
and then forwards as shown in the figure, a slight tooth a little distad
of middle, in anterior view the process bends first somewhat ectad.

Male (Type, Cuzco Valley). Length, 29 mm. Length of cephalo-
thorax, 14 mm.; width, 12.3 mm.

fem.

tib. + pat.

met.

tar.

total

Leg I

10.8

mm.

14 mm.

8.1 mm.

5.4 mm.

38.3 mm

Leg II

10

12.2

7.8

5

35.0

Leg III

9

11.2

8.3

4.5

33.0

Leg IV

11.2

14.2

11.2

5.S

42.4

Tibia

I, 7.1 mm.

Til.ia IV,

S.9 mm.

chamberlin: the arachnida. 201

Female (Urubamba). Length, 37 mm. Length of cephalothorax,
15.8 mm.; width, 13.1 mm. (In a female from Cuzco Valley the
cephalothorax is proportionately broader, the ratio of length and
breadth being 16:14.3). '

fern. tib.+pat. met. tar. total

Leg I 11.8 mm. 1.5.3 mm. 7.5 mm. 5.6 mm. 40.2 mm.

Leg II 11.0 13.0 7.1 5.3 36.4

I^glll 10.0 11.8 8.8 5.5 36.1

Leg IV 12.5 15.5 13.0 6.4 47.4

Localities. — Cuzco Valley. (Type, M. C. Z. 142, one male; para-
type, M. C. Z. 143, one female) E. D. Flint coll. Urubamba, 9,500
feet, July. (M. C. Z. 144, one female).

Hemirrhagus sp.

Locality. — Urubamba, 9,500 feet, July. (M. C. Z. 326, several
immature females).

Eurypelma aymara,^ sp. nov.

Integument of carapace when dry black or nearly so, when wet
appearing of slight chestnut tinge. Sternum chestnut, darker cepha-
lad. Labium nearly black. Endites chestnut. Legs proximall\'
blackish like the carapace, becoming slightly more chestnut distad.
Carapace clothed with a coat of sandy grey and light brown hair
of a golden lustre. Hair of femora dusky brown of rufous tinge, that
of more distal joints with a larger proportion of grey intermixed;
bristles numerous and long, of rufous or rust color proximally, paler,
greyish yellow distad. Hair of venter of abdomen brown and black
intermixed, chiefly the latter. Dorsum with a thick coat of long
rufous bristles.

Pars cephalica moderately high, highest a little caudad of eye-
tubercle.

Eye-tubercle rather high, sharply limited ; highest along the median
longitudinal line which descends a little from between the posterior
median eyes cephalad. Eye-area considerably less than twice as
wide as long (25:16). Anterior row of eyes slightly shorter than

' Aymara, a tribe of the indigenes of Peru.

202 bulletin: museum of comparative zoology.

the second (25 : 24) ; in dorsal view considerably procurved, a line
tangent to the anterior edges passing through the anterior third of
the lateral eyes. Anterior median eyes much smaller than the laterals,
their diameters being about as 2 to 3; medians three fourths their
diameter apart and near the same distance from the laterals. Pos-
terior lateral eyes of nearly same size as the anterior laterals from which
they are separated by less than their radius. Posterior median eyes
much smaller than the laterals; oblong, with sides straight and ends
more rounded. A line tangent to caudal edges of the two posterior
eyes of each side intersects the anterior median eye of opposite side.

Metatarsus I and II scopulate very nearly to base; metatarsus III
scopulate over distal third; metatarsus IV scopulate only at distal
end.

Female. Length 38 mm. Length of cephalothorax, 21.2 mm.;
width, 18.1 mm. Length of pars cephalica, 13.8 mm.

fem. tib.+pat. met. tar. total

Leg I 15 mm. 20.9 mm. 10 mm. 6.5 mm. 52.4mm.

• Leg II 13.8 18.1 9.9 6.1 47.9

Leg III 12.3 16 11 6.1 45.4

Leg IV 15.7 20 15.6 6.8 58.1

Locality — Feru: Aymas, Dr. W. H. Jones. (Type, M. C. Z. 145,
one female).

This seems to be the only true Eurypelma thus far recorded from
Peru.

DiPLURA MONTICOLENS,^ sp. nOV.

Plate 7, fig. 9-10.

Cephalothorax dusky brown, the margins of carapace and the eye-
area darkest. Edges of sternum darker than median portion. Cheli-
cera palpi and legs testaceous. Integument of abdomen dark above,
light beneath; a series of three pairs of obscure light stripes extend-
ing from near dorsal median line obliquely ventrocaudad across sides.
Spinnerets testaceous, in part dusky, especially at ends. Hair of
carapace light brown, those of sternum sparse, darker. Bristles of
abdomen dark rufous brown, the shorter hairs mostly lighter.

Eye-area twice as wdde as long or very nearly so (67:34). Eye-
tubercle highest beneath anterior median eyes, behind which it
descends almost immediately and also slopes laterally from this

' iVIons, mountain, incolens, inhabiting.

chamberlin: the arachnida. 203

point, the eyes being set obliquely to the horizontal plane. Anterior
row of eyes in dorsal view a little procurved, in anterior view strongly
recurved. Anterior median eyes inclusive of rim with diameter more
than three fourths as great as that of the anterior laterals, less than
their radius from each other and closer to the laterals; exclusive of
rim the median eyes are less than their diameter apart and are near
their radius from laterals. Anterior laterals very oblique, the anterior,
narrow end lying in front of the ectal edge of the median eye. Pos-
terior lateral eyes clearly smaller than the anterior laterals (diameters
nearly as 3:4). Posterior median eyes smaller than laterals, closer
to them than to the anterior medians. A line tangent to the caudal
edges of posterior eyes on each side is tangent to or passes near caudal
edge of anterior median eye of opposite side, (Plate 7, fig. 9, 10).

Cephalothorax broad and low anteriorly (the width across eye area
to greatest width about as 47:76). Thoracic fovea small, oval,
transverse, situated considerably behind middle of length of carapace.

Labium much wider than long, with anterior margin mesally in-
curved. On anterior edge with two spinules present, a scar in t>T)e
seeming to indicate the normal presence of a third one.

Endites at proximal corner with a small group of few (8-10) spinules.

Sternum as wide as long. The sigillae marginal.

Tarsi of legs slender, all conspicuously curved, those of the anterior
pairs most strongly so. Paired claws each with a double row of mostly
eight teeth. Inferior claw smooth, slender. Metatarsus I with three
pairs of ventral spines; femur I with a spine at distal end on ante-
rior side and above and another smaller one on the posterior side;
other femora with corresponding spines, the posterior one increasing
in size in going caudad. Posterior metatarsi much more strongly
spined. None of the tarsi spined.

Posterior spiimerets as long as abdomen; slender; articles sub-
equal, the first thickest, the distal one most slender, narrowing distad.
Anterior spinnerets short, distally acuminate.

Female. Length 8.4 mm. Length of cephalothorax 3.4 mm.;
width, 2.8 mm.

fern. tib.+pat. met. tar. total

Leg I 3.4mm. 4.5 mm. 2.9 mm. 2.2 mm. 13.0mm.

Leg II 3.1 4.1 2.8 2.2 12.2

Leg III 2.9 3.8 2.8 2.0 11.5

Leg IV 4.0 5.0 4.0 2.5 15.2

Loca%. — Huadquina, 5,000 feet, July. (Type, M. C. Z. 146,
female).

204 bulletin: museum of comparative zoology.

Brachythele keithi, sp. nov.
Plate 7, fig. 11.

Integument of cephalothorax and legs light chestnut-brown; that
of the chelicerae dark chestnut; integument of abdomen and spin-
nerets clear brown. Pubescence of carapace brown of golden lustre ;
that of abdomen dense, of similar golden brown color or of more
coppery lustre.

Eye-tubercle strongly convexly elevated between the posterior
median eyes. Eye-area twice as wide as long (45:22); anterior row
a little shorter than the posterior (45:42). x\nterior row a little
procurved. Anterior median eyes a little more than their radius
apart, closer to the lateral eyes; diameter but little more than two
thirds the longer diameter of the lateral eyes (ratio near 23:31).
Posterior lateral eyes slightly longer than the anterior (33: 31). Pos-
terior median eyes elongate, narrowed caudad, length to width
about as 3: 5 — smaller than lateral eyes (lengths as 25: 33), nearer to
these than to anterior medians. A line tangent to caudal edges of
posterior eyes on each side in type passing through anterior third of
opposite anterior median eye, (Plate 7, fig. 11).

Labium clearly wider than long; distal margin mesally incurved.
Spinules none.

Spinules on proximal corner of endites short and stout, constricted
near or a little above base, distally expanded and rounded.

Thoracic fovea recurved.

Sternum longer than w4de; gently convex. The sigillae at \e\e\
between second and third legs most distinct, marginal. Sternum
with numerous fine tubercles from each of which arises a hair.

Paired claws with teeth in two rows; teeth of each row mostly
seven or eight in number. Unpaired claw smooth. Tarsal scopulae
dense, extending to base; none divided by a setose line or band.
Spines long and moderately stout, black.

Posterior spinnerets more than two thirds as long as the abdomen
ventrally ; the three joints subequal in length, decreasing in diameter
distad, the second being more slender than the first and the third
than the second. Anterior spinnerets twice or more their diameter
apart at base.

Female. Length, 30 mm. Length of cephalothorax, 14 mm.;
width, 12 mm.

chamberlin: the arachnida. 205

fern. tib. + pat. met. tar. total

Leg I 12 mm. 15 mm. 9.5 mm. 8 mm. 44.5 mm.

Leg II missing

Leg III 10 12.5 10 7.5 40.0

Leg IV missing

Length of spinnerets 12.2 mm.

Locality. — Huadquina, 5,000 feet, wet season. (Type M. C. Z.
147, one female).
Named for Minor C. Keith of New York.

Brachythele incursus, sp. nov.
Plate 7, fig. 12.

Integument of cephalothorax and legs brown, that of carapace
and femora of legs above deeper in color; chelicerae above blackish
brown. Integument of abdomen above dusky brown; below light
brown. Spinnerets blackish. Hair of carapace and abdomen gold-
en brown. Hair of sternum blackish. Fringe of chelicerae rufous.

Eye-area less than twice as wide as long (57:30). Anterior row of
eyes of same length as posterior. Anterior row straight. Anterior
median eyes slightly more than their radius apart (6:5.5), nearer to
the laterals; diameter but little more than half that of lateral eye
(11:20). Posterior lateral eyes clearly smaller than the anterior
laterals (ratio of diameters 3:4), lateral eyes nearly contiguous as
usual. Posterior median eyes elliptic, about three fourths as wide
as long, four fifths as long as the laterals. A line tangent to caudal
edges of two posterior eyes of each side passes clearly caudad of
anterior median eye of opposite side, (Plate 7, fig. 12).

Labium wider than long in the ratio 9.5:6.8; distal margin a
little incurved. A row of low stout, rounded spinules across apical
portion.

Endites at proximal corner with the patch of spinules as usual.

Thoracic fovea a little recurved.

Sternum longer than wide. Hairs not dense, black, longer about
margins.

Paired claws of legs each with two rows of from six to eight teeth.
Single claw smooth. In the stage of growth represented by the
type the anterior tarsi are not densely scopulate and the metatarsi
are scopulate only distally; the posterior tarsi are scopulate only

206 bulletin: museum of comparative zoology.

distally, ordinary bristles covering the proximal portion while the
metatarsi are not at all scopulate. The tarsi are not armed. Patel-
lae, tibiae, and metatarsi armed with stout black spines as usual.

Spinnerets short, about one half the length of the abdomen on ven-
tral side, much less slender relatively than in keithi: each joint atten-
uated distad. Black in color.

Female {not fully mature). Length, 13 mm. Length of cephalo-
thorax, 6.2, mm.; width, 5 mm.

fern. tib.+pat. met. tar. total

Leg I

4.3 mm.

5.6 mm.

2.8 mm.

2.0 mm.

14.7 mm

Leg II

4.0

4.7

2.3

1.8

12.8

Leg III

3.2

4.1

2.2

1.8

11.3

Leg IV

4.3

5.5

3.2

2.0

15.0

Loca/i/?/.— Huadquina, 5,000 feet, July. (Type, M. C. Z. 148, one
young female).

The female described above is apparently not fully grown; but the
differences it presents especially in eye sizes and relations added to
the differences in proportions of joints of legs, the presence of spinules
on the labium and especially the pronounced differences in form and
relative length of spinnerets render it impossible to regard it as identi-
cal with the preceding species which comes from the same locality.

ULOBORIDAE.

Orinomus,^ gen. nov.

Posterior eyes small, subequal, the series strongly recurved, nar-
rower than the cephalothorax, the medians farther apart than from
the laterals. Anterior eyes in a procurved series; medians farther
from laterals than from each other. Area of median eyes much wider
than long and much narrower in front than behind.

First leg much the longest, the others in order IV, II, III. Meta-
tarsus I clearly shorter than tibia + patella I. Calamistrum not
reaching distal end of metatarsus IV. .

Cribellum narrow; entire.

Abdomen subglobose; bigibous above, not produced or acuminate
behind; spinnerets not quite terminal.

' opeii-o/ios, mountain ranging.

chamberlin: the arachnida. 207

Genotype. — Orinonius lamprus, sp. nov.

Apparently closest to Uloboms, a genus occurring widely in this
and adjoining regions as well as elsewhere.

Orinomus lamprus/ sp. nov.
Plate 8, fig. \-4.

Carapace nearly black, a narrow median longitudinal line yellow-
ish, the clypeus also paler. Carapace clothed with grey hair. Ster-
num black; a pale median mark. Coxae dusky beneath. First
legs with femora, patellae, and tibiae black, with a few small light
spots; metatarsi and tarsi yellow. Other legs yellow, with dusky
annul i. Abdomen somewhat yellowish white, finely spotted with
black, a dark median longitudinal line over entire length and a
deeper, more distinct one on each side from the hump caudad; venter
with a geminate black stripe along the middle.

Posterior row of eyes conspicuously recurved, a line tangent to
caudal edges of median eyes passing through the anterior third of
laterals; median eyes three times their diameter apart, half as far
from the laterals which are nearly of same size. Posterior lateral eyes
as far from the anterior laterals as from the posterior medians (once
and a half their diameter), the anterior laterals with diameter two
thirds as large. Anterior median eyes only a little more than their
diameter apart but more than once and a half their diameter from the
laterals; row conspicuously procurved. Area of median eyes wider
than long nearly in ratio 5:3; twice as wide behind as in front. Cly-
peus slanting ventrocephalad ; about 6qual in w^idth to diameter of
median eye.

Labium wider than long; subtriangular, the sides being straight
and meeting at an angle, (Plate 8, fig. 4).

Sternum caudally produced as a short tongue between the separated
posterior coxae, the tongue rounded at end. More than twice as
wide as long (20:9), (Plate 8, fig. 2).

Female. Length 3 mm.

fem. tib. + pat. met.

tar. total

Leg I

1.6 ram. 1.6 mm. 1.2 mm.

.6 mm. 5.0 mm.

Leg 11

1.0 1.1 1

.3 3.4

Locality.—
Tiale).

-Urubamba, 9,500 feet, July.

* \aiivp6s, distinct.

(Type, M. C. Z. 149, one

208 bulletin: museum of comparative zoology.

DICTYNIDAE.

Amaurobius, sp. a.

An immature male and female from Ollantaytambo (9,000 feet,
July. M. C. Z. 150). The abdomen is marked with a distinct median
longitudinal black stripe of narrowly hastate form and with indented
edges and with a bright white spot each side of middle of its length.
Carapace with head dark brown, elsewhere dusky. Sternum dark,
dusky brown.

Two immature males apparently of same species from Urubamba
(9,500 feet, July. M. C. Z. 151). They are darker throughout, the
sternum being nearly black and the abdomen so dark that the dorsal
median black line is almost obliterated ; but the two white spots are
conspicuous.

Amaurobius sp. b.

One immature female with lateral eyes nearer together than usual
in the genus. Carapace yellow with each eye enclosed in black.
Sternum light brown, a little dusky. Legs yellow, tibiae very ob-
scurely ringed. Abdomen grey of olive cast; above conspicuously
marked with a series of dark chevron lines from middle caudad, all
of which are broken at middle.

ioca%.— Urubamba, 9,500 feet, July. (M. C. Z. 152).

Amaurobius platei Cambridge.

Journ. Linn. soc. London, 1898, 27, p. 18, pi. 2, f. 3.

One specimen probably this species was secured at Ollantaytambo,
9,000 feet in July. (M. C. Z. 166).

Aymarella,^ gen. nov.

Pars cephalica relatively broad, convex, and high.

Anterior row of eyes substraight or slightly procurved, the medians
much smaller than the laterals, farther from the laterals than from
each other. Posterior row of eyes recurved ; subequal, the medians

' Diminutive of Aymara.

chamberlin: the arachnida, 209

farther from the laterals than from each other. Area of median eyes
distinctly narrower in front than behind. Lateral eyes close together,
not more than their radius apart.

Clypeus narrower than the anterior lateral eyes.

Chelicera with lower margin short, but little oblique, armed with
two (or three) small teeth; upper margin with three teeth of which
the median one is largest.

Labium longer than wide, narrowed distad, obtuse.

Length of legs in order I-IV, II, III. Robust. Moderately spined.
Metatarsi much longer than the tarsi.

Cribellum bipartite. Calamistrum uniseriate.

Genotype. — Aymarella viunda, sp. nov.

This genus would seem to be closely related to Calleva, a genus
based by Simon upon an Argentine species; but it differs clearly in
having the lateral eyes much closer together, in having the legs
spined, and in having the metatarsi much longer than the tarsi.
Callevopsis, a Chilean genus, differs, e. g., in having the eyes of the
anterior row equidistant and the lower margin of the chelicera armed
with four teeth.

Aymarella munda,^ sp. nov,
Plate 9, fig. 1-5.

Carapace brown to fuscorufous with darker lines radiating from
thoracic furrow. Sternum fuscous to nearly black; clothed with
numerous black hairs. Legs testaceous, typically of rufous tinge
distad on anterior pairs; femora darker, dusky to nearly black.

Palpi fuscous. Chelicerae nearly black. Labium and endites very
dark, pale across tips. Abdomen olive-grey; without distinct mark-
ings but in some with a median hastate mark very vaguely outlined
in dark and followed behind by several faint light colored chevron
lines, the lines very thin, and also a small light dot each side of middle
median mark.

Posterior row of eyes decidedly longer than the anterior (3.4:2.9),
a little recurved; median eyes smaller than laterals (7:9), L7 their
diameter apart, a third farther from the laterals. Anterior row of
eyes viewed from above with line of apices a little recurved; viewed
from in front the line of centres is a little procurved; median eyes

' mundus, fine, neat.

210 bulletin: museum of comparative zoology.

with diameter about three fifths that of the laterals ; laterals on well-
marked tubercles. Area of median eyes narrower in front than behind
in ratio 10:13. Lateral eyes usually their radius apart, the distance
being somewhat variable.

Pars cephalica large, conspicuously elevated, highest immediately
behind eye-area, smooth and shining, (Plate 9, fig. 1).

Sternum longer than wide in about ratio 5:4. Caudal process
short, narrow, (Plate 9, fig. 2).

Labium grooved across base as usual; narrowed conspicuously
from above groove distad; distally obtuse, (Plate 9, fig. 3).

Tarsus of palpus armed with a claw and a number of spines ; other
joints unarmed.

Femur I armed with a single spine on anterior surface toward distal
end; other femora unarmed. Patellae unarmed. Tibia III armed
on anterior surface in subapical position with one spine; other tibiae
unarmed. Metatarsi I and II armed ventrally with three pairs of
spines; IV also with spines which are not distinctly arranged in pairs
but are more distributed along surface of joint.

Superior margin of furrow of chelicera armed with three teeth of
which the median is largest; lower margin nearly transverse and with
two rather small teeth.

Abdomen broadly subelliptic in outline.

Cribellum bipartite, (Plate 9, fig. 5).

Feviale {San Miguel). Length 8 mm. Length of cephalothorax
3.2 mm.; width, 2.2 mm.

fern. tib. + pat. met. tar. total

Leg I 2.5 mm. 3 mm. 2 mm. 1.2 mm. 8.7 mm.

Leg II 2.25 2.7 1.9 1 7.85

Leg III 2.1 2.2 1.5 1 6.8

Leg IV 2.5 3.1 2 1.1 8.7

Localities.— San Miguel, 6,000 feet, July. (Type, M. C. Z. 153
female; paratypes, M. C. Z. 154, females). Torontoy, 8,000feet, July.
(M. C. Z. 155, several females). Lucma, 7,000 feet, August (M. C. Z.
156). Urubamba, 9,500 feet, July. (M. C. Z. 157, one female).
Huadquina, 5,000 feet, July. (M. C. Z. 158, one female). Ollan-
taytambo, 9,000 feet, July. (M. C. Z. 159, one female).

AuxiMUS sp.

A single immature male of a species of this genus was secured at
Lucma in August. (M. C. Z. 160).

chamberlin: the arachnida, 211

AuxiMUS PRODUCTUS, sp. nov.
Plate 8, fig. 5-7.

Carapace dusky brown or fuscous. Caelicerae similarly dusky
over a more reddish or chestnut background. Legs testaceous;
the tibiae, tarsi, and metatarsi of legs I and II dusky, almost black or
tibia with an apical and a broader subbasal dark annulus indistinctly
set off; tibiae of legs III and IV with apical and subbasal dusky annu-
lus distinct and metatarsus of these legs with an apical and a broader,
less distinctly limited subbasal dark band. Patellae dusky. Femora
without rings or distinct marks, usually dusky beneath proximally.
Tibia and tarsus of palpus dusky or almost black like the distal articles
of anterior legs. Sternum dusky brown. Labium and endites darker
than sternum, pale across tips. Abdomen above with background
of greyish yellow of slight olivaceous tinge; at base above with a
black median sagittate stripe the acute apex of which is at middle;
this mark is followed caudad by a series of four, successively smaller
and smaller chevron-shaped black lines the angle of the first of which
is at the point of the sagittate mark; chevron marks with ends on
each side united by a longitudinal black line. Dorsum elsewhere with
small black, in part angular, dots, these more numerous on the sides
where they are confluent with enclosed areas dusky and entire surface
thus appearing blackish. A broad median band along venter immacu-
late greyish yellow of slight olivaceous cast like background of dorsum.

Posterior row of eyes considerably procurved, a line through middles
of median eyes being nearly tangent to caudal edges of laterals;
median eyes clearly less than twice their diameter apart (11:7) and
nearly the same distance from the laterals; smaller than the laterals
(7:9). Anterior row of eyes with line of apices as seen from above
very slightly procurved; viewed from in front the row is nearly
straight, the centres of the laterals being slightly lower; median eyes
with diameter half as great as that of laterals, four fifths their diameter
apart, nearly same distance or slightly farther from laterals. Ante-
rior laterals equal to posterior laterals from which they are separated
by a distance not more than equalling half the radius.

Labium about two thirds as broad at base as long; narrowed distad;
apically truncate.

Sternum longer than wide, the length inclusive of caudal process
being to width as 9 : 6.5 and the length only to level of anterior proximal
corner of coxa IV being as 7:6.5.

212 bulletin: museum of comparative zoology.

Abdomen in outline broadly subovate.

Lower margin of furrow of chelicera with six teeth of which the two
farthest from claw are much smaller and the first is smaller than
the succeeding three.

All femora dorsally with a submedian spine and at distal end with
one toward anterior side and there may be a smaller one in corre-
sponding position on posterior side. Tibia I with a submedian pair
of spines on ventral side and a single one on anterior side between
middle and proximal end. Tibia II with a single submedian ventral
spine; and tibiae III and IV unarmed. Metatarsus I with three
pairs of ventral spines.

Femur of palpus with a spine in submedian position above; tibia
with three long, almost bristle-like, spines, two toward mesal side
above and one ectal in position; tarsus with numerous spines, the
proximal ones long and slender, the distal ones shorter and stouter.

Female. Length, 7.5 mm. Length of cephalothorax, 3 mm.;
width, 2.5 mm.

fern. tib.+pat. met. tar. total

Leg I 2.4 mm. .3.25 mm. 2.1mm. 1.1mm. 8.85 ram.

Leg 11 2.2 2.9 2 1.1 8.2

Leg III 2.2 2.2 1.6 1 7.0

Leg IV 2.4 2,9 2 1.25 8.55

Locality.— Tincochaca, 7,000 feet, August. (Type, M. C. Z. 161,
one female).

While a number of species of Auximus have been described from
Ecuador, only one was previously known from Peru, A. funestus
(Keys.), a species differing rather widely from the present one in hav-
ing its posterior row of eyes straight, in the sparser spining of legs,
in the different form of epigynum as well as in color.

DiCTYNA HESPERIA,^ sp. nOV.

Plate 8, fig. 8.

Thoracic part of carapace fuscous, yellow along edges; head testa-
ceous. Legs and palpi yellow, the tarsi in part dusky. Sternum
yellow, faintly dusky. Labium dusky yellow. Abdomen opaque
yellowish white, venter covered with a broad, more or less dusky
band, a more blackish area about the spinnerets; above there is a

' f<riripios, western.

chamberlin: the arachnida. 213

median dorsal accent shaped black stripe with point cephalad and
base at middle; enclosing dorsal area is a dark band of elliptical form
with a cross connection toward caudal end, also united with anterior
band along middle line, and a network of finer dark lines connecting
with the median dark mark elsewhere within the ellipse.

Posterior row of eyes a little recurved ; a little longer than the ante-
rior row; median eyes farther from each other than from the laterals;
laterals equal to the medians. Anterior row of eyes in dorsal view
recurved in such degree that a line tangent to anterior edges of laterals
passes through the posterior fourth of medians; in anterior view the
row is straight; median eyes but little smaller than the laterals, near
their diameter apart and three fourths as far from the laterals. Area
of median eyes somewhat trapeziform, the width anteriorly being less
than that posteriorly in about ratio 5:6. Anterior lateral eyes their
diameter from lower edge of clypeus.

Epigynum with openings widely separated, protected by laminae,
(Plate 8, fig. 8).

Sternum much longer than wide (18:13).

Cribellum undivided as in most species.

Female. Length 3.25 mm. Length of cephalo thorax 1.1 mm.;

width .86 mm.

fem.

tib. + pat.

met.

tar.

total

Leg I

1.38 mm.

1.3 mm.

.9 mm.

.6 mm.

4.18 mm.

Leg II

1.3

1.12

.74

.48

3.64

Leg III

1.2

1

.7

.44

3.34

Leg IV

1.38

1.3

.86

.5

4.04

Locality — San Miguel 6,000 feet, July. (Type, M. C. Z. 162, one
female).

Four species, all based similarly upon females, have been previously
described from Peru. The present species is most readily separated
from these and other South American species through the different
form of the epigynum, the eye relations, and the coloration.

SCYTOTIDAE.

LOXOSCELES RUFESCENS (Dufour).

Scytodes rufescens Dufour, Ann. gen. sci. phys., 1820, 4, p. 203, pi. 77, f. 5.

Numerous specimens of both sexes were taken at Huadquina in
July and August, 5,000 feet. (M. C. Z. 163, 164). This species is
widespread in both hemispheres.

214 bulletin: museum of comparative zoology.

Thomisoides Nicolet.

Nicolet, Gay hist. ChQe. zooL, 1847?, pi. 1 Arachn., 1849, 3, p. 352.

As Plate I of Nicolet's work on spiders in Gay's Histoire de Chile
was issued and in Walekenaer's hands before the publication of volume
four of the Histoire naturelle des Insectes Apteres, the species illus-
trated on that plate and named at the bottom must be dated at the
latest from 1847 rather than from 1849 when the text was published,
since the issuance of figures of this kind with accompanying names
constitutes publication. Thomisoides is used in the legend of the
plate in combination with specific names and must accordingly be
regarded as having been established at the same time. Walekenaer's
genus Sicarius, like his species thomisoides, was based upon the pub-
lished plate of Nicolet's work, and I can see, therefore, no other way
than to drop both Sicarius and thomisoides as clear synonyms.

Thomisoides terrosus Nicolet.
Plate 9, fig. 6-10; Plate 10, fig. 1-3.

Thomisoides terrosus Nicolet, Arachn. pi. 1, fig. 9, Gay's Hist. Chile, 1847?,

Gay's Hist. Chile, 1849, 3, p. 352, Arachn., pi. 1, fig. 9.
Sicarius terrosus Simon, Hist. nat. Araign., 1893, 1, p. 271.

In alcohol the cephalothorax is reddish brown, the abdomen grey-
brown, and the legs and sternum brown of lighter shade than carapace.

Cephalothorax wider than long; width across head decidedly less
than half the greatest width; low; shorter than tibia I but longer than
tibia IV. The margins and the anterior surface of the head bearing
numerous short, stout, spiniform bristles which are distally l)lunt and
toothed like those of the legs.

Sternum almost circular, equal in length and breadth (3.15 mm.)
or very slightly wider than long; depressed behind at middle in front
of caudal margin and giving the appearance of being deeply emargi-
nate though the caudal edge at middle is really convexly rounded;
with two or three rows of stiff bristles along the edge and smaller
bristles arranged on radii with their free ends towards centre of ster-
num; between the bristles short fine hair.

Labium gradually narrowing from base distad; tip obtusely

chamberlin: the arachnida. 215

rounded ; about two thirds as wide at middle as long, length in meas-
ured specimen 1.6 mm., (Plate 9, fig. 8).

Endites very long, arched and meeting in front of labium in usual
way. Endites and labium covered with stiff short hairs like those of
sternum.

Claw of chelicera small ; apparently less strongly curved than usual
in the family; chitinous appendage rather large, overlapping apical
portion of claw, (Plate 9, fig. 7, 9) covered with stiff bristles.

Clypeus very wide, the distance from lower edge to edge of anterior
median eyes being 1,35 mm.

Anterior median eyes rather less than their radius from each other;
six times their diameter from the anterior laterals. Anterior row of
ej'-es seen from above slightly recurved, a line tangent to caudal edges
of median eyes being tangent to anterior edges of the laterals. Lateral
eyes of each side equal in size and clearly larger than the anterior
median eyes, a little less than their diameter apart, (Plate 9, fig. 6).

Palpi lacking claw. Stridulating tubercles of palpus only four in
number in specimen described. The stridulating plate of the cheli-
cera, (Plate 9, fig. 7). Palpi covered with stiif bristles like those of
mandibles, etc.

Claws of legs curved; each with a single row of usually 12-13 teeth,
(Plate 9, fig. 10). The spines and bristles on legs and body with regu-
lar dark longitudinal ribs which under high magnification are seen to
be serrated. The distal end of the spines is acute but that of the
bristles is in the form of a crown of teeth, (Plate 10, fig. 3).

Bristles on legs short and stiff, spinescent, arranged in regular
rows, the spines inserted in line with the bristles. On the femora
are two rows of bristles below^ and three rows above, the sides being
covered with bristles irregularly arranged. Patellae with five less
regular rows above, of which the three median rows are best developed;
below with irregular bristles. Tibiae, metatarsi, and tarsi with two
rows above, two on each side, and two below. Fine short hair be-
tween the rows on all the joints, (Plate 10, fig. 2, 3).

The tibiae of all legs have six pairs of spines beneath and the meta-
tarsi four or five pairs. On the anterior surface of tibiae I and II are
five spines in lower row of bristles and four in the upper; tibia III has
five in the lower and three in the upper; and tibia IV has four in the
lower and one in the upper. On the posterior surface of tibia I there
are six spines in the lower row of bristles and six in the upper; tibia
II has six in the lower row and seven in the upper; and tibiae III and
IV six in the lower and five in the upper. On the anterior surface

216 Bulletin: museum of comparative zoology.

of all metatarsi there are four spines in the lower row of bristles and
four also in the upper. On the posterior surface of metatarsus I
there are four spines in the lower row and four in the upper; on
metatarsus II there are four and three respectively; on III five and
three; and on IV, four and three.

Abdomen, (Plate 10, fig. 1), covered with small scattered groups
of short, curved, stiff bristles and over whole surface uniformly with
very short hair.

Spinnerets small, dark brown, removed from end of abdomen in
usual way. Colulus thin, pale, almost half as long as spinnerets.

Female. Length, 18.3 mm. Length of cephalothorax, 7.3 mm.;
width, 7.6 mm.; width in front, 3.4 mm.

fem. tib.+pat. met. tar. total

Leg I 8 mm 10.5 mm. 5.8 mm. 3.8 mm. 28.1mm.

Leg II 8.6 10.8 6.3 3.6 29.3

Leg III 8.5 10.0 5.7 3.8 28.0

Leg IV 8.2 9.8 5.8 3.8 27.6

Tibia I, 7.8 mm. Tibia IV, 7 mm.

Locality.— Santa Ana, 3,500 feet, August. (Type, M. C. Z. 165,
female).

DYSDERIDAE.

Ariadna hotchkissi, sp. nov.
Plate 10, fig. 4, 5.

Carapace dusky brown or fuscous, not adeolate. Sternum, cheli-
cerae and coxae of palpi dusky, not areolate or only slightly so; labium
darker, nearly black. Abdomen beneath dark grey; above obscurely
mottled with partly confluent small areas of dusky purple, a more
solid median longitudinal stripe at base ; hairs long and short, numer-
ous, dark.

Eyes large, the medians and laterals equal or very nearly so. Line
formed by two anterior lateral eyes shorter than the posterior in ratio
17:15. Median eyes contiguous, their diameter or a little less from
the laterals. Lateral eyes contiguous, (Plate 10, fig. 4). Clypeus
equal in width to the diameter of an anterior lateral eye or a little
narrower.

Chelicerae short; smooth.

chamberlin: the arachnida. 217

Labium nearly four fifths as wide as long ; strongly narrowed distad ;
apically obtusely rounded, the proximal notches long, (Plate 10, fig. 5).

Sternum long, the ratio of length to width being 85 : 53 ; caudal end
obtusely angular.

Femur of leg I armed on anterior (inner) side with one large spine;
patella unarmed; tibia armed ventrally with four pairs of uniform
spines, not armed laterally; metatarsus armed ventrally with a
double row of spines (6-6 or 6 + 7 in number) of which in each row
the first three or four and the distal one are longer than the others.
In leg II the femur and patella are unarmed; the tibia and metatarsus
are armed like those of leg I. The femur and patella of leg III are
also unarmed; the tibia is armed ventrally in the median line with
two spines, one proximad of and one distad of the middle; the meta-
tarsus is armed ventrally toward the caudal side with two spines, one
median and one subbasal, and at distal end with three spines. Leg IV
wholly unarmed as usual.

Female. Length 7 mm. Length of cephalothorax 2.8 mm.;
width 2 mm.

fern. tib.+pat. met. tar. total

Leg I 2 mm. 2.6 mm. 1.3 mm. .9 mm. 6.8 mm.

Leg II 1.8 2.2 1.2 .8 6.0

Leg III 1.7 1.8 1 .7 5.2

Leg IV 2 2.2 1 .8 6.0

Tibia I, 1.6 mm.' long.

Locality.— Lucma, 7,000 feet, August. (Type, M. C. Z. 167, one
female).

Named for H. Stuart Hotchkiss of New Haven, a patron of the
expedition.

CAPONIDAE.

NOPS BELLULA, Sp. nOV.

Plate 10, fig. 6-8; Plate 11, fig. 1-3.

Carapace and sternum light orange-yellow; legs more lemon-yellow
or with femora of legs I and II tinged with orange. Eyes on a black
spot. Abdomen pale grey-green; dorsally at base with a small
deep brown or blackish triangular spot of which the apex is cephalad,
followed by a series of chevron marks of a color a little darker than
that of the general surface, (Plate 11, fig. 1).

218 BILLETIX: MUSEUM OF COMPARATIVE ZOOLOGY.

Pars cephalica rather broader and less convex along anterior margini
than in related species.

Clypeus slanting.

Eyes large; five sixths their diameter apart, a little less than three
times their diameter (25 : 9) from lower edge of clypeus. Eye-tubercle
rather low but distinct, (Plate 10, fig. 7).

Sternum longer than wide in ratio 25:19; narrowing from middle
caudad with the caudal end obtusely rounded, (Plate 10, fig. 6).

Labium subcordate; distally acute; basal notches short, relatively
deep; a little longer than wide, the ratio being about 17:16, (Plate 10,
fig. 8).

Coxae I and IV longer than others, subequal or I but slightly longer
than IV. Trochanter IV longest, I next.

Claws of anterior legs with five long, stout teeth. An unpaired
claw is present on anterior legs, this being rather long, parallel with
paired claws, smooth. Membranous laminae below claws as in other
species, (Plate 11, fig. 3). Membranous appendage at base of ante-
rior tarsi and the membranous keel beneath anterior metatarsi as in
other species of the genus. All tarsi distinctly clavate excepting the
fourth which is but slightly so and more slender. The division of
tarsus most complete in fourth legs in which the parts are flexible at
joint. Hairs of tarsus all simple, none clubbed as c. g., in coccineus,
{Cf. Plate 11, fig. 2).

Inferior piece of lorum of pedicel as usual.

Inferior and median spinnerets in a recurved trans^*erse row as in
the other species. The superior spinnerets shorter than in most
species, not so much exceeding those of the lower row.

Female. Length 6.7 mm.

total

coxa

fern .

pat.

tib.

met.

tar.

(excl. coxa)

Leg I

. 76 mm .

1.6 mm.

.8 mm.

1.03 mm.

.84 mm.

.56 mm.

4.84 mm.

Leg II

.6

1.36

.8

.92

.8

.56

4.52

Leg III

.52

1.2

.68

.88

.8

.56

4.12

Leg IV

.72

1.6

.8

1.28

1.36

.72

5.76

Locality. — Ollantaytambo, 9,000 feet, July. (Type, M. C. Z. 168,
one female).

This form differs from all other species of the genus, according to
the published accounts, in having a large median claw on the anterior
tarsi in addition to the paired membranous laminae. Simon, in dis-
cussing the Caponidae, says in regard to the inferior claw in this

chamberlin: the arachnida. 219

famih' (Hist. nat. Araign., 1, p. 325) "la griffe inferieure est assez
petite et toujours mutique; cliez les Nops elle est remplacee aux
paires anterieures par deux petites lames membraneuses." But
since in the case of the present species both the claw and the laminae
are present at the same time, we cannot look upon the development
of the laminae as such a replacing. The median claw is higher than
usual in position, being inserted on a level with and lying between
the paired claws so that it is easily overlooked; it is quite possible,
therefore, that it will be found in some degree of development in
other species. The superior spinnerets seem to be relatively consid-
erably shorter than usual. The color-pattern of the abdomen is
distinctive. In our present state of knowledge, it is impossible to
give a wholly satisfactory judgment as to the position and relation-
ships of the present form.

DRASSIDAE.

Drassodes sp.

An immature specimen of uncertain species was taken at Cuzco,
11,500 feet, in July. (M. C. Z. 169).

Drassodes araucanius,^ sp. nov.

Plate 11, fig. 4-8; Plate 12, figs. 1-2.

Carapace light brown or testaceous; eyes, excepting sometimes the
posterior medians, ringed with black and the intervening area in some
dusky. Chelicerae brown to dilute chestnut. Sternum and endites
like the carapace; labium dark dusky brown to dusky chestnut.
Legs light brown or testaceous, the anterior pairs darker distally.
Tarsus of palpus abruptly darker than proximal joints. Abdomen
clear brown-grey without markings, paler beneath than above. Cara-
pace clothed with numerous fine greyish hairs, a large proportion of
which are plumose, with coarser, black, more erect hairs sparsely
intermixed. Sternum densely clothed with similar mostly plumose
hairs and sparsely with coarser, more bristle-like, black hairs. Legs
clothed with finer appressed plumose grey hairs, more erect, stiffer

1 Araucania, a tribe o( Soutli American lodiiins.

220 bulletin: museum of comparative zoology.

simple grey hairs proportionately more numerous distad, and with
numerous intermixed stiff black hairs. The abdomen is clothed
with similar plumose grey pubescence and numerous longer stiff grey
hairs with the black bristles more sparsely intermixed.

Posterior row of eyes longer than the anterior in the ratio 63:53.
Posterior row of eyes moderately procurved; median eyes placed
very obliquel}', subelliptic (ratio of axes 11:8), half of their smaller
diameter apart, about twice their diameter from the laterals and
their long diameter or a little more from the anterior medians ; laterals
with diameter smaller than long diameter of medians (9:11) and
slightly exceeding the lesser diameter (9:8). Anterior row of eyes
in dorsal view a little recurved, in anterior y'lerw considerably pro-
curved; median eyes subequal to the laterals, their diameter apart
and less than their radius (three fifths) from the laterals; eyes their
diameter from lower margin of clypeus. Anterior lateral eyes some-
what larger than the posterior laterals from which they are separated
by less than the diameter of the former, nearer together than the
anterior and posterior medians. Area of median eyes equal in length
and breadth, longer in front than behind in ratio 29:25, (Plate 11,
fig. 4).

Labium longer than wide in ratio 7:5; narrowing moderately distad,
apically broadly obtusely rounded, (Plate 11, fig. 6).

Sternum longer than wide in ratio 13:9. Acutely pointed caudad,
(Plate 11, fig. 5).

Leg I with femur above in median line with a distally bristle-like,
but basally stout, subbasal spine and a submedian one, and with one
on anterior side between middle and distal end; patella unarmed;
tibia armed at distal end with one spine; metatarsus at base with
two; metatarsus scopulate to base. Leg II with femur spined as in
I with in addition a submedian one in line with the one on anterior
surface and one in corresponding position on caudal side; patella
unarmed; tibia ventrally with a distal and a submedian pair of spines
and a single spine at base ; metatarsus with a pair of spines at proximal
end, scopulate to base. Leg III with femur spined as in II excepting
for the addition of a subapical one (and sometimes a second one
proximad of it) in the median dorsal line and another one in line with
the one on the caudal side between middle and distal end, there being
thus three or four median dorsal, two anterior, and two posterior
spines; patella armed on the caudal side with one spine; tibia with
three pairs of ventral spines, two in line on ventral part of anterior
side, two in corresponding position on the posterior, one subdorsal

chamberlin: the arachnida. 221

on anterior side and one subbasal toward caudal side of dorsal surface;
metatarsus with three pairs of ventral spines, two spines in line on
caudal and two in line on dorsal surface, and two pairs (apical and
submedian) on dorsal surface; metatarsus not at all scopulate. Leg
IV with femur armed above in median line with two spines, on caudal
side toward distal end above with one (or exceptionally two close
together) spine and on anterior with two in line (the distal of which
may be doubled); patella with one spine on caudal surface; tibia
with three pairs of ventral spines, two in line on caudal surface and
three on the cephalic, the most proximal of the latter more ventral
in position; metatarsus with three pairs of ventral spines (or one
spine absent from subbasal pair), two (apical and submedian) on
cephalic side, and two pairs and a single subbasal spine on caudal side;
metatarsus not at all scopulate. Claws of leg I with five slender
teeth, (Plate 12, fig. 2).

Female. Length 12 mm. Length of cephalothorax 5 mm.; width
3.8 mm.

fern. tib +pat. met. tar. total

Leg I

3.6 mm.

4.4 mm.

2.1 mm.

1.3 mm.

11.4 mm

Leg II

3

4.1

2

1.2

10.3

Leg III

3

3.2

2

1.2

9.4

Leg IV

3.7

5

3

1.9

13.6

Tibia I, 2.4 mm.

Tibia IV,

3.1 mm.

Locality — Cuzco, 11,500 feet, July 4. (Type M. C. Z. 170, female;
paratypes, no. 171, females). Ollantaytambo, 9,000 feet, July.
(M. C. Z. 327, several females).

Apodrassus,^ gen. no v.

Cephalothorax low, pars cephalica in profile descending moderately
anteriorly, much narrowed in front; thoracic stria fine, distinct,
moderate in length.

Eyes relatively large. Posterior row of eyes strongly procurved,
the medians considerably farther from each other than from the
laterals. Anterior row of eyes also rather strongly procurved; the
medians the larger, widely separated from each other but very close

' 'aTTo, away from, Drassus.

222 bulletin: museum of comparative zoology.

to anterior and also to posterior laterals. Lateral eyes on each side
subcontiguous. Clypeus narrow, its width not exceeding the radius
of an anterior lateral eye.

Labium wider than long.

Lower margin of ehelicera with one tooth; upper with three of
which the median is largest.

Legs slender; wellspined; tarsi scopulate but not densely so.

Inferior spinnerets widely separated; median spinnerets moderate,
cylindrical.

Genotype. — Apodrassus andinus, sp. nov.

Related to Leptodrassus Simon, but differing in the much narrower
clypeus, more strongly recurved eye-rows, in having the posterior
median eyes decidedly farther from each other than from the laterals,
the tarsi moderately scopulate, and the inferior spinnerets widely
separated instead of connivent.

Apodrassus andinus, sp. nov.
Plate 12, fig. 3-8.

Carapace and legs light brown, the carapace and femora slightly
dusky. Sternum and coxae of legs beneath clearer, more testaceous.
Abdomen beneath light brown, above darker, dusky brown.

Posterior row of eyes very strongly procurved, a line tangent to the
anterior edges of the medians being nearly tangent to the posterior
edges of the laterals; medians elongate (axes to each other as 8:5),
set very obliquely, separated by a distance equal to the lesser diam-
eter, closer to the laterals about three fifths as far; lateral eyes a little
larger than the medians (long diameters about as 9:8), about half
their radius from the anterior laterals to which they are subequal in
size. Anterior row of eyes scarcely longer than the posterior; in
dorsal view a little procurved, in anterior view strongly so, a line
tangent to lower edges of medians passing through caudal third of
laterals ; medians circular with diameter larger than the long diameter
of the laterals (10:9), three fifths their diameter apart, much closer
both to the anterior and the posterior laterals; laterals separated
from lower edge of clypeus by a distance equal to their radius or a
little less, the medians bv about four fifths their diameter, (Plate 12,
fig. 5).

chamberlin: the arachnida. 223

Sternum longer than wide in ratio 7:5 (Plate 12, fig. 3).

Labium somewhat wider than long (nearly as 8 :7) ; distad of proxi-
mal third broadly, almost semicircularly rounded, but distally a little
depressed, (Plate 12, fig. 6).

Tarsi III and IV a little curved, I and II straight. Anterior
tarsi not densely scopulate, the scopular hairs of tarsi III and IV more
sparse; metatarsi I and II with a few scopular hairs extending nearly
to base, III and IV without any. Scopular hairs very broad and
strongly clavate. Leg I with femur bearing two long spines in mid-
dorsal line (between middle and base and middle and distal end
respectively) and one on anterior surface between middle and distal
end; tibia with three pairs of ventral spines; metatarsus with one
subbasal pair. Leg II with spines same as those of I. Leg III with
femur bearing three spines along middorsal line, two on anterior side
and two on posterior side; patella with a spine on anterior and also
one on posterior surface ; tibia on ventral surface with a pair of spines
at distal end and two single ones in line more proximad, two on ante-
rior and two on posterior surface, and one subbasal one on dorsal
surface; metatarsus on ventral surface with a distal pair and three
(or four?) more proximal ones rather irregularly arranged, a spine at
distal end on each side and dorsally a median distal one, a submedian
pair and a single subbasal one. Leg IV with femur armed with three
spines along middorsal line, two on anterior side and one on posterior;
patella with a spine on anterior and one on posterior side; tibia with
two pairs of dorsal spines, the anterior spine of each pair much smaller
than the caudal one, ventrally with one distal pair, a submedian pair
and a single subbasal spine; metatarsus with two pairs of dorsal
spines, (between middle and base and middle and distal end respec-
tively), three on anterior side and two on posterior, on ventral surface
with two at distal end, one toward middle and one toward base. Claw
of leg I with seven long teeth, (Plate 12, fig. 7).

Inferior spinnerets widely separated, more than their diameter
apart; spinning tubules very large, arranged in a semicircle.

Spermatheca large, broadly fusiform, densely chitinized and dark
in color, showing distinctly through surface integument, the openings
small. The integument over the spermatheca in type is torn so that
its precise configuration cannot be made out, but its modifications
seem to be very slight, (Plate 12, fig. 8).

Female. Length 6.2 mm. Length of cephalothorax 2.5 mm.;
width 2 mm.

224

bulletin: museum of comparative zoology.

Leg I
Leg II
Leg III
Leg IV

fern.
2.1 mm.
1.9
1.65
2.25

Tibia

tib. + pat.
2.4 mm.
2.35
2.0

2. 85

, 1.55 mm.

met.
1 . 6 mm .
1.6
1.4
2.25

tar.
1 . 1 mm.
1.1
.9
1.1

total

7.2 mm.

6.95

5.95

8.45

Tibia IV, 1.8 mm.

Locality. — Huadquina, 5,000 feet, July. (Type, M. C. Z. 172,
female).

PHOLCIDAE.
Hypsorinus,^ gen. nov.

Eyes elevated on a tubercle. Anterior row of eyes strongly re-
curved; median eyes with radius one third more or less than that
of the laterals, near together and only about their radius from laterals.
Posterior row of eyes a little procurved, the space between medians
in type only one and a third times their long diameter. Area of
median eyes trapeziform, clearly longer than posterior width.

Abdomen short, strongly elevated, and in type distinctly higher
than long.

Legs in both sexes without spines.

Chelicera in male with a short stout tooth near base of claw on
upper surface, (Plate 14, fig. 3).

Tarsus of palpus much longer than thp tibia; pointed distad but
not abruptly more slender than tibia.

Genotype. — Hypsorinus binghamae, sp. nov.

This genus is most closely related to Smeringopus of Simon from
which it is here separated chiefly because of the following differences : —
the abdomen is high and subglobose instead of elongate and slenderly
cylindrical; the tarsus of the female palpus is much longer than the
tibia instead of abruptly much shorter and more slender; and the pos-
terior row of eyes is procurved instead of recurved, while the ante-
rior median eves are much closer to the laterals.

Hypsorinus binghamae, sp. nov.

Plate 13, figs. 1-9; Plate 14, figs 1-7.

Carapace and sternum very light greyish brown, a darker median
longitudinal stripe narrowing up the clypeus over the eye-area and

' vtf'os, top, 'opupcn, mountaineer.

chamberlin: the arachnida. 225

thoracic groove to the caudal margin where it is reduced to a mere
line, the carapace elsewhere with scattered small dark dots. Legs
dilute testaceous to nearly grey; femora toward distal end with two
darker, somewhat reddish brown, annuli incomplete dorsally and
with more indistinct dark cross-marks more proximad ; patellae with a
dark annulus over most of length; tibiae with three annuli, one sub-
basal, one distad of middle, and one distal or with the dark color more
diffused and no distinct annuli indicated. Abdomen dull grey, more
or less closely mottled wdth very small, not strongly contrasting, silver
spots, a clear narrow median stripe free from spots extending three
fourths the distance to the caudal end; on each side of median stripe
two or more pairs of dark dots of which one pair near the caudal and
one near the anterior end are more conspicuous.

Eye-tubercle conspicuous, subdivided by a y-shaped furrow with
the branches cephalad, the tubercle between them bearing the anterior
median eyes and each lateral tubercle bearing two lateral and the one
posterior median eye of the corresponding side. Anterior row of
eyes strongly procurved both in dorsal and in anterior view as in
Smeringopus; median eyes much smaller than laterals the diameter
of which is somewhat more than 1.5 times greater, less than their
radius apart and only their radius from the laterals. Posterior row
of eyes a little procurved; median eyes smaller than the laterals
(ratio of long diameters near 5:6), very close to laterals, once and a
third their long diameter apart. Lateral eyes equal or nearly so,
their radius or a little less apart. Area of median eyes in dorsal view
trapeziform, narrower in front than behind in about ratio 25:43,
longer than greatest width (ratio 7:6), (Plate 13, fig. 2).

Clypeus high, nearly twice as high as length of median eye area.

Cephalothorax convex and very high. Thoracic groove very deep,
radial impressions distinct. Pars cephalica small, not elevated.

Sternum wider than long nearly in ratio 3:2.

Labium large, convexly rounded at tip, (Plate 13, fig. 3).

Paired claws with eight to twelve long teeth, (Plate 14, fig. 5).
Unpaired claw with a single long tooth, (Plate 13, fig. 7). Bristles of
legs with slender teeth near base, mostly two to four in number on one
side but one or more may also be present on the other, while in a
number near the claw the number of branches or teeth is much larger,
(Plate 13, fig. 9; Plate 14, fig. 2). A feather-hair, of which a pair
occur near claw of leg I, is represented, (Plate 14, fig. 1).

The abdomen is distinctly higher than long. The spinnerets are
borne considerably farther forward beneath abdomen in the female
than in the male, {(^. Plate 13, fig. 1).

226 bulletin: museum of comparative zoology.

Male (Type). Length 5.5 ram. Length of eephalothorax 2.3 mm.;
width, 2.4 mm.

fern. tib.+pat. - met.

Leg I 10.8 mm. 12.2 mm. 16.2 mm.

Leg II 8.2 8.8 11.0

Leg III 6.5 7.0 8.8

Leg IV 9.0 9.0 11.8

Female. Length 5.5 mm. Length of eephalothorax 2.2 mm.;
width 2.4 mm.

fem.

tib.+pat.

met.

Leg I

10.8 mm.

11.8 mm.

14.5 mm

Leg II

7.8

8.0

10.0

Leg III

6.3

6.3

8.2

Leg IV

8.8

8.8

11.0

(Because of the curHng of the tarsi it is very difficult to determine
their lengths accvu-ately, so these are not given above).

Xoca%.— Huadquina, 5,000 feet, July. (Type, M. C. Z. 173,
male; paratypes, M. C. Z. 174, one adult female and two immature
specimens).

Named for Mrs. Alfreda Mitchell Bingham,

LiTOPORUS ABERRANS, sp. UOV.

Plate 14, fig. 8, 9; Plate 15, fig. 1-3.

Carapace dilute yellow, a deep brown or blackish median band
extending from caudal margin across eye-area and down the clj^eus
to its lower edge, this band widening up the posterior declivity but
of uniform width in front of this, enclosing a narrow paler longi-
tudinal area between the eyes ; a dark band extends on each side from
the caudal end of the median band along the margin ectad and for^^ard
some distance in front of middle of length. Sternum brown, paler
along borders. Legs brown, the femora each with a black annulus
at distal end set off by a light ring adjacent to it on its proximal side,
also somewhat darker at proximal end; patella dark; tibia with a
dark annulus at each end; other joints not ringed. Abdomen with
ground color whitish; above weakly suffused with blue and with
deeper solid blue spots in a row each side of a middle stripe the row
curving ectad at caudal end and not attaining the posterior end of
abdomen; on each side a few less deeply colored spots ; caudal surface

chamberlin: the arachnida. 227

of abdomen with a large dark area of continuous dilute blue closely
covered with deeper blue spots, this area reaching to the spinnerets;
ventrally there is a moderately large quadrate area of brown in the
genital region.

Posterior row of eyes with median eyes essentially contiguous with
the laterals as usual, separated from each other by a distance some-
what greater than their diameter (9:7), smaller than the laterals
(diameters as 7:10) which are about equal to the anterior laterals.
In dorsal view the anterior row of eyes is distinctly procurved; in
anterior view by their centres they are strongly procurved but with
upper edges in a straight line ; the median eyes are nearly contiguous,
separated by a distance clearly less than their radius, separated from
the laterals by a distance greater than their radius but less than their
diameter, (Plate 15, fig. 1).

Chelicera of male without the usual tooth on the anterior face and
also lacking a true carina toward base; inner chitinous edge bent out-
ward at clypeal margin and showing as a dark chitinous line or ridge.

Labium broad, distally truncate, (Plate 14, fig. 8).

Sternum wider than long in ratio 6:5; the wide caudal margin be-
tween coxae of fourth legs convex, (Plate 14, fig. 8).

Male. Length 4 mm. Length of cephalothorax L5 mm. (to edge
of clypeus) ; width L4 mm.

fem. tib.+pat. met.

Leg I 5.2 mm. 6 mm. 6.2 mm.

Leg II 4.2 4.2 4.3

Leg III 3.2 3.1 4

Leg IV 4.6 5. 5

Locality.— Urubamba, 9,500 feet, July. (Type, M. C. Z. 175, one
male).

Aberrant in its genus in lacking any distinct tooth or true carina
on anterior face of chelicera but in other respects apparently conform-
ing to the genus in its typical form. The abdomen is higher than
usual.

THERIDIIDAE.

Argyrodes vittatus Keyserling.

Spinnen Amerikas. Theridiidae, 1884, 2, pt. 1, p. 191, pi. 9, f. 114.

One female of this species from San Miguel, 6,000 feet, July.
(M. C. Z. 176).

228 bulletin: museum of comparative zoology.

Argyrodes lucmae, sp. nov.
Plate 15, fig. 4-6.

Carapace pale yellow. Sternum dusky brown or blackish. Labium
blackish, pale across tip. Endites yellowish with some dusky mark-
ings. Legs pale yellow or, especially the caudal pairs, proximally
whitish. Abdomen covered with scales of shining silver over a pale
yellow ground.

Abdomen moderately elongate; roundly elevated above toward
base, (Plate 15, fig. 5).

Posterior row of eyes straight or very nearly so; medians about
five sevenths their diameter apart, a little closer to the somewhat
smaller laterals (6:7). Lateral eyes contiguous, subequal. Anterior
row of eyes straight; medians their radius or slightly more apart,
about half as far from the laterals which are smaller (diameters about
as 6 : 7). Clypeus slightly narrower than the diameter of a median eye.

Labium wider than long {cir. 4:3); sides but little convex and
moderately converging distad ; distally widely truncate.

Sternum broadest across anterior end, rounded caudad and extended
as a slender acute process between the fourth coxae; total length to
the width about as 9:8, (Plate 15, fig. 4).

Male. Length 3.8 mm.

fem. tib. + pat. met. tar. total

Leg I 5 mm. 6.2 mm. 7 mm. 1.6 mm. 19.6 mm.

Legll 3.2 3.7 3 1.1 11.0

Leg III 2.6 2.5 2 .9 8.0

Leg IV 3 2.2 2.3 1.0 8.5

Tibia I, 5.8 mm.

Locality. — Lucma, 7,000 feet, August. (Type, M. C. Z. 177, one
male).

Theridion sp. a.

A young female of doubtful species from Lucma, 7,000 feet, August.
(M. C. Z. 178).

Theridion sp. b.

An immature female from the Conservidayo River, August.
(M. C. Z. 179).

chamberlin: the arachnida.

229

Theridion tosum/ sp. nov.
Plate 16, fig. 1-4.

Carapace and chelicerae pale testaceous; sternum, labium, and
endites yellow; legs yellow except proximally above where similar
to carapace; all these parts without special markings. Abdomen
grey, a longitudinal wavy white line on each side of middle of dorsum
the two lines converging to spinnerets ; a few small light dots between
the wavy lines caudally; otherwise the abdomen is unmarked.

Posterior row of eyes substraight or slightly recurved; eyes nearly
equidistant, the medians being about their diameter apart and the
same distance or slightly farther from the laterals. Posterior lateral
eyes equal to the anterior laterals or very nearly so. Anterior row of
eyes in front view straight ; median eyes slightly smaller than laterals,
to which they are closer than to each other, more than their diameter
apart. Area of median eyes equal in width in front and behind, equal
in length and breadth or slightly wider, (Plate 16, fig. 3).

Labium wider than long as usual; distally widely semicircularly
rounded, (Plate 16, fig. 1).

Sternum longer than wide in ratio 6 : 5, more narrowly attenuated,
and rounded caudad than, e. g., in the succeeding species, (Plate 16,
fig. 2).

Anterior paired claw of leg IV with four teeth of which the most
distal is very long and reaches about the same level as tip of claw;
teeth of other claw shorter. Unpaired claw untoothed.

Female. Length 4.3 mm. Length of cephalothorax, 2 mm.;
width, L47 mm.

fern. tib.+pat. met.

tar.
.9 mm.

.7
.6

.8

total
8.0mm.
6.3
5.0

7.8

Leg I 2.. 3 mm. 2.9 mm. 1.9 mm.

Leg II 2 2.1 1.5

Leg III 1.9 1.5 1.0

Leg IV 2.9 2.2 1.9

Locality.— Ruaidquma, 5,000 feet, July. (Type, M. C. Z. 180,
one female).

Theridion leguiai, sp. nov.
Plate 15, fig. 7-10.

Carapace on sides obscure olive-white, a broad black median longi-
tudinal band over clypeus and eye-area and narrowing from there

Gosiute tosibit, iosa, white or light colored.

230 bulletin: museum of comparative zoology.

caudad down posterior declivity of cephalothorax ; lateral borders also
black, the lateral bands confluent with the median one in front; cly-
peal margin pale. Sternum testaceous, broadly bordered with black.
Labium and endites blackish. Chelicerae testaceous in front, darker,
chestnut, distally. Legs black, conspicuously annulate with yellow or
testaceous; femora with a broad light band proximally and a narrower
one toward distal end; patellae with one subdistal annulus; tibiae
with an annulus at proximal end and one between middle and distal
end; metatarsi with annuli corresponding to those of tibiae; tarsi en-
tirely light or light at the ends. Abdomen black; on dorsum toward
caudal end in median line a row of three small light dots and in front of
these two pairs of separated oblique light marks of which the anterior
send a slender branch forward. Hairs of abdomen short, numerous.

Posterior row of eyes a little procurved ; median eyes only slightly
elongate and oblique, about seven tenths their longer diameter apart
and nearly the same distance from the laterals; laterals equal in size
to the medians, larger than the anterior laterals with which they are
contiguous. Anterior row of eyes in dorsal view strongly recurved,
the median eyes being carried considerably forward as usual, larger
than the laterals (diameters as 11:7), a little more than their radius
apart (six elevenths of diameter), only half as far from the laterals;
in front view the row is straight or nearly so. Area of median eyes
subquadrate, being very slightly wider in front than behind. Cheli-
cerae much longer than the height of the clypeus, (Plate 15, fig. 8).

Labium wider than long in ratio 22:13; distally widely truncate.

Sternum subtriangular; longer than wide in ratio 6:5.5; caudal
process obtusely rounded, (Plate 15, fig. 7).

Paired claw of leg I with seven or eight teeth of which the proximal
ones are very short, (Plate 15, fig. 9).

Abdomen subglobose, smoothly rounded, with no irregularities or
tubercles.

Female. Length, 5.1 mm. Length of cephalothorax, 2.2 mm.;
width, 1.85 mm.

fem.

tib.+pat.

met.

tar.

total

Leg I

3 . 6 mm.

4 mm.

3 mm.

1.3 mm.

11.9 mm

Leg II

2.3

2.8

2

1.2

8.3

Leg III

2

2

1.3

1

6.3

Leg IV

3.1

3

2.2

—

—

Locality. — ■ Conservidayo River, August. (Type, M. C. Z. 181,.
one female).

Named for Mr. A. B. Leguia, former President of Peru.

chamberlin: the arachnida. 231

Garricola,^ gen. nov.

Cephalothorax broadly ovate, much narrowed cephalad; front
narrow, convex. Thoracic furrow transverse, wide, recurved.

Eyes of the posterior row procurved; subequal; the medians
farther apart than from the laterals from which they are separated
by much less than their diameter. Anterior eyes subequal, in a
strongly procurved row, medians well separated but very close to
the laterals. Area of median eyes quadrate, of equal width anteriorly
and posteriorly.

Sternum as wide as or wider than long; posteriorly obtuse, with
the coxae rather widely separated.

Labium not free; much wider than long, distally narrowed, convex.

Endites bent, conspicuously converging over labium which extends
rather beyond the middle of their length. ,

Legs slender, of moderate length, clothed with slender bristles.
Teeth of claws few, divaricate.

Genotype. — Garricola sanctits, sp. nov.

Garricola sanctus, sp. nov.
Plate 16, fig. 5, 7.

Carapace and sternum dilute yellow. Legs dilute yellow, dusky
especially distad. Abdomen grey.

Cephalothorax broad, subcircular in outline caudad, strongly
narrowed cephalad. Eye-tubercle elevated. Thoracic furrow trans-
verse, broad, recurved, (Plate 16, fig. 5).

Posterior row of eyes distinctly procurved; medians a little larger
than the laterals, about three fourths their long diameter apart
and about their radius from the laterals. Lateral eyes contiguous,
scarcely differing in size. Anterior median eyes three fourths their
diameter apart and about one half their radius from the laterals,
which are of an equal size or nearly so ; anterior row of eyes strongly
prociu"ved. Area of median eyes equal in length and breadth, of
same width anteriorly and posteriorly; clypeus lower than length of
chelicerae and a little wider than length of median eye-area (6:5),
oblique, slanting cephalomesad.

Sternum convex and prominent; wider than long (10:9); broad

1 Gosiute garri, mountain range, cola.

232 bulletin: museltm of comparative zoology.

anteriorly and triangularly narrowing caudacl, but truncate between
the widely separated fourth coxae, (Plate 16, fig. 6).

Labium not free; distally strongly convexly rounded; much wider
than long (cir. 7:4).

Abdomen subglobose. Spinnerets terminal, (Plate 16, fig. 5).

Bristles in comb of fourth tarsus eight or nine in number. Teeth of
claws few, divaricate.

Epigynum proportionately large, (Plate 16, fig. 7).

Length of female, 3 mm.

Locality.— San Miguel, 6,000 feet, July. (Type, M. C. Z. 182, one
female). ■

Latrodectus mactans (Fabricius).

Aranea mactans Fabr., Ent. syst., 1775, 2, p. 410.

A species occurring widely in the western hemisphere from New
England to Terra del Fuego.

Localities.— OWa^ntaytaimho, 9,000 feet, July 21. (M. C. Z. 183,
five females and one male). Cuzco, 11,500 feet, July 12. (M. C. Z.
184, two females).

Lithyphantes nigrofemoratus Key'serling.

Spinnen Amerikas. Theridiidae, 1884, 2, pt. 1, p. 139, pi. 6, f. 87.

A species previously known from Monterico in Peru and from
Guatemala.

Localities. — Huadquina, 5,000 feet, July. (M. C. Z. 185, one
female). Cuzco, 11,500 feet, July. (M. C.^ Z. 186, one female).

Enoplognatha sp.

A young female of uncertain species from Tincochaca, 7,000 feet,
July. (M. C. Z. 187).

Enoplognatha peruviana, sp. nov.

Plate 16, fig. 8-11; Plate 17, fig. 1-2.

Carapace and legs light brown or testaceous; sternum darker, of
somewhat chestnut cast. Abdomen over anterior and caudal ends

chamberlin: the arachnida. 233

and over upper part of sides dark, somewhat greyish black, a wavy
or zig-zag light line dividing the dark along the upper part of the
dark area of each side ; dorsum light, the light area divided by a me-
dian longitudinal stripe formed by two contiguous triangular areas of
which the apices are cephalad ; lower portion of sides lighter, crossed
by several more or less obscure light lines; median portion of venter
dark, a light longitudinal line on each side. Spinnerets light testa-
ceous or yellow.

Posterior row of eyes a little procurved; median eyes much the
largest of all, elongate, long axis paralled with that of body, only
about one third their long diameter apart, much farther, about four
fifths their diameter, from the laterals ; laterals about one third their
diameter from the anterior laterals which are considerably smaller.
Anterior row of eyes straight; median eyes distinctly smaller than the
laterals (diameters about as 11:13), their radius apart and only their
diameter from the laterals. Area of median eyes longer than wide
(9:8), wider behind than in front not quite as 8:7, (Plate 17, fig. 1).

Labium as usual much wider than long (ratio 11:8); semi-circularly
rounded distad, (Plate 17, fig. 2).

Sternum subtriangular, narrowed caudad to a slender process ex-
tending between coxae of last legs, (Plate 16, fig. 8).

Paired claws with numerous long teeth, the tips of which are in a
straight line, these on leg I numbering ten or eleven, (Plate 16, fig. 9).
Unpaired claw with a single small denticle, (Plate 16, fig. 10).

Female. Length, 8 mm. Length of cephalothorax, 3 mm. ; width

2.4 mm.

feni.

tib. + pat.

met.

tar.

total

Leg I

3 mm.

3.6 mm

2.2 mm.

1 mm.

9.8 mm

Leg II

2.6

3

2

.9

8.5

Leg III

2.3

2.9

1.8

.9

7.9

Leg IV

3.1

3.6

2.2

1.2

10.1

Loca%.— Urubamba, 9,500 feet, July. (Type, M. C. Z. 188,
female; para type, M. C. Z. 189, 2 females).

Enoplognatha dubia, sp. nov.
Plate 17, fig. 3.

Carapace dilute testaceous, a blackish median longitudinal stripe
.over stria thoracica, this becoming wider and more dilute cephalad

234 bulletin: museum of comparative zoology.

toward eyes; lateral margins dusky. Sternum dusky testaceous,
more blackish caudad. Labium blackish; endites paler, testaceous,
Legs testaceous; the femora with three wide dusky annuli; patellae
dusky around distal end and tibiae with three dark rings, the annuli
of all joints most distinct on anterior legs. Abdomen somewhat pale
testaceous over sides; above a narrow, solid black, foliate mark over
entire length, the edges wavy and bordered on each side by a distinct
white line; venter with a black longitudinal stripe over entire length
which narrows somewhat from anterior end caudad.

Posterior row of eyes straight; median eyes a little smaller than
the laterals diameters about as 7:8), a little less than their diameter
apart and a little more than their diameter from the laterals. Lateral
eyes contiguous, subequal. Area of median eyes of nearly equal
length and breadth ; of same width in front as behind. Anterior row
of eyes straight; median eyes their diameter apart, considerably
nearer to the laterals (near their radius). Clypeus of nearly same
width as area of median eyes, but not much more than half as wide
as the length of the chelicerae.

Labium about three fourths as long as wide; the distal margin
convex.

Sternum not quite five sixths as wide as long; process between
posterior coxae distally rounded, not acute.

Female. Length, 5.2 mm. Length of cephalothorax, 2 mm.;
width, L6 mm.

Locality. — Sorontoy, 7,000 feet, September. (Type, M. C. Z. 190,
one female).

LINYPHIIDAE.

Erigone taibo,^ sp. nov.
Plate 17, fig. 5.

Carapace, sternum, mouthparts, palpi, and legs brownish black.
Abdomen yellowish white; epigynum and spinnerets blackish.

Posterior row of eyes a little recurved; eyes equidistant, the medians
one and two thirds their diameter apart. Lateral eyes contiguous,
the anterior the larger. Area of median eyes wider behind than in
front in the ratio 10:7; about equal in length and greatest width.

' Gosiute laibo, a white person.

chamberlin: the arachnida. 235

Anterior row of eyes slightly procurved; median eyes two thirds their
diameter apart, one and two thirds or more their diameter from the
laterals; median eyes nearly four times their diameter from lower
margin of the clypeus.

Carapace smooth ; not marginally dentate.

Lower margin of fm-row of chelicera with three or four teeth of
which the one nearest the claw is double or bifid.

Length of female 3.5 mm.

Zoca/t^i/.— Urubamba, 9,500 feet, July. (Type, M. C. Z. 191;
para types M. C. Z. 192, three females).

Erigone niwina,^ sp. nov.
Plate 17, fig. 4.

Carapace, sternum, and coxae of legs light red or dilute chestnut.
.Legs distad of coxae dusky brown or blackish. Entire abdomen
grey-black.

Carapace smooth.

Posterior row of eyes straight; median eyes their diameter apart,
a little farther from the laterals {cir. one and a fifth). Lateral eyes on
each side subcontiguous, the anterior distinctly the larger. Area of
median eyes wider behind than in front in the ratio 11:7; slightly
longer than the greatest width (12:11). Anterior row of eyes slightly
procurved ; median eyes smaller than the laterals in about ratio 3 : 4,
two thirds their diameter apart, one and two thirds their diameter
from the laterals and four times their diameter from the lower margin
of clypeus.

Sternum of the usual general form ; equal in length and breadth or
slightly wider.

Upper margin of furrow of chelicera typically with a triple or trifid
tooth nearest claw and two isolated teeth; lower margin with five
teeth.

Epigynum of same general form as that of E. taibo, (Plate 17, fig. 4).

Length of female 3.5-4 mm.

Loca%.— Urubamba, 9,500 feet, July. (Type, M. C. Z. 193,
female; paratype, M. C. Z. 194, one female).

1 Gosiute niwina; aD iDclian.

236 bulletin: museum of comparative zoology.

Oedothorax melacra/ sp. nov.
Plate 17, fig. 6, 7.

Carapace and legs light brown or the latter somewhat more yellow-
ish, the former dusky anteriorly and along lateral edges. Sternum
black. Abdomen t;yT)ically pink, more rarely brown of only faint
pinkish tinge; tip of abdomen about bases of spinnerets black;
epigynum black.

Posterior row of eyes straight; eyes equal in size; median eyes but
little more than their radius apart, three fourths their diameter from
the laterals. Lateral eyes contiguous, equal in size or the anterior,
but slightly larger. Anterior row of eyes straight or very slightly
recurved ; median eyes with diameter three fourths that of the laterals,
not fully their radius apart, two thirds their diameter from the laterals,
three times their diameter from the lower edge of clypeus. Area of
median eyes wider behind than in front in ratio 10: 7.

Carapace of male smooth, without processes.

Chelicera of male with a rather long, acute process or tooth near
middle of length on anterior side, this directed anteroventrad.

Sternum with process between posterior coxae with sides convex, the
process widening a little distad of its middle and distally truncate.

Tibia with two spines on outside at distal end on a common base.

Locality.— Ciizco, 11,500 feet, July. (Type, M. C. Z. 195, male;
para types M. C. Z. 196, three females).

Oedothorax orinus,^ sp. nov.
Plate 17, fig. 8.

Carapace and legs light brown. Sternum dusky. Abdomen dark
grey; a median longitudinal black line; caudal portion with several
black che\Ton-marks of which the more anterior are crossed by the
median line.

Carapace smooth, without teeth.

Posterior row of eyes very slightly procurved; eyes subequal;
median eyes three fifths their diameter apart and the same distance,
or very nearly so, from the laterals. Lateral eyes on each side con-

' tJii\ai black, ccKpa, tip.
2 opetvos, mountaineer.

chamberlin: the arachnida. 237

tiguous, equal. Quadrangle of median eyes wider behind than in
front in the ratio 6:5, equal in length and breadth or scarcely longer.
Anterior row of eyes straight or but very slightly recurved; median
eyes smaller than the laterals (diameters about as 4:5), one fourth
their diameter apart, their radius from the laterals, two and a half
times their diameter from the lower margin of the clypeus.

Sternum equal in length and width or slightly wider ; the intercoxal
caudal process narrow.

Upper margin of furrow of chelicera typically with four teeth of
which the three nearest the claw are long and conical or with the one
nearest claw smaller, the fourth much smaller than others ; sometimes
only three teeth of which the one nearest claw may be bifid.

Locality — Cuzco, 11,500 feet, July. (Type, M. C. Z. 197, female;
paratypes, M. C. Z. 198, two females).

TuTAiBO,^ gen. nov.

Cephalothorax short; broadly ovate; frons obtuse and moderately
wide.

Posterior row of eyes substraight or weakly recurved; median eyes
farther apart than from the laterals, distinctly more than their diame-
ter apart. Anterior row of eyes decidedly procurved, its eyes sub-
equidistant. Area of median eyes wider thaij long and wider behind
than in front.

Legs slender; anterior tarsi shorter than the metatarsi.

Frons without special process in the male.

Tibia of palpus in the male with a conspicuous dorsal spine.

Genotype. — Tutaiho debilipes, sp. nov.

TUTAIBO DEBILIPES,^ sp. nOV.

Plate 17, fig. 9-10.

Carapace and sternum black. Legs black or blackish brown,
somewhat paler distally. Abdomen entirely shining black.

Posterior row of eyes slightly recurved; median eye a little more
than once and a half their diameter apart, only about their diameter
from the laterals. Lateral eyes on each side elevated on a common

' Gosiute lutaibo, a negro.
' debilis, feeble, pes, foot.

238 bulletin: museum of comparative zoology.

tubercle, contiguous, subequal. Quadrangle of median eyes wider
behind than in front in the ratio 13: 11, and wider than long in the
same ratio or nearly so. Anterior row of eyes conspicuously pro-
curved; eyes subequal; median eyes two thirds their diameter, or
slightly more, apart and the same distance from the laterals, three
and a half times their diameter from lower margin of clypeus (male).

Sternum with process between posterior coxae moderately wide,
the distal margin a little incurved.

Tibia of male palpus with a stout, acutely pointed, subconical
process from dorsal surface near the proximal end.

Length of male 2.2 mm. ; of female, 3 mm.

Locality. — Huadquina, 5,000 feet, July. (Type, M. C. Z. 199,
male; para types M. C. Z. 200, one male, five females).

ARGIOPIDAE.

Tetragnatha tincochacae, sp. nov.
Plate 18, fig. 2, 3.

Carapace and chelicerae yellowish, a black marginal line on each
side and a dusky median longitudinal stripe which widens and becomes
more diffuse upon the head. Sternum dusky brown or blackish.
Legs yellow, either unmarked or with femora and tibiae of anterior
pairs darker at distal ends and patellae also more or less darkened.
Ground color of abdomen silvery white as usual, overlaid with a net-
work of dark lines; venter with a median longitudinal brown stripe;
a narrow black stripe along each side expanded near its middle into a
larger spot; typically with a fine more or less broken longitudinal
black line on each side of dorsum just above the lateral stripe; dorsum
with a median longitudinal grey line giving off side branches as usual.

Posterior row of eyes (in female) conspicuously recurved as usual;
median eyes twice their diameter apart, nearly same distance from the
laterals. Lateral eyes separated by about their diameter, their tu-
bercles touching at base. Area of median eyes wider behind than in
front in ratio 6:5; shorter than wide (5:6), the length equalling the
width at anterior end. Anterior row of eyes recurved; median eyes
equal to the posterior medians or scarcely smaller, once and a half
their diameter apart and twice their diameter or more from the
laterals.

chamberlin: the arachnida. ' 239

Chelicera of male with fang having a nodule or cusp at base; the
conspicuous dorsal spine well removed from the line of teeth; upper
row of teeth with the second tooth from claw; the large tooth, con-
spicuously long and stout, six teeth proximad of large tooth. Tooth
nearest fang on lower margin the largest, the second small. Chelicera
of female lacking the dorsal spine; teeth of upper margin similar to
those of the male but the tooth nearest fang more reduced and the
"big" tooth shorter, (Plate 18, fig. 2).

Legs with rather numerous spines on the femora, these equal in
length to diameter of joint or but little longer; spines of more distal
joints fewer but proportionately longer.

Male. Length 6.1 mm. Length of cephalothorax 2.1 mm.; width,
1.4 mm.

fern. tib. + pat. met. tar. total

Leg I 6 mm. 6.8 mm. 5.8 mm. 1.8 mm. 19.4mm.

Legll 4.2 4.2 3.6 11 13.1

Leg III 2.3 1.6 1.5 .9 6.3

Leg IV 4 3.3 3 1.1 11.4

Locality — Tincochaca, 7,000 feet, August. (Type, M. C. Z. 201,
male; paratypes, M. C. Z. 202, three females).

Tetragnatha SCOPUS,^ sp. nov.
Plate 18, fig. 1.

Carapace and chelicerae light brown to testaceous. Legs brownish
yellow to yellow. Sternum from dusky yellow to dusky brown.\
Abdomen with ground-color silver-white covered with a fine network
of gre;s'. A longitudinal band over the venter in which the light
dots are much reduced and almost obliterated limited on each side
by a more silvery line on the outside of which is typically an irregular
line of dark grey or blackish. Dorsum of abdomen lighter than
sides and venter; a middorsal longitudinal grey line presenting several
uneven pairs of branches extending in a caudoectal direction, in
most specimens a large dark brown or black spot on each side near
middle of length contiguous with a dark stripe along each side.

Abdomen conspicuously broadened and gibbous in front; about
2.7 times longer than the greatest width; dorsal line convex.

1 oKoiTos, a watcher.

240 bulletin: museum of comparative zoology.

Posterior row of eyes recurved as usual; median eyes near two
and two fifths their diameter apart and about the same distance from
the laterals. Lateral eyes their diameter apart, their tubercles in
contact at base. Anterior row of eyes slightly procurved; median
eyes once and a half their diameter apart, twice their diameter from
the laterals. Area of median eyes much wider behind than in front
(ratio 5:4).

Upper margin of furrow of chelicera with six (or five) teeth of which
the one nearest the claw is much the largest; this separated from the
second by a wide space, the second, third, and fourth about equal in
size and spacing, the two most proximal smaller. Lower margin with
four teeth equally spaced but with the first considerably largest as
in the upper row, (Plate 18, fig. 1).

Legs with the spines of the distal joints long, slender, and sub-
appressed, clearly longer than in the succeeding species, T. quechua.

Female. Length 7.7 mm. Length of cephalothorax 2 mm.; width
1.6 mm.

fern. tib.+pat. met. tar. total

Leg I

4.9 mm. 5. 1 mm.

4 . 2 mm.

1.2 mm.

15.4 mm

Leg II

3.2 3.3

3.8

1

11.3

Leg III

1.7 1.7

1.2

.6

5.2

Leg IV

3.1 3

2.2

1

9.3

Tibia I, 4.2 mm.

Tibia IV,

2.3 mm.

Locality. — Conservidayo River, August. (Type, M. C. Z. 203,
female; paratypes, M. C. Z. 204, numerous specimens, chiefly imma-
ture males and females).

Tetragnatha quechua,^ sp. nov.

Plate 18, fig. 4.

Carapace and chelicerae light brownish yellow. Sternum somewhat
dusky yellow. Legs yellow, femora, and tibiae darker at the distal
ends but legs not truly annulate. Abdomen above silvery white in
numerous spots separated by a close network of fine grey lines; a
median longitudinal grey line throughout length, this widest at the
anterior end and giving off a principal pair of branches in a caudo-

• The Quechuas are the indigenous people of Peru and Ecuador.

CHAMBERLIN : THE ARACHNIDA. 241

lateral direction at about one fourth length from anterior end; on
caudal half on each side a row of four moderately large dark dots.
Venter of abdomen with ground-color and network like dorsum but
darker; a narrow median longitudinal stripe solid grey, a darker line
along border of venter on each side.

Posterior row of eyes strongly recurved; median eyes about two
and one fifth their diameter apart and about three times their diameter
from the laterals. Lateral eyes on each side separated by their diam-
eter, their tubercles distinct but contiguous at base or nearly so.
Anterior row of eyes recurved as usual; median eyes equal in size to
the posterior medians, one and a fifth their diameter apart and more
than three times their diameter from the laterals. Area of median
eyes much wider behind than in front, the ratio being 11:9, slightly
wider than long.

Upper margin of furrow of chelicera in female with nine teeth; of
these the first and third from the fang are the largest and are separated
by a wide space in which the much smaller second tooth lies, (Plate 18,
fig. 4). Teeth of lower margin with nearly the same arrangement as
those of the upper.

Abdomen long and slender, somewhat more than four times longer
than the greatest width, narrowly rounded caudad and gradually
widening cephalad; anterolateral corners subrectangular, not gibbous.

Legs with few spines of which those on femora are shorter than the
diameter of the joint and those on the more distal joints are longer.
Hairs more numerous on distal joints, somewhat oblique, of moderate
length.

Female. Length 11 mm. Length of cephalothorax 3 mm.; width,
1.9 mm,

fem. tib.-f-pat. met. tar. total

Leg I 8.1mm. 9.6 mm. 9.1mm. 2.1mm. 28.9 mm.

Leg II 5.3 5.3 5 1.5 17.1

Leg III 8.1 2.2 2.2 1 8.5

Leg IV 5.6 5.6 5 1.3 17.5

Tibia I, 8.8 mm. Tibia IV, 4.8 mm.

Locality.— Sorontoy, 7,000 feet, September. (Type, M. C. Z. 205,
female).

Tetragnatha sp.
Locality.— San Miguel, 6,000 feet, July. (M. C. Z. 328, immature).

242 bulletin: museum of comparative zoology.

Leucauge MARIANA (Keyserling) .

Meta tnariana Keyserling, Verb. Zool. bot. ges. Wien, 1880, 30, p. 560, pi. 16,

f. 10.
Argyroepeira mariana Keyserling, Spinnen Amerikas. Epeiridae, 4, 1893,

p. 345, pi. 18, f. 10.

Locality. — Conservidayo River, August. (M. C. Z. 211, one fe-
male). Previously known from Amable Maria and Lima, Peru.

Leucauge idonea O. P. Cambridge, var.

Argyroepeira idonea O. P. Cambridge, Biol. Cent. Americana, 1889, 1, p. 4,
pi. l,f. 7.

Locality. — Panama, June. (M. C. Z. 212, three females).

Meta explorans, sp. nov.
Plate 18, fig. 5.

Carapace yellow with a brown oeellate spot on each side of head, a
brown line from each spot extending to the thoracic groove. Sternum
black. Legs brown, typically with femora marked with one or two
dark annuli at distal end; patellae with one dark annulus; tibiae w^ith
three dark annuli and metatarsus with two. Sides of dorsum of
abdomen silvery white covered with a network of fine dark lines, the
median portion crossed by a rather narrow dark band extending from
spinnerets to anterior third or fourth of length, widening cephalad, its
lateral edges wavy and the anterior margin arcuate and in front of it
often an inverted v-shaped mark ; sides and anterior face of abdomen
dusky but showing the same areolation as the lateral portions of dor-
sum ; venter of abdomen with a median longitudinal blackish brown
band between spinnerets and genital furrow, this bordered on each
side by a white, finely areolated stripe.

Posterior row of eyes slightly recurved; median eyes five sixths
their diameter apart, one and a third their diameter from the laterals.
Lateral eyes on each side contiguous ; anterior with diameter exceeding
that of posterior in ratio 7 : 5. Anterior row of eyes recurved ; median
eyes equal to the posterior medians, two thirds their diameter apart,
once and a half their diameter from laterals. Median eves once and

chamberlin: the arachnida. 243

a third their diameter from the lower edge of clypeus. Area of median
eyes wider behind than in front in ratio 4:3, scarcely wider than long.

Sternum triangular, the caudal apex narrowly truncate between
last coxae. Nearly equal in width and length or scarcely longer
(25:24). Lower margin of furrow of chelicera with four teeth of
which the end ones are longer than the two intervening.

Length of female 5.5 mm.

Locality. — Conservidayo River, August. (Type, M. C. Z. 206;
paratypes, M. C. Z. 319, fifteen specimens). Tincochaca, 7,000 feet,
August. (M. C. Z. 207, two immature specimens).

Argiope argentata (Fabricus).

Aranea argentata Fabr., Ent. syst., 1775, 2, p. 414.

A very widespread species in tropical and subtropical portions of
the western hemisphere, occurring from the southern United States
to Patagonia.

Localities. — Panama, June. (M. C. Z. 208, one immature female).
Huadquina, 5,000 feet, July. (M. C. Z. 209, one female).

Gea panamensis, sp. nov.
' Plate 19, fig. 8.

Carapace brown with the sides dusky. Sternum blackish at sides,
a median longitudinal stripe yellow. First pair of legs with femora,
patellae, and tibiae black, the tibiae each with two narrow pale rings;
metatarsi yellow, with three black annuli; tarsi j^ellow. Second legs
similar to the first but femora mostly yellow marked with black espe-
cially distad, the third legs being like the second. Fourth legs colored
like the first, the femora being more nearly entirely black than that
of second or third. Abdomen above with a dark, wavy edged folium-
mark embracing a pale sagittate area in the anterior portion which is
limited by a dark longitudinal line; venter with a black median
longitudinal stripe which widens caudad to spinnerets limited by a
yellow line.

Posterior row of eyes very strongly procurved; median eyes twice
their diameter apart or nearly so. Lateral eyes contiguous, the pos-
terior on each side much the larger, its diameter being about thi-ee

244 bulletin: museum of comparative zoology.

times that of the anterior. Anterior row of eyes a little procurved;
median eyes carried forward on a conspicuously bulging tubercle,
six sevenths their diameter apart, distinctly larger than the posterior
medians and much larger than the laterals. Area of median eyes
wider behind than in front in about ratio 10:9. Clypeus viewed in
projection directly from in front distinctly narrower than anterior
median eye.

Sternum sub triangular, acutely pointed caudad. Winder than
long in about ratio 22:19.

Labium with apical portion subtriangular as usual. Endites typi-
cal.

Spines of legs few, moderate in length.

Locality. — Panama, June. (Type, M. C. Z. 210, one male).

Acacesia peruviana, sp. nov.
Plate 18, fig. 6.

Carapace brown. Sternum a paler brown. Legs brown, with the
femora of first pair black except at ends, the femur of leg II also dark
except proximally, the femora of posterior pairs dark at distal end
only. Abdomen above with a dusky or blackish subtriangular mark
or folium with wavy edges, this area embracing a wavy edged sagit-
tate mark in its anterior half much as in foliata; but the folium nar-
rower and more elongate; venter marked with a deep black median
longitudinal stripe from the epigynum to the spinnerets, this narrow-
ing moderately caudad.

Median eyes elevated on a conspicuous rounded tubercle projecting
dorsocephalad. Posterior row of eyes strongly recurved;, median
eyes facing ectad on tubercle, once and a half their diameter apart
or nearly so. Lateral eyes not elevated on a tubercle, separated by
more than their radius, the anterior one the larger. Anterior row of
eyes substraight; median eyes a little smaller than the posterior medi-
ans, nearly twice their diameter apart. Area of median eyes slightly
wider than long, of equal width anteriorly and posteriorly. Sternum
with a short triangular process at caudal end; longer than wide in
ratio 6.5 : 4.5.

Teeth of lower margin of furrow of chelicera three, these small
well separated.

Legs with spines of tibiae few, very slender, more sparse on the
metatarsi. Anterior legs much longer than the posterior.

chamberlin: the arachnida. 245

Cephalothorax with head strongly narrowed. Abdomen more
nearly strictly rhomboidal than in foliata.

Length of female 6.7 mm.

Locality — SsLii Miguel, 6,000 feet, July. (Type, M. C. Z. 213,
one female).

Among other features this species would seem to be clearly sepa-
rated from foliata by the presence of the conspicuous black band
along the venter.

EusTALA FUSCOVITTATA (Key serluig) .

Epeira fusco-vittata Keyserling, Sitzungsb. Naturw. ges. Isis, 1863, p. 129,

pi. 6, f. 18.
Cyclosa thorelli McCook, Amer. spiders, 1893, 3, p. 228, pi. 19, f. 11.
Eustala caudata Banks, Proc. Cal. acad. sci., 1898, 1, p. 255, pi. 15, f. 5.
Eustala fusco-vittata O. P. Cambridge, Biol. Cent. Americana, 1904, 2, p. 505,

pi. 48, f. 3, 4.

A species previously known from Mexico and the West Indies south
to Brazil and Paraguay.
Locality. — Panama, June. (M. C. Z. 214, three females).

Eustala andina, sp. nov.

Carapace with thoracic part pale yellow, the head light brownish.
Sternum light brown, a pale median longitudinal line. Legs testa-
ceous ; femora with a broad dark band at distal end, a broader median
one, and an indistinct proximal one; patellae dusky, black at distal
end; tibiae with three broad dark rings not sharply delimited; meta-
tarsi also with three dark rings; tarsi dark except at proximal end,
the mesal portion darkest. Abdomen deep brown in an area covering
most of the dorsum, this area with sides concave and converging
caudad, the edges wavy, dark, and bordered with yellow; sides
dusky or brownish over a yellowish background. Median portion
of venter more blackish, enclosing a yellow median dot midway be-
tween the genital furrow and the spinnerets and a pair of these nearer
the spinnerets.

Abdomen subtriangular with the anterior corners rounded and the
caudal end narrowly truncate, this truncate caudal end presenting
three low elevations or crenulations.

Posterior row of eyes conspicuously recurved; median eyes once
and a fourth their diameter apart and about three times their diameter

246

bulletin: museum of comparative zoology.

from the laterals. Lateral eyes equal in size, their radius or scarcely
more apart. Anterior row of eyes recurved; median eyes larger
than the posterior median eyes (11:8.5), just their diameter apart,
farther from the laterals. Area of median eyes wider in front than
behind in the ratio 6: 5 and wider in front than long in the ratio 10: 9.

Lower margin of furrow of chelicera with three teeth of which the
most proximal is largest; upper margin with three teeth of which
the median is much the stoutest and longest.

Sternum with caudal portion narrowly triangular as in the suc-
ceeding species, and the caudal process similarly narrowly truncate.

Spines of legs sparser, short and slender.

Female. Length, 8 mm. Length of cephalothorax, 3.2 mm.;
width, 2.7 mm.

fem. tib.+pat.

met.

tar.

total

Leg I

4 mm. 4.2 mm.

3 mm.

1 . 1 mm.

12.3 mm

Leg II

3.3 4

2.6

1

10.9

Leg III

2.3 2.2

1.3

.9

6.7

Leg IV

3.7 4

2.6

1.1

11.4

Tibia I, 2.8 mm.

Tibia

IV

, 2.2 mm.

Locality. — Huadcjuina, 5,000 feet, July. (Type, M. C. Z. 215,
one female, not quite adult).

EusTALA monticola, sp. nov.
Plate 18, fig. 7.

Cephalothorax low, only moderately convex; head low; thoracic
groove deep, longitudinal. Abdomen in outline from above subtri-
angular.

Carapace yellow of very dilute brownish tinge. Sternum and coxae
of legs beneath yellow. Legs testaceous, the femora with a wide
dusky ring at distal end and one near middle with in some legs a nar-
rower one near base, these rings in the female type not sharply limited,
the dusky color being more or less diffused, but in the male more
sharply defined; tibiae with three broad dusky rings between which
the dark color may also be more weakly diffused; metatarsi of pos-
terior pairs with three annuli of which the proximal one is least dis-
tinct, these less clearly defined on the anterior pairs; tarsi with a
submedian dark annulus. Abdomen above dark greyish brown,
embracing numerous fine dots and spots of white, several pairs of
transverse dark lines running from outside a little caudad of mesad

chamberlin: the arachnida. 247

but not meeting at middle, the median dorsal line being occupied by a
fine dark line which presents short lateral branches; in the male
there are caudad on each side two large proximate white spots and
farther forward two short white marks. General background of
venter of abdomen pale; just back of epigynum a median white spot
on each side of which is a black spot of about equal size and farther
laterad a longitudinal fine black line which reaches to a dusky area
about the spinnerets.

Posterior row of eyes strongly recurved; median eyes a little more
than their diameter apart, nearly three times their diameter from the
laterals. Lateral eyes on each side their radius apart, scarcely differ-
ing in size. Anterior row of eyes recurved ; median eyes a little smaller
than the posterior medians (diameters as 7:8), twice their diameter
apart, and between two and a half and three times their diameter from
the laterals. Area of median eyes wider in front than behind in ratio
25:22; wider in front than long (25:22).

Sternum longer than wide in about ratio 5:4. Caudally triangular,
the tip narrowly truncate.

Lower margin of furrow of chelicera with three stout conical teeth;
upper margin also with three.

Legs well spined, the spines moderate, appressed, blackish.

Female. Length, 7.3 mm. Length of cephalothorax, 3.5 mm.;
width, 2.9 mm.

fem. tib.+pat.

met.

tar.

total

Leg I

4.1 mm. 5.2 mm.

3 mm.

1.2 mm.

13.5 mm

Leg II

4 4.7

3

1

12.7

Leg III

2.3 2.1

1.2

1

6.6

Leg IV

4.6 3.9

2.2

1.1

11.8

Tibia I, 3.8 mm.

Tibia

IV, 2.3.

Locality.— San Miguel, 6,000 feet, July. (Type, M. C. Z. 216,
female; paratype, M. C. Z. 217, immature male).

Aranea sp. a.

An immature female from Huadquina, 5,000 feet, July. (M. C. Z.

218).

Aranea sp. b.

An immature female of a second species also from Huadquina.
(M. C. Z. 219).

248 bulletin: museum of comparative zoology.

Aranea sp. c.

An immature female of uncertain species from San Miguel, 6,000
feet, July. (M. C. Z. 220).

Aranea sp. d.

An immature female from Paltaybamba, 5,000 feet, August.
<M. C. Z. 221).

Aranea abunda (Taczanowski).

Epeira abunda Taczanoski, Horae Soc. ent. Ross., 1878, 14, 152, pi. 1, f. 7.

Reported by Taczanowski from many localities in Peru.

Localities.— San Miguel, 5,000 feet, September. (M. C. Z. 222,
one somewhat variant female). Tincochaca, 7,000 feet, August 9.
(M. C. Z. 223, one female also apparently variant).

1^ Aranea nigroventris (Taczanowski).

Epeira nigroventris Taczanowski, Horae Soc. ent. Ross., 1878, 14, p. 151, pi. 1,
f. 6.

Locality.— Cuzco, 11,500 feet, July 6, 12. (M. C. Z. 320, one female;
July 12. M. C. Z. 224, numerous specimens).

Aranea zelotypa (Keyserling).

Epeira zelotypa Keyserling, Verb. Zool. bot. ges. Wien, 1882, 32, p. 202, pi. 15,
f. 7.

Previously known from San Mateo, Peru.

Localities. — Tincochaca, 7,000 feet, August. (M. C. Z. 225, one
female). Urubamba, 9,500 feet, July. (M. C. Z. 226, one female).

Aranea orina,^ sp. nov.
Plate 19, fig.[3.

Carapace brownish yellow, dusky in an area on each side of caudal
portion of head, with also a transverse row of four small black dots
across the furrow. Sternum and coxae of legs beneath clear yellow.
Endites and labium dusky yellow, whitish across tips. Legs yellowish;

' 'opeicOT, pertaining to mountains.

chambeelin: the arachnida. 249

femora I and II with two broad dark annuli, one just proximad of mid-
dle and one at about distal third; femora III with a narrower band
at distal end; femora IV black from proximad of middle to distal end;
patellae I and II unmarked, but III and IV black at both ends; ante-
rior tibiae with two black bands, one about middle and one at distal
end; metatarsi and tarsi dark at distal end, the posterior metatarsi
also with a vague dark ring near middle. Bulb of palpus blackish.
Abdomen yellow ; venter dusky back of the genital furrow except in a
narrow transverse band in front of the spinnerets, the dark area divided
by a longitudinal median yellow line; dorsum with a folium outlined,
a black mark at base and a diamond shaped one near the middle.

Carapace broad, cordate, being very strongly narrowed cephalad.
Groove, fine, longitudinal.

Posterior row of eyes strongly recurved; median eyes a little more
than their diameter apart, four times or more their diameter from lateral
eyes. Lateral eyes on each side separated by about their radius. Area
of median eyes conspicuously wider in front than behind {cir. 15: 11);
and wider in front than long in ratio 15:13. Anterior row of eyes
nearly straight, the median eyes clearly larger than the posterior me-
dians (ratio of diameters about as 10: 7), nearly once and a half their
diameter apart and two times their diameter from the laterals.

Sternum not fully two thirds as wide as total length; acutely nar-
rowed caudad.

Lower margin of furrow of chelicera with three teeth; upper margin
also with three.

Legs with all joints excepting the tarsi aculeate or spined as usual.

Patella of palpus of male with a very long spine above; tibia on
ventral side extended into a thin blade-like process as wide as the
length of the joint; process of cymbium on exterior side narrow
proximally and expanded distad into a tomahawk form with the blade
partially double.

Male. Length 7.7 mm. Length of cephalothorax 3.7 mm.; width
3.1 mm.

fern.

tib.+pat.

met.

tar.

total

Leg I

5.6 mm.

6. 1 mm.

3.9 mm.

1.5 mm.

17.0mm

Leg II

4.2

5

3.2

1.4

13.8

Leg III

3

2.7

1.6

1

8.3

Leg IV

4

4

2.9

1.1

12.0

Tibia

I, 4.3 mm.

Tibia IV,

2.8 mm.

Locality.— S&n Miguel, 6,000 feet, July. (Type, M. C. Z. 227,
one male).

250 bulletin: museum of comparative zoology.

Aranea quechuana, sp. nov.
Plate 19, fig. 1.

Carapace testaceous, unmarked. Sternum dusky brown or black-
ish. Labium and endites dusky, pale across tips. Legs pale yellow;
femora with a wide dusky ring near one third the distance from the
distal end or in less deep shade including also the distal portion;
posterior patellae darker and the posterior tibiae with darker annulus
at distal end. Abdomen whitish yellow; venter with a black area
imiUediately back of the genital furrow but this typically not extend-
ing to spinnerets, weakly dusky over each anterolateral corner and
caudad from there over upper portion of each side; dorsum with a
fine median longitudinal dark line extending from in front of middle
caudad and sending off laterad a number of pairs of fine branches, a
row of from two to four black dots on each side caudad of the middle,
these two rows somewhat converging toward the spinnerets.

Carapace broad and low; groove longitudinal, distinctly impressed^
crossed behind bj- a vague transverse impression.

Posterior row of eyes distinctly recurved as usual ; median eyes only
five eighths their diameter apart, two and a half times their diameter
from the laterals. Lateral eyes subequal, contiguous, raised on a
common low black tubercle. Anterior row of eyes slightly recurved;
median eyes a little longer than the posterior medians (diameters as
5:4), their diameter apart, once and a third their diameters from the
laterals. Area of median eyes wider in front than behind in ratio
14:11; wider than long in about ratio 14:13.

Sternum slenderly extended caudad between the posterior coxae.

Legs abundantly aculeate or spined as usual; two rows of spines
on anterior face of tibia II shorter and distinctly stouter than those
found elsewhere.

Patella of palpus above at distal end with two long slender spines.
Process of cymbium undivided, somewhat clavately expanded distad,
moderately curved.

Male. Length 6.2 mm. Length of cephalothorax 3.1 mm.; width
2.8 mm. Tibia I, 3.6 mm.; Tibia IV, 2.1 mm.

fern. tib. + pat. met. tar. total

Leg I 4.25 mm. 5 mm. 4 mm. 1.5 mm. 14.75 mm.

Leg II 4 4.1 3 1.2 12.3

Leg III 2.5 2.25 1.9 1 7.65

Leg IV 3 3.2 2.6 1 9.8

chamberlin: the arachnida. 251

Localities — Huadquina, 5,000 feet, July. (Type, M. C. Z. 228,
male; paratypes, M. C. Z. 229, one immature male). Lucma.
(M. C. Z. 329, one male).

Aranea tigana/ sp. nov.
Plate 19, fig. 2.

Carapace pale brown, a broad stripe along each side chocolate or
blackish. Sternum blackish. Legs yellow; femora dark at distal ends
in a broad band; patellae and tibiae also dusky at distal ends. Abdo-
men with dorsum white faintly tinged with yellowish, the dorsal light
area enclosing four dark dots and in the caudal portion a median black
line with side branches; the sides blackish, the lateral dark bands con-
verging caudad and meeting considerably in front of the caudal end.

Posterior row of eyes strongly recurved; medians their radius apart,
two and a fourth or more times their diameter from the laterals.
Lateral eyes nearly contiguous, the anterior one the larger, elevated
on a low common tubercle. Area of median eyes wider than long in
about ratio 25 : 22 and wider in front than behind in about ratio 5 : 4.
Anterior row of eyes distinctly recurved; median eyes larger than the
posterior medians (diameters as 5:4), three fifths their diameter apart
and not fully their diameter from the laterals.

Sternum of usual shape, being conspicuously narrowed caudad.

Labium of usual form, distinctly triangular.

Legs abundantly spined as usual; spines on the anterior face of
tibia II shorter and stouter, in two series.

Palpus with patella above having two long spines at distal end.
Tibia flattened and extended ectad as in various other species. Pro-
cess of cymbium of nearly uniform width throughout, bent nearly at
right angles near middle.

Male. Length 6.6 mm. Cephalothorax, length 3 mm.; width,
2.2 mm.

fem. tib. + pat. met. tar. total

Leg I 3.8 mm. 4.4 mm. 3.2 mm. 1.6 mm. 13.0mm.

Legll 3.1 3.3 2.5 1.1 10.0

Leg III 2.2 2 1.2 1 6.4

Leg IV 3.1 2.7 2 1 8.8

Tibia I, 3.2 mm.

Locality. — Lucma, 7,000 feet, August. (Type, M. C. Z. 230, one
male).

' Gosiute tigana, teguna, close to.

252 bulletin: museum of comparative zoology.

Aranea compsa/ sp. nov.
Plate 19, fig. 6.

Carapace black excepting the anterior region of the head about the
eye-area which is abruptly lighter, pale brown; the pale color extend-
ing farther caudad on lower side of head clothed with long grey or
white hair. Sternum black with a bright yellow median longitudinal
stripe which is interrupted across its anterior portion. Labium and
endites black, pale across tips. Legs with femora pale yellow-brown
proximally, the distal half blackish; patellae black; tibiae black
excepting for a light annulus at the middle; metatarsi and tarsi pale
with no distinct black annuli or metatarsus black at extreme distal
end. Abdomen brownish grey; venter, anterior face and a broad
band reaching back across each anterolateral corner and along side
black. A series of yellow dots and marks on each side between black
of venter and that of side ; dorsum covered with a broad folium-mark
a little darker than adjacent parts, the folium embracing a narrow
median stripe limited by whitish marks and extending over entire
length and marked with light colored paired transverse lines each side
of this middle stripe, especially caudad.

Posterior row of eyes decidedly recurved; median eyes three fourths
their diameter apart and two and a half or a little more their diameter
from the laterals. Lateral eyes on each side contiguous, moderately
elevated. Anterior row of eyes clearly recurved; median eyes a
little smaller than the posterior medians, more than their diameter
apart (once and a half), nearly two and a half times their diameter
from the laterals. Area of median eyes nearly equal in length and
breadth or slightly wider; wider in front than behind in ratio 11 : 10.

Thorax of typical form, smooth, without processes; head narrow;
groove transverse, not profound.

Labium with distal end triangular.

Sternum about six sevenths as wide as long, constricted to a narrow,
distally rounded tongue behind.

Upper margin of furrow of chelicera with four teeth ; lower margin
with three.

Legs strongly aculeate, the spines long and slender, all joints ex-
cepting the tarsi being well armed.

■ KoiJ.4'CK, pretty.

chamberlin: the arachnida.

253

Epigynum with scape short; broad at base but narrowed to a
slender tip projecting caudoventrad, (Plate 19, fig. 6).

Female. Length, 6.7 mm. Length of cephalothorax, 2.9 mm.;
width, 2.3 mm.

fern. tib.+pat.

met-

tar.

total

Leg I

3.25mm. 3.2mm.

3 mm.

1 . 1 mm.

10.55 mm.

Leg II

3.1 3

2.2

1

9.3

Leg III

2.1 2

1

0.9

6.0

Leg IV

3 2.S

2

1

8.8

Tibia I, 2.1.

Tibia IV,

1.9 mm.

Localities.— OWa^ntSiytamho, 9,000 feet, July. (Type, M. C. Z.
231, female; paratype, M. C. Z. 232, female). Urubamba, 9,500
feet, July. (M. C. Z. 233, female).

Aranea plesia,^ sp. nov.
Plate 19, fig. 5.

Carapace deep chestnut or blackish, paler along middle part of
head. Sternum black. Legs light brownish yellow, the joints an-
nulate with dark about distal ends, the tibiae of anterior pairs also
with a subbasal and narrow basal band. Abdomen with median
portion of venter black, the black area limited on each side by a nar-
row yellow longitudinal stripe ; lower portion of side black, the rest of
sides and the dorsum paler, whitish yellow covered with a close net-
work of fine dark lines ; back of middle of length a row of short, solid
black, transverse spots or lines each side of the middle, the anterior
of these bordered in front with white, on anterior portion of dorsum
two curving white lines meeting at an angle on the median line, the
figure thus formed having its edges lined with black.

Posterior row of eye conspicuously recurved; median eyes their
diameter apart and between two and a half and three times their
diameter from the laterals. Lateral eyes subcontiguous and slightly
elevated together as usual. Anterior row of eyes straight or slightly
recurved; median eyes equal in diameter to posterior medians,
nearly one and three sevenths their diameter apart and about the
same distance from the laterals or but slightly more. Area of median
eyes wider in front than behind (11: 10), nearly equal in length and
breadth.

1 TrXTjffios, close to.

254

bulletin: museum of comparative zoology.

Sternum of usual general shape; caudad abruptly narrowed into a
narrow tongue projecting between coxae.

Legs spined as Usual; spines of tibiae II not specially modified.

Scape of epigynum long and a little clavate, projecting caudad,
(Plate 19, fig. 5).

Female. Length, 5.1 mm. Length of cephalothorax, 2.5 mm.;
width, 2 mm.

fem.

tib.+pat.

met.

tar.

total

Leg I

2.25 mm.

3.1 mm.

2.1 mm.

1 mm.

8.45 mm.

Leg II

2

2.3

1.6

.7

6.6

Leg III

1.3

1.8

.9

.5

4.5

Leg IV

2.1

2.1

1.3

.9

6.4

Tibia

I, 2 mm.

Tibia IV,

1.2 mm.

Locality.— Sorontoy, 7,000 feet, September. (Type, M. C. Z. 234).

Aranea SANTA, sp. nov,
Piatt 19, fig. 10.

Carapace with pars cephalica testaceous, the pars thoracica abruptly
darker, dusky. Sternum testaceous, somewhat dusky, with a clear
white T-shaped mark of which the cross-piece is at the anterior end.
Labium and endites dusky testaceous. Legs light brown or testa-
ceous, the joints vaguely darker at distal ends but not distinctly
annulate. Abdomen above whitish, covered with a close network
of fine dark lines and on each side with a series of dark oblique areas;
back of middle a median longitudinal black line and a little each side
of and parallel with this a broader dark stripe embracing about five
small triangular black spots, these two more lateral stripes converging
moderately caudad to spinnerets, a small white spot each side of and
a little in front of the spinnerets. Anterior face of abdomen brown
or dusky brown.

Abdomen suborbicular or but slightly extended at ends, not at all
angulate or tuberculate.

Posterior row of eyes conspicuously recur\'ed as usual ; median eyes
only about six sevenths their diameter apart, twice their diameter
from the laterals. Lateral eyes contiguous, subequal. Anterior row
of eyes distinctly recurved; median eyes equal to the posterior
medians, a little more than their diameter apart, and the same dis-

chamberlin: the arachnida. 255

tance or but little farther from the laterals. Area of median eyes
equal in length and breadth, wider in front than behind in ratio
10:9.

Sternum of usual general shape; posterior portion triangular,
acutely pointed caudad.

Legs abundantly armed with the usual long slender spines.

Length of female 4.7 mm.

Locality.— Santa Ana, 3,000 feet, August. (Type, M. C. Z. 235,
one female).

Aranea sexta, sp. nov.
' Plate 19, fig. 7.

Carapace broad behind with the head narrow; thoracic furrow
transverse. . '"'. '

Abdomen angulate above on each side near middle, (Plate 19,
fig. 7).

Carapace pale testaceous, the head-itgion darker and with a fine
light median longitudinal line extending caudad from between eyes.
Sternum light testaceous. Legs testaceous, not at all annulate with
dark. Abdomen in front of level of angles pale testaceous or even
of whitish cast cephalad; a white transverse band at level of angles,
the abdomen behind this a darker brown; venter pale, somewhat
whitish mesally, brown laterally and caudally.

Posterior row of eyes recurved as usual ; median eyes circular, their
diameter apart or slightly less, nearly three times their diameter from
the laterals. Lateral eyes on each side nearly contiguous, the anterior
a little the larger. Anterior row of eyes clearly recurved; median
eyes equal in diameter to the posterior medians, once and a half their
diameter apart and no farther from the laterals. Area of median
eyes much wider in front than behind (16:13) and a little wider in
front than long (about 8:7).

Sternum with process between posterior coxae long, distally rounded.

Labium and endites of the usual form.

Femora and patellae of legs unspined; the tibiae, metatarsi, and
tarsi with fewer long spines and numerous slender short, seriate or
subseriate ones chiefly on the anterior surface.

Length of female 4.4 mm.

Locality. — Panama, June. (Type, M. C. Z. 236, female).

256 bulletin: museum of comparative zoology.

Aranea duocypha/ sp. nov.
Plate 18, fig. 8-10.

Abdomen above with two conspicuous conical tubercles, one on
each side toward anterolateral corner.

Carapace testaceous, unmarked. Sternum and coxae of legs be-
neath yellow. Legs dark testaceous, the tibiae obscurely marked
with three darker bands, the metatarsi and tarsi darker. Palpi dark
at tips. General color of abdomen above yellowish; a black line
extending transversely between the tips of the two angular tubercles,
this line bent forward at middle and continued cephalad as a median
line; caudad of this line a series of other parallel transverse black
lines, (Plate 18, fig. 8). Venter yellow, marked obscurel;^' with a
network of somewhat darker lines; epigynum very dilute chestnut.

Posterior row of eyes conspicuously recm-ved medians nearly once
and a third their diameter apart ; nearly twice their diameter from the
posterior laterals. Lateral eyes on each side separated by about
their radius. Anterior row of eye straight or slightly procurved;
medians slightly smaller in diameter than the posterior medians, one
and two thirds their diameter apart and nearly an equal distance from
the laterals; medians just their diameter from the lower edge of the
clypeus. Area of median eyes wider in front than behind (10:9)
and wider in front than long (also as 10:9).

Caudal end of sternum narrowly triangular, not projecting between
coxae; longer than wide in ratio 47:43, (Plate 18, fig. 9).

Labium and endites of the usual form.

Epigj^num with scape very broad, long triangular, projecting much
caudad of the genital furrow, (Plate 18, fig. 10).

Legs sparsely and weakly aculeate, the aculei more numerous
distally and on anterior surface.

Length 4.2 mm.

Loca%.— Huadquina, 5,000 feet, July. (Type, M. C. Z. 237,
one female).

Aranea calotypa,^ sp. nov.

Plate 19, fig. 4.

Abdomen in outline subelliptic, narrowly obovate when viewed
from above, being narrower caudad than cephalad.

1 Svo, two, KU0OS, a hump. /

' xaXAj, pretty, ri'iros, mark.

chamberlin: the arachnida. 257

Carapace yellowish or pale testaceous. Eyes black. Sternum
yellow. Legs clear yellow, wholly unmarked. Abdomen beneath
pale yellowish or whitish yellow, unmarked, but the spinnerets ab-
ruptly darker and together appearing like a black spot at the tip of
the abdomen; lower part of sides with fine dark dots and streaks;
dorsum above dark grey, the color mesally solid and laterally in a
fine network enclosing light spots, along the middle line with a series
of whitish spots extending over whole length, the most conspicuous
marks being three pairs of widely separated black dots and in addition
toward caudal end a quadrangle of four more closely approximate
black spots.

Posterior row of eyes strongly recurved; medians four fifths their
diameter apart and three and three fifths their diameter from the
laterals. Lateral eyes contiguous, nearly equal. Anterior row of
eyes a little recurved; medians their diameter or a little more apart,
not quite fully three times their diameter from the laterals. Area of
median eyes a little wider in front than behind (13: 121), very slightly
wider than long.

Sternum longer than wide in the ratio 11:9, caudad narrowly
triangular, the caudal tip not sharply defined.

Labium and endites of typical general form.

Legs of male conspicuously spined as usual, the spines of tibiae,
especially of tibia I, longer, stouter and more numerous than those of
femora and patellae as usual, the metatarsi and tarsi with but few
spines. Legs in the female more weakly aculeate as usual.

Process of cymbium in palpus of male on ectal side of base, arising
from a broad base, curved, ending in a swollen tip or button. Patella
with a single long spine at distal end above.

Length 5-6 mm.

Localities. — Below Lucma. (Type, M. C. Z. 238; paratypes,
M. C. Z. 239, one adult male, one immature male, and three immature
females). Tincochaca, 7,000 feet, August. (M. C. Z. 240, one male).

Anawixia,^ gen. nov.

Thoracic furrow deep, longitudinal, continued upon head, not
separated from cervical depression.

Area of median eyes much wider in front than behind. Posterior
median eyes decidedly smaller than the anterior, separated by a
distance somewhat greater than their diameter. Anterior median

1 'avd, towards, Wixia.

258 bulletin: museum of comparative zoology.

eyes prominent. Lateral eyes on each side contiguous, on a common
prominent tubercle. Posterior row of eyes strongly procurved.

Clypeus much narrower than the eye area; but little exceeding
diameter of an anterior eye.

Labium not discrete from sternum ; wider than long.

Femur I with three very short ventral spines in a series, five longer,
stouter spines along antero ventral surface; tibia I with two ventral
spines, five on anterior surface and two on the posterior. Femur II
with six ventral spines ; tibia II moderately incrassate, with two stout
ventral spines and on anterior surface with two series of short, stout
spines, six in each series. Femur III unarmed beneath. Femur IV
beneath with a series of about eight, mostly very short, spines. Tro-
chanter IV in type with a single stout spine beneath. Coxa I with a
hook as in related genera. Posterior tarsi setose beneath ; with some
accessory claws or stout seriate bristles at the distal end.

Abdomen elongate; much extended caudad beyond the spinnerets
which are ventral and submedian in position and with a conspicu-
ous cylindrical slender caudal process; with spinous points above
at proximal end and near base of caudal process.

Coxa of pedipalp (male) with a cone-shaped spur distally and femur
with the usual chitinous ridge. Patella with a single apical spine.
Tarsal sheath with a strongly chitinized non bifid process at base.

Genotype. — Anawixia atopa, sp. nov.

Differing from Cyclosa, in the wider separation of the posterior
median eyes, and in the spining of the legs.

Anawixia atopa,^ sp. nov.
Plate 20, fig. 1-3.

Carapace with thoracic part black, the head yellow. Sternum
blackish over a yellow background; labium and endites similar except
at tips which are clear yellow. Chelicera yellowish, dusky especially
proximally. Leg I with femur dusky beneath, deeper black distad,
yellowish above; patella black; distal joints yellowish or the tibia
somewhat dusky or black distad. Other legs with the femora yellow
except at distal end where dusky or black, and two less distinct dark
rings, one at middle and one more proximad, the tibiae black at
distal end beneath as in Leg I. Abdomen blackish; dorsum with a

' (XTowos, strange.

chamberlin: the arachnida. 259

small pale median spot at base and vague paler markings over middle
and caudal regions ; venter with a vague pale line on each side and also
paler in front of genital furrow.

Cephalothorax broadly ovate; the head much narrowed, projecting
conspicuously forward above and over the clypeus, highest midway
between eyes and caudal end.

Abdomen slender, narrowed caudad, at caudal end produced caudo-
dorsad into a long, slender, cylindrical process, on each side at base
of which is an acute spinous process or point, an acute process or point
also occurring toward each anterolateral corner; lower, less distinct
cornicles also occurring elsewhere over the dorsum, (Plate 20, fig. 2).

Posterior row of eyes strongly recurved; median eyes a little more
than their diameter apart, and between three and four times their
diameter from the laterals. Lateral eyes on each side contiguous,
borne upon a common low tubercle at a considerably more ventral
level than the medians, the posterior one smaller than the anterior.
Quadrangle of median eyes distinctly narrower caudad than cephalad
(4:5), wider in front than long (about 25: 22), the median eyes raised
on a broad, common elevation. Anterior row of eyes straight or
slightly recurved; median eyes larger than the posterior medians
(diameters about as 9 : 7), not fully their diameter apart, two and a half
times their diameter from the laterals; median eyes projecting con-
spicuously forward over the receding clypeus. Median eyes about
once and a third their diameter from edge of clypeus but seen in
projection directly in front appearing only about one half their diame-
ter removed from this edge, (Plate 20, fig. 1).

Labium not distinctly separated from the sternum; much wider
than long; distal portion triangular, its sides straight and meeting
at an angle in the median line; proximal portion transversely de-
pressed. Endites well bent, curving over the labium, the inner side
conspicuously curved.

Sternum longer than wide in nearly ratio 13:10 or 13: 11; widest
at level between second and tTiird legs, abruptly indented as usual
opposite bases of first coxae slender acute extensions between coxae
of legs.

Spinnerets borne on venter not far caudad of middle of length,
the abdomen projecting widely over and caudad of them.

Trochanter IV with an acute black spur beneath at the distal
end; tibia I with a series of slender spines beneath, (Plate 19, fig. 3);
tibia II proportionately stouter, more strongly spined with short
stouter spines on anterior surface, having also two spines toward

.9

9.1

.6

6.05

1

8.8

260 bulletin: museum of comparative zoology.

base on ventral surface; femora I and II with series of ventral spines;
femur III with no spines beneath.

Male. Length (not including caudal process), 7.2 mm. Length
of cephalo thorax, 3.1 mm.; width 2.25 mm.

fem. tib.+pat. met. tar. total

Leg I 4.1mm. 4.3 mm. 2.3 mm. 1mm. 11.7 mm.

Leg II 3 3.1 2.1

Leg III 2.25 2.1 1.1

Leg IV 2.9 2,9 2

Tibia I, 3 mm.

Locality — San Miguel, 6,000 feet, July. (Type M. C. Z. 241, one
male).

Scoloderus hybus,^ sp. nov.
Plate 19, fig. 9.

Caudal region of head strongly elevated as usual, the tubercle
rounded; posterior declivity of cephalothorax steep. Abdomen
large and very high, its surface clothed with numerous very short,
straight and acutely pointed hairs as is also that of the carapace.

Carapace and legs dusky red-brown, the latter obscurely annulate.
Sternum dusky brown, the coxae of legs beneath paler. Abdomen
above brownish grey, with vague chevron-lines caudad, the anterior
face and sides dusky with on the anterior face a small median white
spot, the dark of the sides in part in oblique stripes more vaguely
extending upon the dorsum; venterdusky or blackish mesally, paler
laterally.

Posterior row of eyes strongly recurved ; median eyes one and three
fifths their diameter apart, four times their diameter, or somewhat
more, from the laterals. Lateral eyes at clypeal corners as usual, con-
tiguous. Anterior row of eyes conspicuously procurved in the typical
manner, the median eyes equal in size to the posterior medians, one and
two fifths their diameter apart, twice their diameter from lower edge
of clypeus, the latter narrower than eye area. Area of median eyes
wider than long (9:7); wider behind than in front in the ratio 9:8.

Lower margin of furrow of chelicera with three teeth of which the
most proximal is largest; upper margin with four teeth.

Spines of legs few, long.

Epigynum small, of form, (Plate 19, fig. 10).

' 'v0di, hump-backed.

chamberlin: the arachnida. 261

Length of female 4 mm.

Locality. — Paltaybamba, 5,000 feet, August. (Type, M. C. Z.
242, one female).

MiCRATHENA CALa/ sp. nOV.

Plate 20, fig. 5.

Carapace brown, the lateral and caudal slopes of the thoracic part
dusky. Anterior pairs of legs dark brown, coxae and proximal ends
of femora yellowish; posterior pairs yellowish or light testaceous.
Sternum solid black. Abdomen above white of a slight yellowish
tinge, a dusky mark at each anterior corner and a spot or two in line
with this farther caudad; venter and lower portion of sides blackish,
the caudal end between apices of lobes also black.

Abdomen bifid behind, each lobe ending in two spines, one above
the other; near base of lobe on each side a spinous point and one
farther forward toward anterior corner.

Posterior row of eyes recurved as usual; median eyes just their
diameter apart, a little more than three times as far from the lateral.
Lateral eyes on each side nearly contiguous, the anterior somewhat
the larger. Anterior row of eyes recurved; median eyes slightly
smaller than the posterior medians, four fifths or a little less their
diameter apart, their diameter from lower margin of clypeus, more
than three times their diameter from the laterals. Area of median
eyes wider behind than in front in about ratio 15:13, nearly equal in
length and breadth.

Sternum shield shaped; longer than wide nearly in ratio 7:6.

Lower margin of furrow of chelicera with three teeth of which the
most proximal is largest.

Spines on anterior face of femora longest.

Length of female 5.3 mm.

Locality.— Sein Miguel, 6,000 feet, July. (Type, M. C. Z. 243,
female).

Gasteracantha raimondi Taczanowski, var.

Horae Soc. ent. Ross., 1879, 15, p. 106, pi. 1, f. 25, 26.

Previously known from various other localities in Peru and from
Brazil.

Locality.— Huadquina, 5,000 feet, July. (M. C. Z. 246, one female).

1 KaXoi, beautiful.

262 bulletin: museum of comparative zoology.

MIMETIDAE.

Gelanor innominatum/ sp. nov.

Plate 20, fig. 6.

Carapace from greenish yellow to light brown, either not at all
distinctly marked or, in lighter individuals, with two dark marks on
caudal region of head and one on each side of the pars thoracica.
Sternum yellow to pale brown, with the labium a little darker. Cheli-
cera from yellow to pale brown of a slightly reddish cast. Legs yellow ;
first pair typically with four black or dark brown spots on caudal
side of femur with a fine unbroken black line along the dorsal surface;
tibia dark about the distal end. Second legs marked like the first but
the spots narrower and together appearing more like a broken line.
Femora of legs III and IV with a fine median longitudinal dorsal dark
line at distal end. Tibia IV and metatarsus IV with a fine median
dorsal longitudinal dark line over entire length. Abdomen above
dark brown over proximal half, testaceous over caudal; typically
with five narrow transverse light stripes with each margin limited
by a fine black line, two pale spots in front of the first of these and a
black mark caudad of the last ; venter yellow or somewhat testaceous,
dusky in front of the spinnerets and just in front of the genital furrow.

Posterior row of eyes recurved ; median eyes a little more than their
radius apart, slightly less than three times their diameter from the
laterals which are of equal diameter; laterals slightly larger than the
anterior laterals with which they are contiguous. x'Vnterior median
eyes about one and two thirds the diameter of the laterals ; four fifths
their diameter apart, and about once and a fifth their diameter from
the lateral one on each side.

Labium with sides strongly convex; much narrowing distad; distal
margin subtruncate or slightly convex.

Sternum less than three fifths as wide as total length inclusive of
the process between the posterior coxae.

The first two pairs of legs are much longer and stouter than the two
posterior pairs; but the second are decidedly smaller and less stout
than the first. Anterior tibiae moderately bowed, the metatarsi
more conspicuously so.

' innominatus, nameless.

chamberlin: the arachnida. 263

Tibia I with seven of the longer spines and four or five smaller
distally curved ones between each two longer ones, these smaller ones
increasing in length distad as usual; four large spines on metatarsus
I, the number of small spines between each two large ones increasing
distad, those distad of the last large spine much more numerous.
Large spines of tibia II mostly five, of metatarsus II three.

Length of female, 5.7 mm. Length of cephalothorax, 3 mm. ; width,
2.1 mm.

Locality — Ssin Miguel, 6,000 feet, July. (Type, M. C. Z. 244,
female; para types, M. C. Z. 245, two females).

THOMISIDAE.
MisuMENOPS CONSPERSA (Keyserling) .

Misumena conspersa Keyserling, Spinnen Amerikas. Laterigradae, 1880, 1,
p. 107, pi. 2, f. 59.

Localities. — Huadquina, 5,000 feet, July. (M. C. Z. 321, one
female). Santa Ana, 3,000 feet, August. (M. C. Z., 249, one male,
one female) .

Previously known from Guadalupe, Vacarmayu, and San Malu,
Peru.

MisuMENOPB PALLENS (Keyserling).

Misumena pollens Keyserling, Spinnen Amerikas. Laterigradae, 1880, 1, p.
96, pi. 2, f . 52.

Localities. — Huadquina, Peru, 5,000 feet, July. (M. C. Z. 247,
one female). Panama, June. (M. C. Z. 248, one female).

Previously known from Guatemala, Colombia, Peru, and Chile.

Thanatus taquarae Keyserling.
Spinnen Amerikas. Bras. Spinnen, 1891, 3, p. 252, pi. 10, f. 191.

Localities. — Huadquina, 5,000 feet, July. (M. C. Z. 250, one
young female). Ollantaytambo, 9,000 feet, July. (M. C. Z. 251,
female).

Previously known from Brazil (Taquara) as well as from Peru (Yura).

264 bulletin: museum of comparative zoology.

TiBELLUs punctulatus (Taczanowski).

Thanatus punctulatus Taczanowski, Horae Soc. ent. Ross., 1872, 9, p. 10.

Locality. — Santa Ana, 3,000 feet, August. (M. C. Z. 253, female).
Previously known from Guiana.

CLUBIONIDAE.

Eusparassus shefteli, sp. nov.
Plate 20, fig. 7-8; Plate 21, fig. 1.

Carapace dark chestnut, blackish over anterior cephalic region
and along the lateral margins. Chelicerae black. Femora chestnut,
distally more blackish; distal joints blackish over a chestnut back-
ground. Sternum dark chestnut or mahogany, the labium and en-
dites similar excepting for the paler distal ends. Hair of these parts
mostly of a golden lustre, that of the carapace in part grey; hair of
the legs long. Abdomen light brown; hair long, subdense, yellowish.

Posterior row of eyes much longer than the anterior (39: 34), moder-
ately procurved; eyes subequal; median eyes near once and a half
their diameter apart, a little farther from the laterals. Posterior
laterals two thirds their diameter from the anterior laterals, larger in
the ratio 10:9. Anterior row of eyes straight, the median eyes a
little larger than the laterals (ratio 11: 10), about their radius apart
and the same distance from the laterals. Median eye-area wider
behind than in front in ratio 8:7; its length equal to anterior width.

Upper margin of chelicera with two teeth; the lower with three
of which the most proximal is the smallest.

Labium semicircular; more than twice as wide as long (42:19);
notches at base slight, (Plate 20, fig. 8).

Claw of palpus distinct; pectinate. Tarsus densely scopulate over
nearly entire surface both above and below. Femur and tibia each
with a long subbasal spine toward the distal end.

Sternum subtriangular; anterior margin straight; narroVly acutely
extended between posterior coxae, (Plate 20, fig. 7).

Femora of first three pairs of legs with three spines on anterior,
three on posterior, and two on dorsal surface; femur IV with spining
the same excepting that the spines of the posterior surface are lacking.

chamberlin: the arachnida. 265

Each patella with a single spine on its posterior side. Tibiae I and
II with 3-3 spines beneath, two on the anterior side, two on the
posterior, and one above; tibiae III and IV the same except for
absence of the dorsal spine. All metatarsi with two spines beneath,
two in front, and two behind. Claws all strongly pectinate, the
teeth increasing in length distad. All tarsi and metatarsi scopulate
beneath.

Female. Length, 17.5 mm. Length of cephalothorax, 7.5 mm.;
width, 7 mm.

fern.

tib. + pat.

met.

tar.

total

Legl

8 mm.

10 mm.

6 . 5 mm.

2 . 2 mm.

26.7 mm

Leg II

8.8

10.9

7.2

2.3

29.2

Leg III

7.2

8.0

5.0

2.0

22.2

Leg IV

7.2

7.9

5.. 5

2.2

22.8

Tibia I, 6.8 mm. Tibia IV, 5.2 mm.

Locality'.— Santa Ana, 3,000 feet, August 3. (Type, M. C. Z. 252,
female).

This species is named in honor of Mr. Herbert Sheftel of New York.

HoRiocTENUS,^ gen. nov.

Cephalothorax ovate, convex; thoracic stria long and distinct,
radial lines not evident.

Posterior row of eyes strongly recurved or appearing as two rows;
eyes subequal, the medians nearer to each other than to the laterals.
Anterior row of eyes recurved ; eyes subequal and nearly equidistant,
the median eyes in type a little the smaller. Area of median eyes
longer than wide, a little narrower in front.

Clypeus narrower than length of area of median eyes, receding from
the anterior eyes ventrocaudad.

Labium wider than long.

Lower margin of furrow of chelicera with one tooth, the upper with
three.

Legs robust. Tibiae of first and second legs armed beneath with
three pairs of spines; the metatarsi with a single pair of long spines
at the proximal end; metatarsi and tarsi densely scopulate. Tarsi of
posterior legs more sparsely scopulate.

Inferior spinnerets (in type species) not contiguous at base. Su-
perior spinnerets with second article short.

' 8pio?, pertaining to the borders or limits, etc. and Cteuus.

266 bulletin: museum of comparative zoology.

Genotype. — Horiodenus lycosoides, sp. nov.

Readily distinguished from Caloctenus and Odo, both occurring in
the Andean region, in having but a single tooth on the lower margin
of the chelicera, as well as in eye relations and characters of legs.

Horioctenus lycosoides/ sp. nov.
Plate 21,%. 2^.

Integument of carapace dark brown with a median longitudinal
yellow or light testaceous stripe from the eye-area to the caudal
margin; this median stripe anteriorly as wide as the eye-area from
where it first widens and then narrows again and remains of nearly
uniform width to the posterior declivity down which it narrows to
the caudal margin; also a narrow paler supramarginal stripe along
each side; carapace clothed densely with chiefly light grey or white
hairs, dark hairs much sparser. Sternum dusky chestnut, the labium
similar excepting at the tip, the endites paler, all these parts clothed
with grey hair like that of the carapace. Legs brown, irregularly but
abundantly, especially the femora, streaked and spotted with black-
ish; hair chiefly grey but with stiff er dark bristles more abundantly
intermixed than on the carapace. Abdomen with the integument
brown, a sagittate outline in some vaguely indicated at the base
above, this followed by a series of equally vague cross-marks; densely
clothed with grey and coarser brown hairs intermixed.

Posterior row of eyes longer than the anterior; very strongly
recurved; median eyes but little larger, separated from the laterals
by about their diameter, only half as far from each other. Anterior
row of eyes decidedly recurved both in dorsal and in anterior view,
the median eyes being borne well forwards; eyes equidistant, less
than their radius apart ; median eyes smaller than the laterals (ratio
of diameters about as 4: 5).

Labium clearly wider than long (23:19); sides convex, decidedly
converging to the roimded or mesally truncate anterior margin;
basal notches short, (Plate 21, fig. 2).

Endites not impressed.

Palpus strongly spined ; claw with five teeth.

Teeth of upper margin of chelicera rather slender, decreasing from

1 Lycosa, fibo%, fonn.

chamberlin: the arachnida. 267

the one nearest claw proximad; single tooth of lower margin also
small, (Plate 21, fig. 4).

Sternum moderately convex; anterior margin widely weakly con-
vex; caudad acutely narrowly pointed, not separating the fourth
coxae; widest near middle; longer than wide in about ratio 5 : 4.

Epigynum proportionately large, (Plate 21, fig. 3).

Female. Length, 9 mm. Length of cephalothorax, 4 mm. ; width,
3.1 mm.

fem. tib.+pat. met. tar. total

Leg I 3.25 mm. 4.2 mm. 2.2 mm. 1.6 mm. 11.25 mm.

Leg II 3.0 4.0 2.0 1.5 10.5

Leg III 3.0 . 3.5 2.0 1.4 9.9

LeglV 3.5 4.8 3.8 2.0 14.1

Tibia I, 2.9 mm. Tibia IV, 3 mm.

Localities.— 0\\2iiit^Yta.mho, 9,000 feet, July. (Type, M. C. Z.
254, one female). Urubamba, 9,500 feet, July!^ (M. C. Z. 255, two
females).

Gayenna monticola, sp. nov.
Plate 22, fig. 6.

Carapace yellow; a dusky median longitudinal line which is gemi-
nate back of the eyes as in Pirata sp.; also a dusky supramarginal
stripe on each side extending forward to the head. Sternum yellow,
dusky at margins. Endites yellow. Labium dusky yellow. Legs
yellow, dark about the bases of spines and in other small spots, giving
the appearance of broken annulations on the proximal joints. Integu-
ment of abdomen greyish brown; above with a narrowly lanceolate
dark median stripe at base, followed by a series of paired dark marks
which extend to a little in front of the spinnerets, the two lines con-
verging caudad; a dusky stripe along each dorsolateral line proxi-
mally below which are scattered inconspicuous dark spots; venter
clear excepting for an interrupted median longitudinal dark line.

Posterior row of eyes moderately procurved; median eyes once and a
half their diameter apart, their diameter from the laterals which are a
little the larger (diameters about as 7:6). Posterior laterals subequal
to anterior laterals from which they are separated by four sevenths
their diameter. Anterior row pf eyes straight, decidedly shorter than
the posterior; laterals larger than the medians (diameters as 7:6);
medians near their radius apart, contiguous with the laterals or nearly

268

bulletin: museum of comparative zoology.

so. Clypeus about four sevenths as wide as diameter of an anterior
lateral eye. Area of median eyes equal in length and breadth or
slightly longer than wide (22:21); narrower in front than behind in
about the ratio 15:21.

Sternum longer than wide in nearly ratio 35:24. Truncate ante-
riorly; widest near middle; angle at caudal end moderate, scarcely
acute. Labium narrowed from basal notches distad, the sides being
weakly convex; distal margin truncate.

Upper margin of furrow of chelicera armed with the usual three
teeth of which the median is much the largest; lower margin with
two subequal teeth.

Tibia I and II with three pairs of spines beneath, of which the
distal are much the smallest, the two proximal pairs being long and
appressed; metatarsi I and' II with a single basal pair of long spines
beneath. Posterior tibiae and metatarsi strongly spined beneath,
laterally and above. Femur I with five spines above, (three in a
transverse row distally, one submedian and one sul)basal). Meta-
tarsi and tarsi I and II scopulate to base; tarsi III and IV more
sparsely scopulate, the corresponding metatarsi not at all.

Female. Length, 8 mm. Length of cephalothorax, 2.7 mm.;
width, 2 mm.

fern.

tib. + pat.

met.

tar.

total

Leg I

2 . 0 mm.

2.3 mm.

1.3 mm.

1 mm.

6.6 mm

Leg 11

2.1

2.3

1.3

1

6.7

Leg III

2

2.1

1.2

1

6.3

Leg IV

2.3

2.9

'1.5

1.1

7.S

Localities — Cuzco, 11,500 feet, July. (Type M. C. Z. 256, female).
Ollantaytambo, 9,000 feet, July. (M. C. Z. 257).

Anyphaena andina, sp. nov.
Plate 22, fig. 4.

Carapace yellow, of a darker shade along the sides; blackish over
eye-area, in a short narrow streak caudad from each posterior median
eye and in a short median line, bifurcate behind, just in front of the
fovea thoracica. Sternum and endites clear yellow, the labium of a
little darker tinge. Legs yellow, indistinctly marked with broken
annuli of which one at distal end of femora and one on patella (espe-
cially of the posterior legs) are most pronounced. Abdomen ventrally

chamberlin: the arachnida.

269

and laterally yellow; the dorsum covered with a close network of
mostly confluent dusky brown spots among which a median longitudi-
nal pale line limited by uneven dark lines may be traced.

Posterior row of eyes only slightly procurved, nearly straight;
medians a little more than their diameter (cir. once and a seventh)
apart, a little less than their diameter (six sevenths) from the laterals.
Anterior row of eyes much shorter than the posterior (41 : 31) ; straight
or scarcely recurved; medians smaller than the laterals (ratio of
diameters 5:7), slightly more than their radius apart (three fifths
diameter) and almost contiguous with the laterals. Clypeus narrower
than diameter of eyes. Area of median eyes equal in length and
breadth; three fourths as wide in front as behind.

Sternum longer than wide nearly in ratio 15:11.

Labium much longer than wide; crenately notched or incised
mesally at distal end.

Lower margin of furrow of chelicera with four or five teeth of which
the two more distal are longer than the others.

Tibia I with three pairs of ventral spines of which the distal are
short and those of the other two pairs long; also bearing on the
anterior side three spines and one dorsally. Metatarsus I below with
a basal pair of long appressed spines; on anterior side with two spines
of which the basal is smaller than the median, and three pairs of
dorsal or subdorsal spines. Femur I with three spines along mid-
dorsal line. Anterior tarsi scopulate; metatarsi sparsely scopulate
distad ; posterior tarsi scarcely truly scopulate or with but few scopu-
lar hairs.

Distal article of superior spinnerets abruptly narrower than the
proximal, cylindroconical, small.

Rima ventralis nearly equidistant between bases of spinnerets and
the genital furrow.

Female. Length, 6.8 mm. Length of cephalothorax, 2.7 mm.;
width, 2 mm.

fem.

tib.+pat.

met.

tar.

total

Leg I

2.1 mm.

3.1 mm.

1 .8 mm.

1.2 mm.

8.2 mm,

Leg II

2.1

2.6

1.7

1.1

7.5

Leg III

2

2.1

1.2

1

6.3

Leg IV

2.2

3.0

Tibia I

2.2
, 2.2 mm.

1

8.4

ioca%.— Tincochaca, 7,000 feet, August. (Type, M. C. Z. 258,
one female).

270 bulletin: museum of comparative zoology.

Anyphaena apora/ sp. nov.
Plate 22, fig. 2-3.

Carapace a brownish or dusky yellow with no distinct markings.
Chelicerae dilute chestnut. Sternum nearly same as carapace.
Endites pale chestnut, the labium darker excepting at pale tip. Legs
dusky yellow above, darker distad, clearer yellow beneath; darker
annuli vaguely indicated on tibiae. x\bdomen dusky grey over a
yellowish background, the venter somewhat paler than the dorsum,
with no distinct markings.

Posterior row of eyes nearly straight or only very slightly procurved;
medians their diameter apart, slightly closer to the laterals. Anterior
row of eyes shorter than the posterior in the ratio 41:55; straight or
very slightly recurved; medians much smaller than the laterals
(diameters as 7:10), about four sevenths their diameter apart and
one seventh their diameter from the laterals; median eyes nearly
their diameter from the lower edge of the clypeus, the laterals but
little more than their radius. Posterior lateral eyes larger than the
anterior medians from which they are separated by only about one
fourth their diameter. Area of median eyes equal in length and
breadth; wider behind than in front in ratio 28: 17.

Sternum longer than wide in ratio 45 : 37.

Labium much longer than wide (40:25). Basal notches long.
Sides only weakly convex and but little converging distad; distal
margin conspicuously concave from side to side, (Plate 22, fig. 2).

Lower margin of furrow of chelicera with four moderately large,
well-spaced teeth.

Tibia I armed with three pairs (or 2, 2, 1) of ventral spines of which
the most distal are some distance proximad of the distal end of the
article and are of smaller size ; two spines on anterior side and two on
the posterior. Metatarsus I with one pair of long appressed, subbasal
spines on ventral surface and with two pairs of subdorsal (each on
dorsolateral line) in position. Femur I above with three spines along
middorsal line, three in line cephalad of these and two in a line caudad.
Anterior tarsi and metatarsi scopulate to base and tibia I also with
some scopular hairs at distal end.

Female. Length, 8 mm. Length of cephalo thorax, 3.6 mm. ; width,
2.8 mm.

1 ajTopos, difBcull.

chamberlin:

THE ARACHNIDA.

fem.

tib. + pat.

met.

tar.

total

Leg I

4 mm.

5.2 mm.

3 mm.

2 mm.

14.2 mm

Leg II

3.8

5

3

1.8

13.6

Leg III

3

3.2

2.5

1.2

9.9

Leg IV

3.2

4.2

3.3

1.3

12.0

Tibia

I, 4 mm.

Tibia IV,

2.9 mm.

271

Locality.— Conservidayo River, August. (Type, M. C. Z. 259).

Anyphaena poicila,^ sp. nov.
Plate 22, fig. 5.

Carapace dusky brown with a clear yellow stripe along each side
above a black marginal line that completely encircles the carapace;
a lighter median longitudinal area on head back of eyes this narrowing
caudad and enclosing a black line over and back of each posterior
median eye; a black line from clypeus over each anterior median eye
and extending back across each posterior lateral eye. Sternum black ;
a narrow interrupted yellow marginal line, a very narrow median
yellow line from anterior margin to middle and on each side a series of
three small yellow dots converging caudad where there is a median
spot. Labium and endites dusky over a yellow background, pale
across tips. Legs yellow; a heavy black annulus opt femora just dis-
tad of middle, one at proximal end of tibia and one near each end of
metatarsus. Tarsus of palpus with a distinct black ring at proximal
end, the tibia also with an interrupted narrower line across proximal
end and small spots across distal; patella and femur also with small
black dots. Abdomen in general yellow; above with a narrowly
deltoid dark brown mark at middle, this continuing forward from its
apex in a narrower median stripe which expands at the base into a
mark of lanceolate form with apex cephalad. Anterior portion of
sides brown, each band followed caudad toward spinnerets by a
number of brown spots, some of which unite in several lines across
dorsum in front of spinnerets. Venter with a median longitudinal
brown stripe which is broadest caudad, is broken across middle, and
extends to the genital furrow; on each side of this with scattered
small dots, some of which are arranged in lines, (Plate 22, fig. 5).

Posterior row of eyes decidedly procurved, a line tangent to anterior

I iroi/ciXos, spotted.

272 bulletin: museum of comparative zoology.

edges of medians intersecting the laterals back of their middles;
medians their diameter apart, five sixths as far from the laterals
which are of the same size or nearly so. Posterior lateral eyes equal
to the anterior laterals from which they are separated by two thirds
their diameter. Anterior row of eyes shorter than the posterior in
ratio 25:33; slightly recurved; medians smaller than laterals, their
diameters being to each other nearly as 2 : 3, their radius apart, nearly
half as far from the laterals. Clypeus a little narrower than diameter
of a median eye. Area of median eyes longer than wide in ratio 7:6;
wider behind than in front in ratio 3:2.

Sternum longer than wide in the ratio 62 : 47.

Labium some more than two thirds as wide as long; apex squarely
truncate, not incised or indented.

Lower margin of furrow of chelicera with four small teeth decreasing
in size proximad as usual.

Tibia I with three pairs of ventral spines of which the two first
pairs are long, those of the basal pair reaching the bases of the median
ones, the distal pair small; two spines on the anterior and two on the
posterior surface. Metatarsus I with the usual pair of long ventral
spines at base; two spines on anterior and two on posterior side.
Scopulae very sparse. Femur I with three spines in middorsal line,
and also one caudad of this line and two cephalad.

Female. Length, 5.1 mm. Length of cephalothorax, 2 mm.;
width, 1.8 mm.

fem. tib.-1-pat. met. tar. total

Leg I

2.1 mm.

3 mm.

1 mm.

1 mm.

7 . 1 mm

Leg II

2

2.8

1.7

1

7.5

Leg III

1.8

1.9

1.1

.9

5.7

Leg IV

2.1

2.2

1.7

1.1

7.1

Tibia

I,

, 1.3 mm.

Tibia

IV

, 21 mm.

ioca%.— Huadquina; 5,000 feet, July. (Type, M. C. Z. 260,
one female not qjuite mature).

Anyphaena sp.

Locality. — Tincochaea, 7,000 feet, August. (M. C. Z. 330, one
immature specimen).

chamberlin: the arachnida. 273

Castaneira quechua, sp. nov.
Plate 22, fig. 1.

Carapace, sternum, and chelicerae black; labium and endites also
nearly black excepting for pale tips. Coxae ventrally yellowish, the
femora black, the succeeding joints more or less dark testaceous,
except the tibia and metatarsus of fourth legs which are nearly black.
Abdomen black, a vague light line across middle above and laterally
but not crossing middle part of the venter. Carapace with numerous
white plumose hairs. Legs with similar hairs but also with numerous
black hairs intermixed. Abdomen with white plumose hairs and
many simple black hairs.

Posterior row of eyes procurbed in such degree that a line tangent
to the anterior edges of the medians cuts the laterals near the begin-
ning of the caudal third ; medians nearly once and a third their diam-
eter apart, less than their diameter from the laterals which are of the
same size or nearly so. Area of median eyes equal in length and
breadth or very nearly so, scarcely narrower anteriorly than pos-
teriorly. Anterior row of eyes procurved; median eyee their radius
apart, half as far from the laterals; medians with diameter exceeding
that of laterals in ratio 9:7; the long diameter of the laterals exceeding
the lesser diameter nearly in ratio 7:5. Clypeus a little wider than the
diameter of a median eye (about as 11:9).

Sternum three fourths as wid6 as long; borders depressed as usual;
anterior margin straight, sides converging to meet at an angle caudad
but the angle not acute.

Labium and chelicerae typical.

Tibia of leg I armed beneath with three pairs of spines, metatarsus
with two pairs. Tibia II with but two pairs of spines beneath, the
metatarsus also with two pairs. Tarsi and metatarsi I and II scopu-
late to base; metatarsi III and IV scopulate only distally.

Female. Length, 8 mm. Length of cephalothorax, 3.25; width,
2 mm.

fem.

tib.+pat.

met.

tar.

total

Leg I

2 . 6 mm.

3 mm.

2 mm.

1.2 mm.

8.8 mm.

Leg II

2.2

2.8

1.9

1.1

8.0

Leg III

2.2

2.4

2.0

1.1

7.7

Leg IV

2.6

3.5

3.1

1.3

10.3

Tibia

I, 2.1 mm.

Tibia IV, 2 mm.

274 bulletin: museum of comparative zoology.

Locality. — Conservidayo River, August. (Type M. C. Z. 261,
one female).

QuECHUELLA, gen. nov.

Cephalo thorax ovate with the frons moderately broad; stria
thoracica fine but distinct, of moderate length.

Posterior eyes subequal (in type the lateral a little the larger),
nearly equidistant, in a weakly proeurved row. Lateral eyes on each
side separated by less than their diameter. Anterior eyes in a re-
curved row; median eyes much smaller than the laterals, separated
from each other but nearly contiguous with the laterals.

Clypeus much narrower than the anterior median eyes.

Lower margin of furrow of chelicera armed with four small teeth.

Labium extending beyond middle of endites; longer than wide
though not greatly so; distally truncate.

Anterior tibiae (compressed in type) without spines.

Furrow of posterior spiracle well in front of middle.

Distal article of superior spinnerets short, conical.

Genotype. — Quechuella lampra, sp. nov.

Distinguished from xVnyphaena and related genera similarly having
the furrow of the posterior spiracle in front of middle in wholly lack-
ing of spines on the anterior tibiae and by its very narrow clypeus.

QuECHUELLA LAMPRA,^ sp. noV.

Plate 21, fig. 5-8.

Carapace pale yellowish brown, dusky over sides and eye-region.
Sternum, endites, and labium yellow. Coxae of legs beneath and
proximal portions of femora yellow; legs more distad, especially
the tibiae and metatarsi, dusky to nearly black, the tarsi again paler.
Abdomen ventrally and over lower portion of sides a dilute yellowish
white; dorsum a very dilute brownish, marked at the base in the
median line with a darker narrow stripe which is followed caudad by a
series of paired dark dots extending to the spinnerets and on each side
of dorsum and upper part of sides also darker, (Plate 21, fig. 7).

Posterior row of eyes very slightly proeurved; medians slightly
smaller than the laterals (ad. 5:6), a little less than their diameter

1 Xafiwpos, distinct.

chamberlin: the arachnida. 275

(four fifths) apart, their diameter from the laterals. Lateral eyes on
each side equal, their radius apart. Anterior row of eyes recurved;
medians smaller than the laterals (diameters as 4.5: 6), one third their
diameter apart, closer to the laterals. Clypeus very narrow, about
one third the diameter of an anterior median eye wide. Area of
median ej'es nearly equal in length and breadth; wider behind than
in front in ratio 2:3, (Plate 21, fig. 8).

Sternimi widest back of middle, much narrowed cephalad; more
abruptly narrowed caudad, the caudal angle not very acute; wider
than long in ratio 15:13, (Plate 21, fig. 6).

Anterior tibiae flattened dorsoventrally, apparently wholly un-
spined; anterior metatarsi with a pair of very short spines beneath.
Posterior tibiae and metatarsi with spines as usual.

Furrow of posterior spiracle situated at about one third the distance
from the genital furrow to the spinnerets.

Length of female, 4.1 mm. Length of cephalo thorax, 1.7 mm.;
width, 1.2 mm.

Locality.— San Miguel, 6,000 feet, July. (Type, M. C. Z. 262,
one female not fully mature).

Trachelopachys bicolor, sp. nov.
Plate 21, fig. 9-10.

Cephalothorax with chelicerae, labium and endites, and the entire
abdomen solid black or the abdomen above with very vague chevron-
lines caudad. Spinnerets yellowish. Legs clear yellow, the scopulae
giving them a dusky appearance at distal ends.

Posterior row of eyes recurved in such degree that a line tangent
to caudal edges of medians cuts through the anterior fourth of the
laterals; median eyes a little smaller than the laterals, twice their
diameter apart, farther from the laterals. Anterior row of eyes a
little procurved; laterals much larger than the medians (diameters
about as 3:2); median eyes their diameter apart, about half as
far from the laterals. Clypeus twice as wide as diameter of median
eye. Area of median eyes equal in length to width behind; wider
behind than in front in ratio 7 : 6.

Sternum longer than wide in a little less than the ratio 4:3; strongly
convex, the borders being much above (dorsad of) level of median
portion.

276 bulletin: museum of comparative zoology.

Labium strongly thickened across base, the thickened portion sepa-
rated by a distinct transverse furrow from the distal portion; sides
moderately converging distad, substraight proximad, convex toward
tip; apically truncate; wider than long in ratio 6:5, (Plate 21, fig. 9).

Upper margin of fmTow of chelicera with the usual three stout
teeth of which the median is largest; lower margin with two stout
and subequal teeth.

Palpus of female unspined; claw smooth.

Tarsi and metatarsi of legs I and II densely scopulate and with
some scopular hairs on distal end of tibia; tarsi and distal portion
of metatarsi also scopulate in legs III and IV. All legs without spines.

Female. Length 8 mm. Length of cephalothorax 4 mm,; width,
3.2 mm.

fem. tib.+pat. met. tar. total

Leg I 3 mm. 3.7 mm. 2 mm. 1.3 mm. 10.0 mm.

Legll 2.8 3.3 1.9 1.2 9.2

Leg III 2.2 2.7 1.8 1 7.7

Leg IV 3.2 3.8 2.7 1.3 11.0

Tibia I, 2.2 mm. Tibia IV, 2.4 mm.

Localities.— Vrnho^rnhdi, 9,500 feet, July. (Type, M. C. Z. 263,
female; paratype, no. 264, one female). Ollantaytambo, 9,000 feet,
July. (M. C. Z. 265, one female).

PISAURIDAE.

Trechalea sp.

Three immature specimens of an uncertain species were secured
at Huadquina, 5,000 feet, July. (M. C. Z. 266).

Trechalea monticola, sp. nov.
Plate 23, fig. 1.

Carapace brown, darker, more black, along the lateral borders;
on each side a pale supramarginal stripe uniting with the one of
opposite side across the clypeus, these stripes more or less zig-zag;
a median longitudinal pale band narrow from caudal margin to the
pars cephalica upon which it expands to nearly the width of the eye-
area and then extends as a narrow tongue between the eyes, this broad
portion embracing a longitudinal dark line back of each posterior

chamberlin: the arachnida.

277

median eye and a pair of broader stripes between this and the one of
opposite side. Sternum yellow, a pair of large dark spots on the mid-
dle portion and a dusky line within each lateral border. Labium
dusky, pale across tip. Endites yellow. Chelicerae yellowish, some-
what darker proximally. Legs yellowish brown or testaceous, the
femora and coxae beneath clear yellow or whitish yellow, the femora
darker distad; over the anterodorsal surface the femora are marked
with a series of four large dark spots. Abdomen with venter dusky
yellowish; sides and dorsum greyish black, a lanceolate outline at
base above in black and a series of about four white spots on each side.

Face and clypeus much sloping as usual.

Anterior row of eyes slightly recurved ; shorter than the row formed
by the two posterior median eyes (28:31); median eyes much larger
than the laterals, the diameter being nearly twice as great (15:8);
median eyes not fully their radius apart (five sevenths) and scarcely
more than half as far from the laterals; medians about once and a
third their diameter from lower margin of clypeus. Posterior median
eyes with diameter one and a third times that of the anterior medians,
their diameter apart and less than their radius from the anterior
medians. Area of posterior eyes wider behind than in front in ratio
115: 62, nearly 2.4 times wider than long.

Lower margin of furrow of chelicera with three stout conical teeth.

Labium longer than wide in ratio 7:6; basal notch one fourth the
total length; distal margin widely but very weakly convex, sub trun-
cate; sides convex and moderately converging distad.

Total length of sternum including caudal process exceeding the width
nearly as 10:9.

Tarsi of legs all slender and distinctly curved or bent, their diameter
least near middle of length. Paired claws of leg I with five slender and
moderately divergent teeth on the proximal half; unpaired claw small,
abruptly bent, with one slender spine or tooth, (Plate 23, fig. 1).

Female. Length, 9 mm. Length of cephalothorax, 4.3 mm.;
width, 4.7 mm.

fern. tih. + pat.

met. tar.

total

Leg I

5.6 mm. 7. 1 mm.

5.25 mm. 3 mm.

20.95 mm

Leg II

7 8.7

6.6 3

25.3

Leg III

5.8 6.5

5.1 3

20.4

Leg IV

6.3 8

7.25 4

25.55

Tibia I, 5 mm.

Tibia IV, 6 mm.

Locality — Santa Ana, 3,000 feet, August. (Type, M. C. Z. 267,
one female, not quite adult).

278 bulletin: museum of comparative zoology.

TuNABo/ gen. nov.

Cephalothorax longer than broad, moderately convex, the posterior
declivity abrupt; thoracic stria long.

The anterior row of eyes weakly procurved; median eyes farther
from each other than from the laterals, a little larger than the laterals.
Posterior eyes in two distinct rows as in Lycosa, much larger than
those of the anterior row. Area of median eyes decidedly wider than
long and much wider behind than in front. Clypeus much narrower
than area of median eyes, only a little exceeding the diameter of an
anterior median eye; sub vertical.

Labium wider than long.

Lower margin of furrow of chelicera with four teeth.

All tarsi, and the anterior metatarsi at least in part, scopulate.
Anterior tibiae armed beneath with five pairs of spines, the distal
pair small and the others very long; the metatarsi armed beneath
with three pairs of spines.

Genotype. — Tmiabo peruvianus sp. nov.

Easily distinguished from Trechalea, the preceding genus, in having
the anterior eyes in a procurved row, and from Hygropoda, also
occurring in the Andean region and a closely related genus, in the
relatively much larger posterior eyes, in the much narrower clypeus,
and in having leg IV longer than leg I.

TUNABO PERUVIANUS, sp. nOV.

Plate 22, fig. 7-9.

A sharply defined median longitudinal stripe over entire length of
carapace and abdomen, this limited on each side on the carapace by a
black stripe below which is a testaceous stripe of about the same
width, this enclosifig some dark dots, the margins darker; on the
abdomen the median stripe is limited on each side by a dark brown
stripe, the sides and venter of abdomen pale, brownish grey with
numerous small dark dots. Sternum yellow or dilute light brown
minutely spotted with black and with a median longitudinal black
line on caudal half. Labium and endites yellow or testaceous.
Legs brown, the coxae and the femora beneath paler; femora marked
above with longitudinal blackish lines and the legs elsewhere minutely
spotted and streaked with dark. Chelicera each with a black longi-
tudinal stripe down its front face.

' Gosiute iuna, straight, and nabo, a mark.

C'HAMBERLIN: the ARkcHNIDA. 279

Anterior row of eyes shorter than the second (about as 11:12);
sHghtly procurved; diameter of medians greater than that of laterals
in ratio 4:3; medians three fourths their diameter (or a little farther)
apart, their radius or slightly more from the laterals; median eyes
more than their diameter from the lower edge of the clypeus (about
once and a fourth), the laterals nearly once and two thirds their diam-
eter from it. Diameter of eyes of second row exceeding that of the
anterior medians in about ratio 5:2; eyes of second row more than
their radius apart (about three fifths diameter), less than their radius
from anterior medians (two fifths diameter). Posterior eyes a
little smaller than the second, three and a half or a little more times
their diameter apart. Quadrangle of posterior eyes contained in
total length of cephalothorax about 3.75 times. Area of median eyes
much wider than long (48: 34), wider behind than in front (12:5).

Total length of sternum exceeding the width in ratio 8.5-9:7,
(Plate 22, fig. 8).

Labium wider than long, (Plate 22, fig. 7).

Upper margin of furrow of chelicera with the usual three teeth;
lower margin with four teeth of which the third from the distal end
is much the smallest, the other three stout, conical, equal, (Plate 22,

fig. 9). , . .

Anterior tarsi and metatarsi scopulate to base ; the posterior meta-
tarsi scopulate lightly on distal portion only. Tibia I with five
pairs of ventral spines, these being very long, the first or basal ones
overlapping bases of those of the third pair, the distal spines much
shorter than those of first and second pairs; also a short spine on
anterior and one on posterior face. Metatarsus I with three pairs
of ventral spines of which the first two pairs are very long like those
of the tibiae ; also with three spines on anterior and three on posterior
surface and a pair of dorsal ones near distal end.

Anterior piece of lorum of pedicel notched behind, the posterior
piece rounded anteriorly and fitting into the notch.

Process of tibia of palpus in male ventral in position; subcorneal
and low in specimen lacking one moult of maturity.

Male. Length, 9 mm. Length of cephalothorax, 3 mm.; width,
2,5 mm.

fem. tib.+pat. met. tar. total

Leg I 3.1mm. 4 mm. 2.1mm. 1.3 mm. 10.5 mm.

Legll 3.1 3.8 2.1 1.3 10.3

Leg III 2.3 3 2 1.1 8.4

Leg IV 3.7 4 3 1.3 110

280 bulletin: museum of comparative zoology.

Locality.— Ruadquina, 5,000 feet, July. (Type, M. C. Z., 322,
male lacking apparently one moult of maturity; para type, M. C. Z.,
268, one male in same stage as type).

LYCOSIDAE.

PORRIMA HARKNESSI, sp. nov.

Plate 23, fig. 2-6.

Carapace with integument from light brown to darker, nearly
chocolate-brown; the lateral margins black with a pale supramarginal,
stripe on each side; eye-region blackish; a pale median longitudinal
line extending from the eye-area caudad and on each side of this a
pale or whitish line converging toward the corresponding one of the
opposite side with which it unites caudad of the stria. Sternum dark
brown to blackish, paler about margin and with a pale median longi-
tudinal mark in the anterior portion. Labium dusky, paler across
the distal end. Endites lighter brown. Chelicerae dark brown to
somewhat mahogany color. Legs dilute testaceous to dark brown or
dusky brown. Abdomen above almost black, the ^ blackish area lim-
ited on each side by a clear white line from which, beginning near
middle, a series of very short lines are given off on the inner side and
extend a little cephalad of mesad ; sides and \'enter paler, from dusky
testaceous to nearly black; the venter showing a vague pale longitudi-
nal line on each side. The light lines of carapace and abdomen
clothed densely with white hairs, the margin of carapace also clothed
with white hair, the hair of other parts dark.

Carapace with dorsal line in profile nearly horizontal, a little de-
pressed at the groove; pars cephalica anteriorly very narrow.

Anterior row of eyes strongly procurved, much longer than the
second row (posterior medians) but shorter than the third (posterior
laterals); median eyes much smaller than the laterals (diameters
nearly as 2:3), about their radius apart and about half as far from the
laterals; lateral eyes not fully their radius from the lower edge of
clypeus. Eyes of the second row but slightly larger than the anterior
lateral eyes, about their radius apart. Eyes of the third row clearly
smaller than those of the second, their diameters being as 4:5; each
its diameter from corresponding eye of the second row, a little more
than three times their diameter apart. Cephalothorax between six

chamberlin: the arachnida. 281

and a half and seven times as long as area of the two posterior rows
of eyes.

Labium a little longer than wide (28:25), reaching a little distad of
middle of endites; proximal notches long; distal margin wide, a
little incurved, (Plate 23, fig. 2).

Sternum but little longer than wide, convexly rounded in front;
caudally ending in a short acute process between the last coxae.

Chelicerae long and cylindrical, not very stout; lower margin with
three subequal teeth; upper margin with three teeth of which the
median is largest as usual.

Tibiae I and II with three pairs of ventral spines of which the fii'st
two are very long and appressed, the distal pair much shorter; two
spines on anterior and two on posterior surface. Femur I with three
spines in middorsal line and with two on the anterior side of this and
three on the posterior. Patellae I and II unarmed; III and IV with a
spine on anterior and one also on posterior surface. Paired claws
with mostly eleven or twelve long teeth, these a little curved distad;
unpaired claw with two short teeth at base, (Plate 23, fig. 6).

Male (Type). Length, 15 mm. Length of cephalo thorax, 7 mm.;
width, 5 mm.

fern. tib.+pat. met. tar. total

Leg I 7 mm. 8.5 mm. 5.7 mm. 3.5 mm. 24.7 mm.

Leg II 6.9 8 5.6 3.5 24.0

Leg III 6.7 7 5.5 3 22.4

Leg IV 7.5 9 8.3 4 288

Tibia I, 5.25 mm. Tibia IV, 6.8 mm.

Female. Length, 13.3 mm. Length of cephalothorax, 6 mm.;
width, 4.6 mm.

fem. tib +pat. met. tar. total

Leg I 5.9 mm. 5.9 mm. 4.2 mm^ 3 mm. 19.0 mm.

Legll 5.1 6.1 4.1 2.8 18.1

Leg III 5 5.3 4 2.2 16.5

Leg IV 6.2 7 6.6 3.1 22.9

Tibia I, 4.9 mm. Tibia IV, 5.2 mm.

ioca%.— Huadquina, 5,000 feet, July 26. (Type, M. C. Z.
269, male; paratypes, M. C. Z. 270, one male and one female).

Named for Edward S. Harkness of New York, a patron of the
expedition.

282 bulletin: museum of comparative zoology.

Lycosa securifer Tullgren.

Arkiv. f. zool, 1905, 2, p. 66, pi. 8, f. 32.

The specimens here listed are, with sUght doubt, referable to L-
securifer described by Tullgren from Argentina.

Localities. — Cuzco, 11,500 feet, July. (M. C. Z. no. 271, one adult
female with egg sac and one immature female taken under stone).
Urubamba, 9,500 feet, July. (M. C. Z. 272).

Lycosa gumia Petrunkevitch.

Lycosa gulosa Tullgren, Arkiv. f . zool., 1905, 2, p. 63, pi. 8, f . 30.

Lycosa gumia Petrunkevitch, Bull. Amer. mus. nat. hist., 1911, 29, p. 560.

Localities. — Tincochaca, 7,000 feet, August. (M. C. Z. 273, six
females). San Miguel, 6,000 feet, July. (M. C. Z. 274, one female).

Lycosa thorelli (Keyserling).

Tarentula thorelli KeyserUng, Verh. Zool. hot. ges. Wien, 1876, 26, p. 650, pi. 1,
f. 28.

Localities. — Huadquina, 5,000 feet, July. (M. C. Z. 275, one
female). Lucma, 7,000 feet, August. (M. C. Z. 276, one female
adult, and one immature female).

Lycosa sp. a.

An immature male of doubtful species collected at Paltaybamba,
5,000 feet, August 27. (M. C. Z. 277).

Lycosa sp. b.

An immature male from Panama, June. (M. C. Z. 278). It is
near L. dilatata F. Cambridge, but is probably undescribed.

Lycosa sp. c.

An immature female from Ollantavtambo, 9,000 feet, July.
(M. C. Z. 279).

chamberlin: the arachnida. 283

Lycosa sp. d.

An immature female of uncertain species from Santa Ana, 3,000
feet, August. (M. C. Z. 280).

Lycosa sp. e.

An immature female from the Conservidayo River collected in
August. (M. C. Z. 281).

Lycosa liopus,^ sp. no^■.
Plate 24, fig. 4.

Carapace with sides - brown, a supramarginal stripe on each side
and a broad median longitudinal stripe light testaceous, the latter
stripe narrowed between eyes and again gradually narrowing caudad,
not distinctly constricted in front of the groove; lateral margins
dusky. Hair of the light stripes when dry grey or somewhat brownish
grey. Sternum and legs light testaceous, the femora of legs beneath
paler, more yellow ; legs without any markings. Labium dusky except
across tip. Endites light chestnut, tips pale. Chelicerae chestnut.
\^enter of abdomen yellow, dorsum and sides darker, especially across
anterolateral corners; dorsum with a median wedge-shaped black
mark at base, the mark abruptly narrowed into a narrow tongue
caudad.

Anterior row of eyes shorter than the second; clearly procurved
median eyes with diameter larger than that of laterals in ratio 16: 13,
their radius apart and only half as far from the laterals. Anterior
median and anterior lateral eye each its diameter from the lower
margin of clypeus. Eyes of second row greatly exceeding those of
the first, their diameter being to that of the anterior medians as 2:1;
not fully two thirds their diameter apart (9/16) scarcely one third
their diameter from anterior median eyes. Eyes of third row about
seven eightlis the diameter of the second, between 2.25 and 2.5 times
their diameter apart. Length of cephalothorax to length of area of
four posterior eyes as 4.75: 1.

Lower margin of furrow of chelicerae with the usual three teeth,
these stout, conical, and subequal.

• Xeios, smooth, tovs. foot.

284 bulletin: museum of comparative zoology.

Sternum 2.8 mm. wide and 3.8 mm. long.

i\nterior tarsi and metatarsi seopulate to the base, the tibiae also
scopulate at distal end; posterior metatarsi only sparsely seopulate
distad. Tibia I with ventral spines as usual, moderate in size, the
basal spines largest, a little overlapping those of the second pair; on
anterior side with two (or on one side with three) spines the posterior
side with none; spines of tibia II the same excepting there are two
anterior spines on both right and left legs in type and there is on one
side a minute distal spine on the caudal side.

Female. Total length, 15.5 mm. Length of cephalothorax, 8.1
mm.; width, 6 mm.

fem. tib.+pat. met. tar. total

Leg I 5.2 mm. 6.4 mm. 4 mm. 2.8 mm. 18.4 mm.

Leg II 5.1 6.4 3.9 2.8 18.2

Leg III 5.1 5.7 4.2 2.5 17.5

Leg IV 6.3 7.1 6.1 3.25 22.75

Tibia I, 4 mm. Tibia IV, 4.7 mm.

Locality.— Santa Ana, 3,000 feet, August. (Type, M. C. Z. 282,
one female).

Lycosa orinus,^ sp. nov.

Plate 24, fig. 1.

Carapace with integument dark brown, a supramarginal pale stripe
on each side and a broad median longitudinal testaceous stripe ex-
tending from between eyes to the caudal margin, this stripe constricted
a little in front of the groove and narrowing down the posterior
declivity ; the pale stripes, and the border beneath the lateral ones as
well, clothed with bronze-brown hair intermixed, especially in the
eye-region, with black bristles ; hair of other parts of carapace darker.
Sternum black. Labium and endites deep chestnut or blackish, pale
across the tip. Chelicerae black clothed in front below with orange
hair. Legs testaceous to brown, the coxae beneath all black; femora
I and II also black beneath but femora III and IV merely with a
black stripe across distal end beneath; all patellae black beneath; all
tibiae with a broad black band across distal end; the anterior meta-
tarsi black beneath, the posterior metatarsi black at the distal ends,

' 'optivbi, mountaineer.

chamberlin: the arachnida. 285

the hair of black areas of legs black, of the other parts above brown
with darker longer bristles intermixed in the usual way, but the light
areas beneath clothed with more grey colored hair. Abdomen be-
neath solid black; sides grey-brown; dorsum very dark brown, dusky,
a median blackish wedge-shaped mark outlined by light lines in the
front third of the length, this followed by a series of black chevron-
lines each bordered caudad by a light line which commonly termi-
nates at each end in a light spot, the light lines clothed with hair
which is yellow when dry, there being other spots and streaks of same
as well.

Cephalothorax in profile with the dorsal line rising gradually and
moderately from the posterior declivity to the third eyes.

Anterior row of eyes a little, shorter than the second (17:18), pro-
curved as usual; medians larger than the laterals (diameters as
19:16) scarcely their radius apart, nearer to the laterals; median
eye separated from eye of second row of same size by a distance
equal to diameter of lateral eye; anterior lateral eye once and a
fourth their diameter from lower edge of clypeus. Eyes of second
row with diameter fully twice that of anterior medians, about five
eighths their diameter apart. Posterior eyes with diameter three
fourths that of the second eyes, not quite three times their diameter
apart. Area of four posterior eyes just one fifth total length of
carapace.

Sternum broadly elliptic, three fourths as wide as long.

Anterior tarsi and metatarsi and distal end of tibiae densely scopu-
late; posterior metatarsi scopulate only at distal end. Anterior tibiae
with the usual three pairs of spines beneath, these very short as are
also the small spines on the anterior and posterior surface.

Female. Length, 18.5 mm. Length of cephalothorax, 9.4 mm.;
width, 7 mm.

fern. tib. + pat. met. tar. total

Leg I 7.1mm. 8.1 mm. 5 mm. .3.5 mm. 23.7 mm.

Leg II 6.5 7.3 5 3 21.8

Leg III 6 6.8 4.6 3 20.4

Leg IV 7.8 8.8 7.0 3.7 24.3

Locality.— Cnzco, 11,500 feet, July 6, 7. (Type, M. C. Z. 283;
paratypes M. C. Z. 284, two mature and one immature females and
three immature males).

286 bulletin: museum of comparative zoology.

Lycosa andina, sp. nov.
Plate 24, fig. 2-3.

Carapace dark brown, a broad median longitudinal stripe from
eyes caudad and narrowing to caudal edge testaceous, this stripe
limited as far back as the thoracic furrow by a narrow black line; a
narrow pale supramarginal stripe on each side. Sternum, labium,
endites, and coxae of legs beneath deep blackish brown in alcohol or
dense blackish when dry. Chelicerae black; a dense coat of orange
colored hair on anterior face, especially distally. Legs brown, femora
paler beneath; patella black across ventral surface; tibiae with a
broad black band across distal end on ventral half. Abdomen grey-
brown, venter wholly dusky; dorsally with a median longitudinal
black stripe at base this first clavately widening and then abruptly
narrowed to continue as a narrow tongue to middle wliere it bifurcates
in two narrow pointed branches; caudad of this mark two or three
slender black chevron-marks are usually distinctly indicated. The
male is colored similarly to the female but the venter back of the
genital furrow is solid black and the abdomen in a stripe across each
anterolateral corner and caudad along side of abdomen is also black.

Cephalothorax in profile with the dorsal line convex, very little
rising cephalad of the groove.

Anterior row of eyes procurved, equal in length to the second;
median eyes decidedly larger than the laterals (diameters as 4:3),
rather less than their radius apart, closer to the laterals; anterior
medians about their diameter from lower edge of clypeus. Eyes of
second row with diameter larger than that of anterior median eye in
about ratio 18:11, nearly two thirds their diameter apart. Eyes of
third row smaller than those of the second (diameters nearly as 5:6);
about five sixths their diameter from eyes of second row, more than
twice their diameter apart. Cephalothorax between 7.25 and 7.5
times longer than the quadrangle of the second and third eye rows.

The three teeth of the lower margin of furrow of chelicera long,
conical, stout, the most proximal one a little largest.

Distal edge of labium broad, very slightly incurved from end to end.

Sternum semicircular, longer than wide in about ratio 3.1:2.9.

All tarsi and metatarsi scopulate to base; anterior tibiae also
densely scopulate over distal half or more. Anterior tibiae with the
usual ventral spines only; posterior tibiae in addition to the ventral

chamberlin: the arachnida.

287

spines with two spines in the middorsal line and with two each on the
anterior and posterior side above.

Male. (Type, Tincochaca). Length, 24.5 mm. Length of cepha-
lothorax, 9 mm.; width, 6.2 mm.

fern.

tib.+pat.

met.

tar.

total

Leg I

6.8 mm.

8.3 mm.

6 mm.

3.5 mm.

24.6 m

Leg II

6

7.3

4.2

3.1

20.0

Leg III

5.4

6

4.2

3

14.6

Leg IV

7.3

8.8

7.2

4.0

27.3

Tibia I, 5.2 mm.

Tibia IV

, 5.6 mm.

Female {Tor ontoy). Length, 24.5 mm. Length of cephalothorax,
9 mm.; width, 6.2 mm.

fem.

tib.+pat.

met.

tar.

total

Leg I

6.5 mm.

8 mm.

5 mm.

3.2 mm.

22.7 mm

Leg II

6

7.3

4.2

3.1

20.7

Leg III

5.4

6

4.2

3

18.6

Leg IV

7.3

8.8

7.2

4

27.3

Tibia

I, 5 mm.

Tibia IV

, 7 mm.

Localities. — Tincochaca, 7,000 feet, August. (Type, M. C. Z.
285, male; paratype, M. C. Z. 286, female). Lucma, 7,000 feet,
August. (M. C. Z. 287, six specimens). Torontoy, 8,000 feet, July.
(M. C. Z. 288, one female). Urubamba, 9,500 feet, July. (M. C. Z.
289, one immature female).

Lycosa algina,^ sp. nov.

Plate 24, fig.

5-6.

Sides of carapace blackish brown; the usual pale supramarginal
line on each side extending forward to pars cephalica; a median longi-
tudinal pale stripe extending from between eyes which are enclosed
in black, to caudal margin, this being a little constricted at a point
between eyes and groove, narrowing down posterior declivity as usual.
Sternum yellow with a solid black median longitudinal stripe across
the entire length. Chelicerae testaceous with a few, in part obscure,
blackish longitudinal lines. Coxae of legs beneath yellow. Leg I

i 'oiKynvoi, giving peiin.

288 bulletin: museum of comparative zoology.

partly regenerated and not normal nearly clear yellow, markings
obscure. Other legs with femora marked by four wavy or serrate
edged black annuli; patellae nearly entirely' l)lackish; tibiae with
two broad black bands leaving between them only a narrow circle of
yellow at the middle; metatarsi with three broad black annuli; tarsi
without dark bands.

Abdomen beneath yellow, black at base of spinnerets; sides above
finely densely spotted with black, the spots fewer below, a black patch
on each anterolateral corner; dorsum pale in a band about a black
basal mark and also paler transversely at caudal end and in a narrow
median line connecting with the anterior light area.

Carapace moderately high, the posterior declivity steep; dorsal
line in profile a little depressed at dorsal groove.

Anterior row of eyes distinctly shorter than the second (9:10);
moderately procurved; median eyes clearly less than their radius
apart (four elevenths of diameter), closer to the laterals, their diam-
eter exceeding that of the laterals in about the ratio 11:9. Anterior
median eyes separated from the lower edge of clypeus by less than their
diameter (about 8:11). Eyes of second row not quite twice the diam-
eter of the anterior median eyes, more than their radius apart (about
thirteen twentieths of diameter); separated from anterior median
eyes by about the radius of the latter. Area of median eyes wider
than long in ratio 10:7; wider behind than in front in ratio 2:1.
Eyes of third row distinctly smaller than the second (diameters about
as 9:11), near 2.22 their diameter apart. Cephalothorax only 3.96
times longer than area of posterior eyes.

Spines of anterior tibiae and metatarsi as usual in number; the first
two pairs very long, the distal ones much shorter.

Labium a little wider than long (15:14); notches about one third
the total length, (Plate 24, fig. 5).

Male. Length, 7 mm. Length of cephalothorax, 4 mm.; width,
3 mm.

fem. tib. + pat. met. tar. total

Leg I - - - - -

Leg II

3.2 mm.

4 mm. 2.6 mm.

1.4 mm.

11.2 mm

Leg III

3

3.5 3

1.6

11.1

Leg IV

4

4.8 4.8
Tibia IV, 3.1 mm.

2.2

15.8

Locality. — Paltaybamba, 5,000 feet, August. (Type, M. C. Z.
290).

chamberlin: the arachnida.

289

A species in several respects much like a Pardosa rather than a
Lycosa. The palpal organ of the male has the structure typical of a
Lycosa.

Arctosa altamontis, sp. nov.
Plate 23, fig. 7-9.

Carapace with the usual wide median pale spot back of the eyes
with radiating lines down the sides, this area abruptly narrowed at
caudal end of thoracic groove and running as a much narrower tongue
down the posterior declivity; on each side above the black margin a
series of large, in part confluent light spots, the row bending up to
the third eye on each side. Eye-area black excepting for short tongue
of light color between the third eyes. The light areas clothed with
grey hair. Sternum black. Labium black except distally. Endites
testaceous. Chelicerae brown or light testaceous. Legs yellow,
closely annulate with black, the annuli commonly incomplete below
and often broken into spots ; femora with four dark rings of which the
most distal is proximad of the end; papilla with one ring at proximal
end; tibia with two rings and metatarsus with three. Abdomen with
venter and lower part of sides clear greyish yellow; a dusky band over
each anterolateral corner broken into irregular spots farther caudad;
at base with a median longitudinal lanceolate mark of brown with two
small triangular black marks on the edge of each side; basal mark
followed behind by a series of pairs of dark, often indistinct, marks
which may in part be confluent.

Cephalothorax low. Head rather broad with sides of face sloping
outward as usual; the eyes removed from lateral edges of the head.

Anterior row of eyes clearly shorter than the second (38:45); a
little procurved; median eyes decidedly larger than the laterals (ratio
of diameters as 10:7), three fifths their diameter apart, not fully a
third as far from the laterals. Clypeus narrower than the anterior
median eyes (about three fifths as wide) . Eyes of the second row one
and three fifths the diameter of an anterior median eye ; their diameter
or very nearly so, apart. Third eyes each slightly smaller than the
second (about as 15:16 in diameters), just twice their diameter apart.
Quadrangle of posterior eyes much wider behind than in front (57:45),
wider than long (57 : 48) ; length to that of cephalothorax nearly as 1 : 4.

Upper margin of furrow of chelicera with the three teeth as usual;
lower margin with three stout conical teeth which commonly de-
crease in size proximad.

290 bulletin: museum of comparative zoology.

Labium a little longer than wide (17:16); notches one fourth the
total length; sides convex, converging distad; distal margin widely
weakly incurved.

Scopulae of tarsi as usual, those of the anterior tarsi entire though
not dense, those of the posterior pairs more sparse and divided by a
wide setose band. Tibia I with the usual three pairs of ventral spines;
tibia II with the usual distal pair of spines but with single median
and basal spines toward caudal side, in place of the usual pairs. No
median dorsal spine at base of posterior tibiae, this as usual represented
merely by a long bristle in each case. Patellae not armed in median
dorsal line.

Female. Length, 10 mm. Length of cephalothorax, 4 mm.; width,
3.1 mm.

fern. tib.+pat. met. tar. total

Leg I

3 mm.

3.6 mm.

2.1 mm.

1.8 mm.

10.5 mm

Leg II

3

3.1

2.1

1.7

9.9

Leg III

2.7

3.1

2.3

1.7

9.8

Leg IV

4

4.3

3

2.1

13.4

Tibia I, 2 mm.

Length of male, 8 mm.; length of cephalothorax, 3.8 mm., width,
2.7 mm.

Localities. — Cuzco, 11,500 feet, July. (Type, M. C. Z. 291, female;
para types, M. C. Z. 292, numerous females). Arequipa, 7,600 feet,
June 28. (M. C. Z. 323, one female). Urubamba, 9,500 feet, July.
(M. C. Z. 293, on male, two females).

Orinocosa,^ gen. nov.

Cephalothorax much as in Lycosa, but with sides of head steeper.

Anterior eyes small, the medians larger than the laterals to which
they are closer than to each other; anterior row much shorter than
the second, strongly procurved. Eyes of second row large, less than
their diameter apart. Quadrangle of posterior eyes wider behind
than in front (in type species the length of quadrangle is to length of
cephalothorax as 1:3.6 or 3.7).

Clypeus receding from median eyes; narrow, but equal in width
to diameter of an anterior median eye.

' Greek 'optivos, mountaineer, Lycosa.

chamberlin: the arachnida. 291

Upper margin of furrow of chelicera with three teeth of which the
median is longest; lower with three teeth of which the most proximal
may be reduced.

Labium as wide as long or nearly so. Basal notch in type short,
about one fourth the total length.

Legs distally slender. Metatarsus IV equal in length to tibia C +
patella IV. Tarsi setose beneath, none truly scopulate. Tibia I
with three pairs of ventral spines, the distal reduced, the first two but
slightly exceeding the diameter of the joint. Tibiae III and IV above
with a stout median apical spine and a median basal spine.

Epigynum with median guide and transverse arms; lateral cavities
deeper cephalad as in Lycosa.

Genotype. — Orinocosa aymara, sp. nov.

Most readily distinguished from related genera by presence of the
stout median dorsal spines on the posterior tibiae. Its affinities seem
to be closest with Lycosa.

Orinocosa aymara, sp. nov.
Plate 24, fig. 7-8.

Carapace with sides nearly black; on each side a narrow, supra-
marginal light line which does not reach upon pars cephalica; median
pale stripe beginning back of second eyes extending between third
behind which it widens to width of eye-area, then abruptly narrower
some distance in front of the thoracic furrow and gradually narrowing
to caudal margin, the stripe embracing a dark mark back of each eye
of the third row. Sternum black, with a median longitudinal yellow
mark in the anterior half. Labium dusky. Endites brown or some-
what chestnut. Chelicerae light chestnut. Coxae of legs beneath
yellow; other joints yellow, with exception of tarsi closely ringed with
black, there being three distinct annuli on femora of which the distal
is very broad and two on the tibia of such breadth that only a small
ring of yellow remains at the middle; metatarsus also with two broad
rings nearly embracing the whole length. Venter of abdomen brown,
black at base of spinnerets; sides with a few black spots; a large
solid black spot over each antero-lateral corner; dorsum greyish yel-
low covered with a very fine areolation in dark, embracing a short,
faint lanceolate outline at base which sends off a pair of short side
branches at tip and one on each side near middle, behind tip of

292 bulletin: museum of comparative zoology.

lanceolate mark a small triangular black spot and on each side of
dorsal light area two black spots of which the most anterior is on a
level with tip of lanceolate mark.

Anterior row of eyes much shorter than the second (33 : 48) ; strongly
procurved; laterals with diameter three fourths that of the median;
medians their radius apart and half as far, or scarcely more, from
the laterals. Clypeus conspicuously receding; width about equal to
diameter of median eye. Eyes of second row more than twice the
diameter of an anterior median eye (about 5:2), not quite three
fourths their diameter apart (about 10:7), one fourth their diameter
from anterior median eyes. Area of median eyes very much wider
than long (ratio about 16:9), wider behind than in front in ratio 48: 17,
or nearly three times. Eyes of third row clearly smaller than those of
the second (diameters about as 17:20), less than 2.5 times their diam-
eter, apart. Area of posterior eyes to total length of carapace nearly
as 1:3.6 or 3.7.

Labium nearh' as wide as long; basal notches one fourth or but
little more of the total length of labium, (Plate 24, fig. 7).

Sternum about six sevenths as wide as the total length.

Lower margin of furrow of chelicera with three stout teeth of which
the most proximal may be smaller than the others; upper teeth as
usual.

Spines of anterior tibiae and metatarsi as usual in Pardosa or nearly
so. Patellae of posterior legs with a spine not only on each lateral
surface but also with two dorsal spines, one basal and one strictly
distal.

Tibia IV with spine at base above. None of tarsi scopulate,
strongly setose beneath. Paired claws with about 10, closely set teeth.

Female. Length 6 mm. Length of cephalothorax 3.2 mm.; width,
2.1 mm.

fem. tib. + pat. met. tar. total

Leg I 2.1 mm. 2.7 mm. 1.8 mm. 1.2 mm. 7.8 mm.

Leg II 2 2.3 1.4 1.1 6.S

Leg III 2 2.1 1.7 1 6.8

Leg IV .3 3.1 3 1.8 10.9

Localities.— Santa Ana, 3,000 feet, August. (M. C. Z. 294, one
female). Huadquina, 5,000 feet, July. (M. C. Z. 295).

chamberlin: the arachnida. 293

OXYOPIDAE.

OxYOPES SALTicus Hentz.

Boston journ. nat. hist., 1845, 5, p. 196, pi. 16, f. 10.

0. varians Taczanowski, Horae Soc. ent. Rocs., 1873, 10, p. 95.

0. gracilis Keyserling, Verh. Zool. bot. ges. Wien, 1877, 27, p. 698, pi. 2, f . 63,

64. O. P. Cambridge, Biol. Cent. Americana, 1902, 2, p. 342, pi. 32, f.

14, 15.

Localities.— Santa Ana, 3,000 feet, August. (M. C. Z. 332, five
females and one male.

Panama, June. M. C. Z., 333, three females and one male).

Peucetia rubralineata Keyserling.

Verb. Zool. bot. ges. Wien., 1876, 26, p. 704, pi. 2, 70, 71.

Localities. — Near Paltaybamba, 6,000 feet, August. (M. C. Zl
334, one female). Huadquina, 5,000 feet, July. (M. C. Z. 335, two
females and one male).

Tapinillus sp. a.

Two immature specimens of uncertain species).

Locality.— San Miguel, 6,000 feet, July. (M. C. Z. 336). "

Tapinillus, sp. b.

An immature specimen of uncertain species.

Loca/?/?/.— Huadquina, 5,000 feet, July. (M. C. Z. 337).

SALTICIDAE.

Dendryphantes bisquinquepunctatus Taczanowski.

Bull. Soc. imp. nat. Moscow, 1878, p. 309.

Locality. — Paltaybamba, originally described from Pumamarca,
Peru, 5,000 feet, August. (M. C. Z.296, one male).

294 bulletin: museum of comparative zoology.

Dendryphantes andinus, sp. nov.
Plate 25, fig. 5, 6.

Carapace with integument reddish brown, darker cephalad, but
covered in Hfe with hair and scales in a broad median longitudinal
band of metallic green color, this band being as wide cephalad as the
eye-area and narrowing strongly caudad, and in a wide supramarginal
band of white hair on each side. Chelicerae chestnut. Sternum,
labium, and endites brown; coxae of legs beneath yellow; other
joints of legs yellowish or testaceous, each with a black annulus at
distal end, but the first pairs darker, dusky chestnut throughout.
Abdomen above encircled with a stripe of white hair and along median
portion five pairs of small white dots outside of which are black dots,
the dorsum elsewhere being clothed with scales of the metallic green
lustre; venter brown, limited on each side with a line of white hair;
sides clothed with white and green hair intermixed.

Ocular quadrangle wider behind than in front (12:11) and wider
than long as usual. Eyes of second row minute, situated considerably
in front of the middle. Anterior row of eyes strongly recurved;
median eyes with diameter two and a third that of the laterals, one
seventh diameter apart and about same distance from the laterals.

Length of male 6.3 mm.; length of cephalothorax 3 mm., width,.
2.2 mm.

Locality.— San Miguel, 6,000 feet, July. (M. C. Z. 297, one male).

Dendryphantes calus,^ sp. nov.

Plate 25, fig. 7, 8.

Carapace with integument black or nearly so, a wide supramarginal
stripe of white scales on each side. Chelicerae black, the claws brown.
Sternum black. Coxae of legs testaceous ; femora and tibiae of leg I
black, patella black distally, testaceous proximally, distal half of
metatarsus I black, testaceous proximally, tarsus testaceous; other
legs paler, more yellowish, with narrow annulus of black at distal end
of each. Abdomen in life with the dorsum black, a band of white hair
across anterior surface and continuing half way back along the sides,

• xaXoj, beautiful.

chamberlin: the arachnida. 295

with three narrow Hnes of white extending from sides a short distance
mesad on the dorsum on each side at and caudad of the middle;
venter brown.

Ocular quadrangle much wider than long and wider behind than
in front in the usual way. Second eyes minute and well in front of
the middle. Eyes of first row about as in the preceding species.

Chclicerae long and slender; the claw slender and as long as the
chelicera, with a double curve; tooth of lower margin bent conspicu-
ously toward distal end of chelicera, (Plate 25, fig. 8).'

Tibia I with two pairs of ventral spines toward distal end and a
single ventral spine toward ectal side well toward base of joint. Meta-
tarsus I with two pairs of ventral spines, one distal and one at middle.

Length, 5.1 mm.; length of cephalothorax, 2.5 mm., width, 1.9 mm.

Localities.— Santa Ana, 3,000 feet, August. (Type, M. C. Z. 298,
male; paratype, M. C. Z. 299, one inale). San Miguel, 6,000 feet,
July. (M. C. Z. 331, one male). Paltaybamba, 5,000 feet, August.
(M. C. Z. 300, one male) Huadquina. 5,000 feet, July. (M. C. Z.
301, an immature female).

Dendryphantes amphibolus,^ sp. nov.
Plate 25, fig.' 9.

Carapace with integument reddish black, clothed over whole surface
with white scales of greenish lustre. Sternum and labium brownish
black, the endites a paler brown. Legs yellowish or pale testaceous,
the femora, patellae, tibiae, and metatarsi typically dark at. tips, at
least ventrally, but the annuli may be clearly evident only on the
posterior pairs. Abdomen above with several pairs of large brown
spots the most anterior of which are confluent across anterior end and
with less distinct chevron-lines between caudal ends of others ; in the
elongate yellowish grey area between anterior pairs of spots is a longi-
tudinal brown line or mark bisecting the same; the hair across the
anterior face and bordering the brown spots is white; venter dilute
yellowish grey, with or without three longitudinal brown lines on the
caudal portion. Spinnerets narrowly enclosed at base with brown.

Tibia I (female) with two pairs of short spines toward the distal
end and one ventral spine on the ectal or caudal side toward the base ;
metatarsus I with two pairs of spines, one distal and one mesal.

• ait4>lffo\os, ambiguous.

296 bulletin: museum of comparative zoology.

Tooth of lower margin of chelicera (female) short, stoutly conical,
acute; teeth of upper margin two, small (or an obsolete third tooth
may be present).

Ocular quadrangle much wider than long, conspicuously wider
behind than in front, with the minute second eyes in front of the
middle as usual. Eyes of the first row as in the preceding species.

Length of female 5 mm.; length of cephalothorax 2 mm.; width,
1.4 mm.

Localities. — Huadquina, 5,000 feet, July. (Type, M. C. Z. 302,
female; para types, M. C. Z. 303, six females). Paltaybamba, 5,000
feet, August. (M. C. Z. 304, three females).

Dendryphantes sp. a.

A female from Huadquina, 5,000 feet, July. (M. C. Z. 309) cannot
be satisfactorily determined.

Dendryphantes sp. b.

An immature female of doubtful species from San Miguel, 6,000
feet, July. (M. C. Z. 310).

Wala sp.

An immature male and female of uncertain species from San Miguel,
6,000 feet, July. (M. C. Z. 305).

Wala noda,^ sp. nov.

^ Plate 25, fig. 2.

Carapace with integument brownish black, rubbed in part but
apparently clothed in life with greyish scales which on the head are
more brownish. Legs brown. Sternum brownish black. Abdomen
grey-brown beneath. x\bove clothed with light grey scales except for
two pairs of elongate subtriangular dark areas of which the apices are
directed forwards and the edges are curved; a narrow chevron-mark
between the two pairs of dark spots.

1 puio^, edentate.

chamberlin: the arachnida. 297

Lower margin of furrow of chelicera unarmed or with but a minute
pale rudiment of a tooth.

Tibia I armed beneath with three pairs of spines; metatarsus I
with two longer pairs; tibia II beneath with an apical pair and a
single spine toward middle in the usual way.

Ocular quadrangle much wider than long (13:9), fully as wide in
front as behind. Eyes of second row slightly in front of the middle.

Length of female, 6 mm. ; cephalothorax, in length, 2.8 mm. ; width,
L8 mm.

Localiiy.— Torontoy, 8,000 feet, July 22. (T^-pe, M. C. Z. 306,
female; paratype, M. C. Z. 307, one female).

One female from the Conservidayo River in bad condition is this
species or very close to it.

EvoPHRYS CRUX Taczanowski.

Bull. Soc. imp. nat. Moscow, 1878, p. 284.

Described originally from Amable Maria, Peru.

Locality — Tincochaca, 7,000 feet, August. (M. C. Z. 311, female).

EvoPHRYS PERUVIANA Taczanowski.

Bull. Soc. imp. nat. Moscow, 1878, p. 280.

Previously known from Amable Maria and Pumamarca, Peru.
Locality. — Conservidayo River, August. (M. C. Z. 312, four
females) .

EvoPHRYS sp. a.

Locality — Santa. Ana, 3,000 feet, August. (M. C. Z. 313, one
immature female).

EvoPHRYS sp. b.

Zoca%.— Santa Ana, 3,000 feet, August. (M. C. Z. 314, one
immature female).

EvoPHRYS sp. c.

Locality.— Lucma, 6,000-7,000 feet, August. (M. C. Z. 315, one
immature male in coloration much like E. crux Tacz.).

298 bulletin: museum of comparative zoology.

EvoPHRYS sima/ sp. nov.
Plate 25, fig. 1.

Carapace with a broad median longitudinal light band anteriorly
as broad as the ocular quadrangle but narrowing caudad to a point
at the caudal edge, this band reddish on head and becoming yellow
caudad; sides blackish brown with a narrow supramarginal pale stripe
on each side. Sternum yellow. Last three pairs of legs yellow;
first pair of legs with femora reddish brown, more yellow distally,
the more distal joints light brown, the metatarsus clothed with a
brush of long spatulate hairs on ventral surface, similar but more
sparse ones on ventral surface of femur and patella. Palpi reddish
brown, the femur with numerous white hairs above and with black
scopulate hairs below which also occur on other joints. Dorsum of
abdomen black with a narrow median longitudinal yellow stripe over
the entire length, the edges of this stripe dentate along caudal half of
length; sides yellow finely dotted with black and a more solid stripe
below at the anterior end ; venter yellow with a few dots at the sides.
Spinnerets dusky.

Tibia I with the usual three pairs of ventral spines; one small
spine toward the distal end on lower anterior surface. Metatarsus I
with two pairs of spines as usual.

Ocular quadrangle a little wider in front than behind and rather
more than one fourth wider than long. Eyes of second row caudad of
middle. Eyes of first row nearly contiguous, medians twice the
diameter of the laterals.

Length of male, 4 mm. Length of cephalothorax, 2.2 mm.; width,
L6 mm.

Locality.— San Miguel, 6,000 feet, July. (M. C. Z. 316, one male).

Phiale panamae, sp. nov.

Plate 25, fig. 4.

Carapace black with a reddish yellow median longitudinal stripe
crossed by a black band at level of the third eye row and narrowing
to the posterior margin; also a reddish supramarginal stripe on each
side. Sternum yellow, dusky at margins. Legs and palpi dusky
brown, the posterior pairs somewhat paler. Spinnerets and venter

' Gosiute sima, one.

chamberlin: the arachnida. 299

of abdomen dusky ; dorsum black, the anterior face and anterolateral
corners and sides whitish, three pale spots in a triangle near middle
and several fainter, smaller ones farther caudad.

Ocular quadrangle slightly wider in front than behind; eyes of
second row minute, behind middle. First row of eyes decidedly
recurved ; eyes contiguous or nearly so, the median eyes with diameter
two and a half times that of the laterals.

Femora I and II with three moderately long spines at the distal end
on anterior part of dorsal surface and with two spines in the middorsal
line farther proximad. Tibia I with the usual three pairs of rather
short ventral spines and two on anterior surface ; metatarsus I with
two pairs of ventral spines and one on anterior surface at the distal end.

Tibia of palpus with apophysis at distal end from dorsoectal corner
subconical, distally curved.

Length of male, 4 mm.; length of cephalothorax, 2.1 mm.; width,
1.3 mm.

Locality. — Panama, June. (M. C. Z. 317, one male).

Phiale huadquinae, sp. nov.

Plate 25, fig. 3. '

Carapace with integument solid shining black, a band of white
hair along each side. Sternum and mouthparts also black, the endites
pale across tips. Last two pairs of coxae yellowish beneath, the
anterior pairs darker; femora, patellae, and tibiae black. Metatar-
sus black distally, testaceous proximally; tarsi testaceous. Integu-
ment of abdomen black clothed with hair of golden brown lustre;
hair of venter grey.

Lower margin of furrow of chelicera with one stout tooth; upper
margin with two teeth united at base.

Ocular quadrangle wider in front than behind (46:43). Eyes of
second row minute, at middle of length of quadrangle or scarcely in
front. Anterior row of eyes strongly recurved; median eyes a little
more than twice the diameter of the lateral, about one seventh their
diameter apart and from the laterals.

Tibia I with three pairs of spines below and three single ones on
anterior surface; metatarsus with two pairs beneath.

Length of male, 7 mm.; length of cephalothorax, 3.1 mm.; width,
2.1 mm.

Locality. — Huadquina, 5,000 feet, July. (M. C. Z. 318, one male).

PLATE 1.

•Chamberlin. — The Arachnida.

PLATE 1.

Tityus foot&i Chamberlin.

Fig. 1. Lateral view of sting and last two segments of abdomen.

Fig. 2. Finger, inner surface, showing arrangement of granules.

Fig. 3. Tarsus IV, anterior view.

Fig. 4. Comb.

Brachistostermis andinus Chamberlin.

Fig. 5. Carapace, dorsal view.

Fig. 6. Finger, inner view, showing granules.

Fig. 7. Tarsus IV.

Fig. 8. Comb.

BULL. MUS. COMP. ZOOl.

Chamberlin— Peruvian Arachnioa. Plate 1

R V. C.DEL.

PLATE 2.

Chamberijn.— The Arachnida.

PLATE 2.

Paravanones peruviantis Chamberlin.

Right mandible, ectal view.

Right pedipalp, ectal view.

Anterior portion of carapace from above and a httle to the left.

Femur and adjoiaing parts of leg IV, the proximal portion in dorsal

view, the distal portion in more dorsoectal view.
Trochanter IV (left), ventral view.
Showing process of coxa IV, dorsal view.

Gonoleptes enoplus Chamberlin.

Fig. 7. Dorsal view.

Fig. 8. Mandible, ectal view.

Fig.

1.

Fig.

2.

Fig.

3.

Fig.

4

Fig.

5.

Fig.

6.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnioa. Plate 2

R V. C.DEL.

PLATE 3.

Chamberun. — The Arachnida.

PLATE 3.

Gonoleptes enoplns Chamberlin.

Fig. 1. Coxa and trochanter of left pedipalp, ectal view.

Fig. 2. Left pedipalp, excepting proximal joints, ectal view.

Fig. 3. Leg III) distal portion, anterior view.

Fig. 4. Right leg IV, dorsoectal view.

Fig. 5. Coxal spur of leg IV, lateral view.

Goiwleptes scolius Chamberlin.

Fig. 6. Dorsal view.

Fig. 7. Mandible, ectal view.

Fig. 8. Left leg III, anterior view.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida, Plate 3

H. V. C..DE1..

^

PLATE 4.

Ghamberlin. — The Arachnida.

PLATE 4.
Gonoleptes scotiiLs Chamberlin.

Fig. 1. Left leg IV, subdorsal view.
Fig. 2. Right pedipalp, ectal view.

Gonoleptes huadquinae Chamberlin.

Fig. 3. Mandible, ectal view.

Fig. 4. Right pedipalp, ectal view (Tjrpe).

Fig. 5. Right pedipalp, ectal view (Paratype).

Fig. 6. Caudal processes of carapace, anterodorsal view (Paratype).

Fig. 7. Left process, lateral view.

Fig. 8. Left leg IV, dorsoectal view.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida. Plate 4

R V. C.,DEL.

PLATEjS.

Cbamberlin. — The Arachnida.

PLATE 5.
Pachylus orinus Chamberlin.

Fig. 1. Dorsal view.

Fig. 2. Right mandible, ectal view.

Fig. 3. Right pedipalp, ectal view.

Liopagus simplex Chamberlin.

Fig. 4. Eye-tubercle, caudal view.

Fig. 5. Eye-tubercle, lateral view.

Fig. 6. Right mandible, ectal view.

Fig. 7. Right pedipalp, mesal view.

Fig. 8. Part of left pedipalp, dorsal view.

Ldobunum monticola Chamberlin.
Fig. 9. Left pedipalp. ectal view.

BULL. MUS. COMP. ZOOL.

Chamberlin,— Peruvian Arachnida, Plate 5

R. V. C.,DEL.

PLATE 6.

Chamberum. — The Arachnida

PLATE 6.
lAobunum monticola Chamberlin.

Fig. 1. Lateral view of eye-tubercle.
Fig. 2. Left mandible, mesal view.
Fig. 3. Left mandible, ectal view.

Hemirrhagus peruvianvs Chamberlin.

Eyes, dorsal view.

Tarsal scale of usual type.

Tarsal scale of a second type.

Tibial spurs, right leg I of male.

Claw of leg IV.

Lock of tarsometatarsal joint (tarsus above).

Palpal organ of male, left ectal view.

Fig.

4.

Fig,

5.

Fig.

6.

Fig.

7.

Fig.

8.

Fig.

9.

Fig.

10.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida. Plate 6

9

9

R. V. C..DFL.

PLATE 7.

Chamberlin. — Tho Arachnida.

PLATE 7.

Hemirrhagus peruvianns Chamberlin.

Fig. 1. Anterior claw of leg I.
Fig. 2. Posterior claw of leg I.

Hemirrhagus viajor Chamberlin.

Fig. 3. Eyes (Type, Urubamba).

Fig. 4. Tibial spurs, left leg I of male, mesal view.

Fig. 5. Caudal claw, leg I.

Fig. 6. Claw of leg III, female.

Fig. 7. Claw of leg III, male.

Fig. 8. Right palpal organ of male, ectal view (T3T)e).

Diplura monticolens Chamberlin.

Fig. 9. Eyes, dorsal view.

Fig. 10. Eyes from in front and a Uttle above.

Brachythele keithi Chamberhn.
Fig. 11. Eyes, dorsal view.

Brachythele incursiis Chamberhn.
Fig. 12. Eyes, dorsal view (more enlarged than fig. 11).

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida. Plate 7

R. V. CDEL.

i

PLATE 8.

Ch.\mbeklin. — The Aracbnida.

PLATE 8.
Orinomus lamprus Chamberlin.

Fig. 1. Lateral view of body,

Fig. 2. Sternum.

Fig. 3. Epigynum.

Fig. 4. Labium.

AuximiLS productus Chamberlin.

Fig. 5. Sternum.

Fig. 6. Cribellum.

Fig. 7. Epigyn\un, caudoventral view.

Dictyna hesperia Chamberlin.
Fig. 8. Epigynum.

SULL. MUS. COMP. 200L.

Chamberlin— Peruvian Arachnida, Plate 8

6

8

R V. C.DEL.

PLATE 9.

Chamberlin. — The ArachnMa.

PLATE 9.

Aymarella munda Chamberlin.

Fig.

1. Lateral view of cephalothorax and mandibles

Fig.

2. Sternum.

Fig.

3. Labium.

Fig.

4. Epigynum.

Fig.

5. Cribellum.

Thomisoides terrosus Nicolet.

Fig. 6. Eyes, dorsal view.

Fig. 7. Mandible, ectal view, showing stridulating plate.

Fig. 8. Labium.

Fig. 9. Distal portion of mandible, ventral view.

Fig. 10. Posterior claw of leg L

BULL. MUS. COMP. ZOOL.

Chamberlin.— Peruvian Arachnida. Plate 9

^.r^-

R- V. C..DEL.

Obamberun. — The Aracbnida.

PLATE 10.

Thomisoides terrosus Nicolet.

Fig. 1. Dorsal view.

Fig. 2. Base of spine from leg.

Fig. 3. Distal end of bristle from leg.

(Figs. 1 to 3 drawn by Prof. A. Petrunkevitch).

Ariadna hotchkissi Chamberlin.

Fig. 4. Eyes, dorsal view.
Fig. 5. Labium.

Nops beUiUa Chamberlin.

Fig. 6. Sternum.

Fig. 7. Eyes and anterior part of carapace.

Fig. 8. Labium and left endite.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida, Plate 10

R V. C.GEl..

PLATE 11.

CHAMBEnLirv. — The Arachnida.

PLATE 11.

Nops bellnla Chamberlin.

Fig. 1. Dorsal view of abdomen.

Fig. 2. Distal portion of right leg I, caudal view.

Fig. 3. Mesal and anterior claw of leg I, caudal view.

Drassodes araucanius Chamberlin.

Fig. 4. Eyes, dorsal view.

Fig. 5. Sternum.

Fig. 6. Labium.

Fig. 7. Epigynum.

Fig. S. Hair of scopula, highly magnified.

BULL. MUS. COMP. ZOOL.

Ghamberlin— Peruvian Arachnida^ Plate 1 1

R V. C.DEL.

PLATE 12.

Cbamberlin. — The Arachoida.

PLATE 12.

Drassodes araucanius Chamberlin.

Fig. 1. Epigj-num, possibly lacking one moult of maturity.
Fig. 2. Tarsal claw.

Apodrassus andiniis Chamberlin.

Fig. 3. Sternum.

Fig. 4. Hair of fascicula of claw.

Fig. 5. Eyes, dorsal view.

Fig. 6. Labium.

Fig. 7. Claw of leg I.

Fig. 8. Spermathecae.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida, Plate 12

"•^"wiltrli^iii- iivr->.-.Mt>

6

8

R V. C. DCL.

PLATE 13.

Chamberlin.- — The Arachnida.

Fig.

1

Fig.

2.

Fig.

3.

Fig.

4.

Fig.

5.

Fig.

6.

Fig.

7.

Fig.

8

Fig.

9.

PLATE 13.

Hypsorinus binghamae Chamberlin.

Abdomen, lateral view.

Eyes, dorsal view.

Sternum, labium, etc.

Endite, subventral view.

Tip of same, sublateral view.

Patella, tibia and tarsus of palpus of adult female.

Unpaired claw of leg I.

Epigynum.

Serrated bristle or accessory claw of tarsus I.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida. Plate 13

R. V. C.DEL.

PLATE 14.

Cbamberlin. — The Arachoida.

PLATE 14.

Hypsorinus binghamae Chamberlin.

Fig. 1. Feather hair from near claw of leg I.

Fig. 2. Bristle from end of tarsus I above claws.

Fig. 3. End of chelicera of male from above.

Fig. 4. Same of female from above.

Fig. 5. Claw of leg I, female.

Fig. 6. Palpus of male, ectal view.

Fig. 7. Sfame, from above.

Litoporus aberrans Chamberlin.

Fig. 8. Sternum and labium.
Fig. 9. Abdomen, lateral view.

BULL. MUS. COMP. ZOO!

Chamberlin— Peruvian Arachnida. Plate 14

i^ V. C .01:..

PLATE 15.

Cb AM BERLIN. — The Arachnida.

PLATE 15.

Litoporus aberrans Chamhei-lin.

Fig. 1. Eyes from above.

Fig. 2. Right palpus of male, ectal view.

Fig. 3. Palpus of male, anterior view.

Argyrodes lucmae Chamberlin.

Fig. 4. Sternum.

Fig. 5. Abdomen, lateral view.

Fig. 6. Right palpus of male, dorsal view.

Theridion leguiai Chambeilin.

Fig. 7. Sternum.

Fig. 8. Eyes from above.

Fig. 9. Middle and anterior claw, caudal view.

Fig. 10. Epigynum.

k.

BULL. MUS. COMP. 200L.

Chamberlin.— Peruvian Arachnida, Plate 15

O Or.1

O

GO

R. V. C.-DEL.

PLATE 16.

Chambeblin. — Tlie Arachnida.

PLATE 16.

Theridion tosum Chamberlin.

Fig.

1.

Labium.

Fig.

2.

Sternum.

Fig.

3.

Eyes from above.

Fig.

4.

Epigynxim.

Garricola sanctiis Chamberlin.

Fig. 5. Lateral view of body.
Fig. 6. Sternum.
Fig. 7. Epigyniun.

Enoplognatha peruviana Chamberlin.

Fig. 8. Sternum,

Fig. 9. Paired claw of leg I.

Fig. 10. Unpaired claw.

Fig. 11. Epigynum.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruviaini Arachnida, Plate 16

^:' ■■■■:■■■■ ■ •■ 1!!5'5»^

'^^-.

R. V. C, DEL.

PLATE 17.

Chamberlin. — The Arachnida

PLATE 17.

Enoplognatha peruviana Chamberliii.

Fig. 1. Eyes dorsal view (anterior row above).
Fig. 2. Labium.

Enoplognatha dubia Chamberlin.
Fig. 3. Epigynum.

Erigone niwina Chamberlin.
Fig. 4. Epigynum.

Erigone taibo Chamberlin.
Fig. 5. Epigynum.

Oedothoraz melacra Chamberlin.

Fig. 6. Palpus.
Fig. 7. Epigynum.

Oedothorax orinus Chamberlin.
Fig. 8. Epigynum.

Tutibo debUipes Chamberlin.

Fig, 9. Epigynum.

Fig. 10. Palpus, distoventral view.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruv'an Arachnida, Plate 17

9

9

R- V. C. PEL.

PLATE 18.

Chambeblin. — The Arachnida.

PLATE 18.
Tetragnatha scopus Chamberlin.
Fig. 1. Chelicera from above.

Tetragnatha tincochacae Chamberlin.

Fig. 2. Chelicera of male from above.
Fig. 3. Palpus of male, mesoventral view.

Tetragnatha quechua Chamberlin,
Fig. 4. CheUcera from above.

Meta explorans Chamberlin.
Fig. 5. Epigyiimn.

Acacesia -peruviana Chamberlin.
Fig. 6. E}pig>Tium.

Eustala monticola Chamberlin.
Fig. 7. Epigynum.

Aranea duocypha Chamberlin.

Fig. 8. Abdomen, dorsal view.
Fig. 9. Sternum.
Fig. 10. Epigynum.

BULL. MUS. COMP. ZOOL.

Chamberlin.-Peruvian Arachnida, Plate 18

5

R V. C. DEL,

PLATE 19.

Crambbhi.in. — The Aracliuida

PLATE 19.

Aranea quechuana Chamberlin.
Fig. 1. Palpal organ of male, meso ventral view.

Aranea tigana Chamberlin.
Fig. 2. Palpal organ of male, mesoventral view.

Aranea orina Chamberlin.
Fig. 3. Palpal organ of male, mesoventral view.

Aranea calotypa Chamberlin.
Fig. 4. Palpal organ of male, mesoventral view.

Aranea plesia Chamberlin.
Fig. 5. Epigynum.

Aranea compsa Chamberlin.
Fig. 6. Epigynum.

Aranea sexta Chamberlin.
Fig. 7. Dorsal view of abdomen.

Gea panamensis Chamberlin.
Fig. 8. Left palpal organ of male, mesoventral view.

Scoloderus hyhus Chamberlin.
Fig. 9. Epigynum.

Aranea santa Chamberlin.
Fig. 10. Epigynum.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian arachnida. Plate 19

R, V. C.DEL.

PLATE 20.

CuAMBERLiN. — The Arachnidu.

PLATE 20.

Anamxia atopa Chamberlin.

Fig. 1 . Eyes from above and a little to the left.

Fig. 2. Abdomen, lateral view.

Fig. 3. Right leg II; anterior view of tibia, etc.

Micrathena cola Chamberlin.
Fig. 5. Dorsal view of abdomen.

Gelanor innominatum Chamberlin.
Fig. 6. Epigynum.

Eusparassus shefleli Chamberlin.

Fig. 7. Sternum.
Fig. 8. Labium.

BULL. MUS. CO MP. ZOOl

Chamberlin.—PcR'jvian Arachnida. Plate 20

R V. C..DEL.

PLATE 21.

Chambe:ri.in. — The Arachnida.

PLATE 21.

Eusparassus shefteli Chamberlin.
Fig. 1. Epigynum.

Horiocte7ms lycosoides Chamborlin.

Fig. 2. Labium.

Fig. 3. Epigynum.

Fig. 4. Distal portion of chelicera, ventral view.

Quechuella Imnpra Chamberlin.
Fig. 5. Labium.
Fig. 6. Sternum.
Fig. 7. Dorsal view of abdomen.
Fig, 8. Eyes from in front and a little above.

Trachelopachys bicolor Chamberlin.
Fig. 9. Labium,
Fig. 10. Epigynum.

BULL. MUS. COMP. ZOOL.

Chamberlin.— Peruvian Arachmida, Plate 21

4 0 ^ ^

8 OgoO

6

9

R V. C.,DEL.

PLATE 22.

Chamberun. — The Arachnida.

PLATE 22.

Castaneira quechua Chamber lin.
Fig. 1. Epigynnm.

Anyphaena apora Chamberlin.

Fig. 2. Labium.
Fig. 3. Epigynum.

Anyphaena andina Chamberlin.

Fig. 4. Epigynum.

Anyphaena poicila Chamberlin.
Fig. 5. Dorsal view of abdomen.

Gayenna monticola Chamberlin.
Fig. 6. Epigynum.

Tunabo peruvianus Chamberlin.

Fig. 7. Labium.

Fig. 8. Sternum.

Fig. 9. Distal portion of chelicera, ventral view.

BULL. MUS. COMP. ZOOL.

Chamberlin.-Peruvian Arachnida. Plate 22

6

_^j^es2rEs!3EgsSEi!st_

2

m

6

R. V. C..DEL.

PLATE 23.

Chamberlin. — The Arachnida.

PLATE 23.

Trechalea monticola Chamberlin.
Fig. 1. Paired and single claw.

Porrima harknessi Chamberlin.

Fig. 2. Labium.

Fig. 3. Male palpal organ, ventral view.

Fig. 4. EpigjTium.

Fig. 5. Lorum of pedicel.

Fig. 6. Claws.

Arctosa altamontis Chamberlin.

Fig. 7. Epigynum.

Fig. 8. Male palpal organ, ventral view.

Fig. 9. Labium.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida. Plate 23

2

8

9

R. V. C.DEL.

PLATE 24.

Chambeiii.in. — The Arachnids.

PLATE 24.

Lycosa orinus Chamberlin,
Fig. 1. Epigynum.

Lycosa andina Chamberlin.

Fig. 2. Epigynum.

Fig. 3. Palpal organ, ventral view.

Lycosa liopus Chamberlin.
Fig. 4. Epigynum.

Lycosa algina Chamberlin.
Fig. 5. Labium.
Fig. 6. Palpal organ.

Orinocosa aymara Chamberlin.
Fig. 7. Labium.
Fig. 8. Epigynum.

BULL. MUS. COMP. ZOOL.

Chamberlin.-Peruvian Arachnida, Plate 24

H V. C.DEL.

PLATE 25.

Chamberun. — The Aracbnida.

PLATE 26.

Evophrys sima Chamberlin.
P'ig. 1. Palpal organ, ventral view.

Wala noda Chamberlin.
Fig. 2. Epigynum.

Phiale huadquinae Chamberlin.
Fig. 3. Palpal organ.

Phiale panamae Chamberlin.
Fig. 4. Palpal organ of male.

Dendryphantes andinus Chamberlin.

Fig. 5. Palpal organ of male.

Fig. 6. Distal end of chelicera of male, caudal view.

Dendryphantes cuius Chamberlin.

Fig. 7. Palpal organ of male, subventral view.
Fig. 8. Distal end of chelicera of male.

Dendryphantes amphiholus Chamberlin. •
Fig. 9. Epigynum.

BULL. MUS. COMP. ZOOL.

Chamberlin— Peruvian Arachnida, Plate 25

R V. C.DEL.

^^ ibio

Bulletin of the Museum of Comparative Zoology
AT HARVARD COLLEGE. *

Vol. LX. No. 7.

A COLLECTION OF BIRDS FROM THE CAYMAN

ISLANDS.

By Outram Bangs.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

March, 1916.

No. 7. — A Collection of Birds from the Cayman Islands.
By Outram Bangs.

In the year 1911 the well-known collector, W. W. Brown, Jr.,
spent the spring and early summer, April to July, in the Caymans.
He visited all three islands and made a practically complete collection
of the resident, breeding land birds of the group. This beautiful lot
of skins, in Brown's inimitable make, fortunately remained intact and
was secured for the Museum of Comparative Zoology. Below I give
an annotated list of the collection which represents only the resident
ornis of the islands, being happily free from migrants.

In the Ibis for 1911, (ser. 9, 5, p. 137-161), Mr. P. R. Lowe published
a list of the birds of the Cayman Islands. Lowe's account of the
islands and his description of them is so good and complete as to leave
nothing more to be said. He also brought up to date all bird collecting
that had been done there. I must, however, give my opinion upon the
sources whence the bird life peculiar to the Caymans has been derived.

Lowe justly states that on account of the very recent origin of the
islands no genus and no very peculiar forms occur there alone. In the
main, this is true, but I think for the moment he had forgotten Mimo-
cichla ravida. This bird bears no close relationship to any other
existing species of the genus. We must, however, bear in mind that
Jamaica at present, alone among the Greater Antilles, is without a
species of Mimocichla. It is highly probable that a form similar to
M. ravida once occurred there and that the Cayman bird, now itself
on the verge of extinction, was derived from that form.

Coereba sharpei is a species of uncertain origin. The genus Coereba
has been in the near past, and perhaps is still, so very plastic that the
relationships of the various forms are hard to trace.

Dendroica viteUina (which also occurs in Swan Island) and D.
crawfordi, quite clearly indicate an instance, rather rare among birds,
of a migratory species establishing itself upon islands that lie on the
line of passage and becoming differentiated there; for clearly the
nearest relation of these two wood-warblers is the migratory North
American Dendroica discolor.

Three other peculiar forms were, I believe, received directly from

304 bulletin: museum of comparative zoology.

the near by mainland. These are the Elaenia which is much more
Uke E. martinica siibpagana, than it is any of the gray Lesser Antillean
forms; the Vircosyha caymancnsis which is very closely related to
V. magister of the coast of British Honduras and unlike any West
Indian form; and the Vireo, which so far as I can see is identical with
V. crassirostris the bird inhabiting the Bahamas (which are of simi-
lar formation). This species is so much like V. ochraccus of the
opposite coasts of Central America and so unlike any of the species
peculiar to the Greater Antilles, — Cuba, Jamaica, or Porto Rico,
that there seems no question of its origin.

The remainder of the Cayman birds have come from either Jamaica
or Cuba, in some case being still identical with the parent stock, in
others having differentiated into what may be called island species
or subspecies according to the degree of change. From Jamaica the
Caymans have received Leptoiila coUaris and Icterus bairdi. From
Cuba the islands have derived the two forms of Amazona peculiar to
them, CoJaptcs guruJIachi, Ccnturus caj/mancusis, Mimocichia cori/i,
two forms of Holoquiscalus, Spindalis salvini, and Mdopyrrha faylori;
probably also Tohnarckus caymanensis, although this species might be
descended from either T. caudifasciatus of Cuba or T. janiaiccnsis of
Jamaica.

From the above which discusses every bird peculiar to the Caymans
it will be seen that I am unable to recognize several forms which ha\e
been described as species or subspecies peculiar to the islands, and
these I comment upon at length in the following list.

At the time of Brown's \isitto Little Cayman and Cayman Brae —
June and July — the Boobies and Man-o'-War-Birds were not breed-
ing and all he saw during his stay were occasional birds off shore.
Besides these and the list of species following he saw and positively
identified only two species, namely Hydranassa tricolor ruficoUis
(Gosse) and Nyctanassa violacea (Linn,e).

Brown took nests and eggs of a number of the species; these are
preserved in the Museum of Comparative Zoology, and I believe
some of them have not before been collected. I have marked Avith
an asterisk each species of which he secured the nest and eggs; and
with a dagger the species of which the eggs only were taken.

As this paper was going to press, an article on the birds of Grand
Cayman appeared in the Ibis, January 1916, p. 17-35, by T. M. Savage
English. Mr. English apparently collected no specimens, but based
all his identifications on living birds observed afield. During his
three years' residence in the island he was able to add twelve species

bangs: birds from the cayman islands. 305

to Lowe's list. A few of these are merely migrants. Four others
I had already included in the following account of Brown's collection,
and I let my words stand as first written.

It would be of interest to know the bird that Mr. English found
breeding in Grand Cayman and called Chordeiles virginicmus, distin-
guishing it from Chordeiles minor by its larger size. On geographic
grounds it certainly could not have been Chordeiles virginianus vir-
ginicmus (Gmelin).

Ardeidae.

*BUTORIDES VIRESCENS BRUNESCENS (Lemb.).

Two specimens, an adult cf and an immature 9 , Grand Cayman,
May.

These are similar to specimens from Jamaica and Cuba. I have
already (Auk, Oct. 1915, 32, p. 481-484) given my reasons for using
the name brunescens for the Green Heron of the Greater Antilles.
Whether or not that form can be maintained as distinct from B. v.
vutculatus (Bodd.) of Martiniciue remains, I think, to be proved.
Oberholser has probablv' subdivided the Green Heron too much, and
he had but three specimens from Martinique when he wrote his
Revision of the subspecies of the Green Heron (Proc. U. S. N. M.,
1912, 42, p. 529-577).

Rallidae.

*Gallinula chloropus CACHiNNANS Bangs.

Brown found the Gallinule breeding in abundance in the many
marshy ponds in Grand Cayman in April and May, and took several
sets of eggs, but made up no skins.

Laridae.
t Sterna antillarum (Lesson).

One adult cf , Little Cayman, July 26. Browfi found the Least
Tern breeding in abundance.

306 bulletin: museum of comparative zoology.

Charadriidae.

Charadrius wilsonius Ord.

Brown noted Wilson's Plover on several occasions on the beaches,
where he thought it was breeding. He took no specimens. I include
it in this list on Brown's identification in spite of the lack of specimens,
in order to correct a rather curious error in Lowe's list where under
Aegialitis semipalmata Lowe says, "This bird is resident in Jamaica
and breeds there. Whether it does so in the Caymans I am unaware.
I have included it among the residents." No mention is made of
Wilson's Plover, and it seems certain that Lowe in some way confused
that species with the Arctic-breeding Semipalmated Plover.

t Hypsibates mexicanus (Miill.).

The Black-necked Stilt was breeding in numbers in the mangroves
in Grand Cayman in May. Brown took several sets of eggs, but did
not shoot any birds.

Columbidae.

CoLUMBA leucocephala Linue.

Three specimens, two males and a female, all adult, Grand Cayman
and Cayman Brae, May and June.

t Zenaida zenaida zenaida (Bp.).

Zenaida spadicea Cory, Auk, 1886, 3, p. 498, Grand Cayman.

Zenaida richardsoni Cory, Auk, 1887, 4, p. 7, Little Cayman.

Thirteen specimens, both sexes, all adult. Grand Cayman, Little
Cayman, and Cayman Brae, May, June, and July.

I can find no difference in specimens from the various islands of the
Cayman group, and after a most careful comparison, with adequate
material, am unable to distinguish in any way Cayman specimens,
which appear to me to be quite like examples from the Bahamas, Cuba,
and Jamaica, in the same condition of plumage.

Judging from Cory's name and description I fancy he took the
Grand Cayman Pea Dove to be darker than true zenaida. This
may have been because he compared Grand Cayman specimens
killed when in worn summer plumage, with skins from elsewhere in
autumn or winter dress, there being quite a change with season in

bangs: birds from the cayman islands. 307

Z. zenaida. This is wholly due, I think, to wear and the loss of the
bloom or sheen characteristic of the fresh plumage.

* Chaemepelia passerina insularis (Ridg.).

Eleven specimens, all adult males. Grand Cayman, Little Cayman,
and Cayman Brae. April, May, and June.

There is no difference in skins from the three islands. On comparing
this series with an enormous number of Cuban birds, I could find no
differences at all, in color, color of the bill, size, or anything else. I
therefore sent the series to W. E. C. Todd, as the latest authority on
this group of birds, and asked him for an opinion that I might quote
in print. He replied that he was now fully prepared to say that the
Cayman and Cuban forms are identical. Chaemepelia passerina
afiavida (Palmer and Riley) therefore becomes a synonym of P. p.
insularis (Ridg.). The Jamaican form, though very close still appears
to be recognizable. •

t Leptotila jamaicensis collaris (Cory).

Five specimens, both sexes, all adult. Grand Cayman, May and June.

Brown considers this Dove to be the rarest of all the peculiar Grand
Cayman birds. The five specimens he took were the result of days
spent hunting especially for it in its favorite haunts.

I can detect no differences whatever in color in comparing these
Grand Cayman skins with our twelve specimens of true L. jamaicensis
(Linne) from Jamaica. The Cayman examples appear to have less
white at the tips of the three outer rectrices. This may be partly,
perhaps wholly, due to their tail feathers being more worn down at
the ends. The Cayman specimens also average slightly smaller than
Jamaican ones, as the following measurements show, but the differ-
ence is so trifling that a larger amount of material might actually
tiu-n it the other way, and I doubt much if the form can be maintained.

Grand

Cayman.

Exposed

No.

Sex

Wing

Tail

Tarsus

Culmen

68334

&

148

103

29

16

68335

cf

146

90

31

16

68338

d^

153

105

31.5

15.5

68336

9

142

95

28

15.5

68337

9

144

102

27.5

16

308

>

bulletin: a

lUSEUM (

OF COMPARAT]
Jamaica.

[VE ZOOLOG

Y.

Exposed

No.

Sex

Wing

Tail

Tarsus

Cvlmen

37732

d^

159

111

31.5

16.5

37733

cT

161

112

31

17

37734

d"

157

109

29

16

54089

cf

154

109

32

16

3695

d"

153

108

30

16.5

3696

d'

158

108

33

17

37735

9

153

107

29

16

37736

9

151

105

29.5

16.5

37737

9

153

100

29

16

41841

9

159

106

32

16.5

54088

9

151

99

28.5

16

71554

9

151

96

28.5

16

PSITTACIDAE,

t x\mazona leucocephala caymanensis (Cory).

Ten specimens, both sexes, all adult, Grand Cayman, May and
June.

This well-marked form is peculiar to Grand Cayman. I agree with
Todd (Annals Carnegie mus., 1911, 7, p. 418) that its relationship to
true leucocephala of Cuba, which is close, is best expressed by the use
of trinomials.

Brown noted that the skin of the orbital region, ^•aried from white
to flesh-color. He took a set of four eggs together with the female
parent on May 12.

Amazona leucocephala hesterna, subsp. nov.

Ten specimens, both sexes, all adult. Little Cayman and Cayman
Brae, June and July.

Type, adult d^, M. C. Z. 68313, Cayman Brae, July 15, 1911, W. W.
Brown, Jr.

Characters. Similar to true A. leucocephala (Linne) of Cuba, but
smaller. Paler green, lime-green to mignonette-green (in the Cuban
bird about Kronberg's green); under tail coverts and under surface
of tail (be;s-ond the red base) paler and more yellowish; red belly
patch always large, more sharply contrasted and brighter red without

bangs: birds from the cayman islands. 309

lavender shimmer — bright hych-anger-red — (dark vinaceous, almost
always, more or less touched with lavender in true leucocephala);
outer surface of closed wing paler and duller blue, more greenish, —
much more as in A. l. caymanensis; shorter upper tail coverts, some-
times also longer upper tail coverts and lower rump feathers, more
or less extensively edged and tipped with red (Cuban examples of
true A. leucocephala seldom show such red markings, and when they
do only to a very slight extent). Brown noted the iris as brown
and the orbital skin as white to gra^-ish white, the tarsus as yellow.
Measurements.

Caymax Brag.

No. Sex

68312 cT

68313 . d"

68315 d"

68316 c?

68309 9

68310 9

68311 9

68314 9

68317 9

68308 9

Remarks. Brown found this Parrot to be not uncommon though
of rather local distribution in Cayman Brae. He also took one speci-
men in Little Cavman on Julv 25.

CUCULIDAE.

Crotophaga ani Linrie.

Three specimens, both sexes, all adult, Grand Cayman and Little
CajTnan, May and July.

* CoccYZUS minor nesiotes (Cab. and Heine).

Eleven specimens, both sexes, all | adult. Grand Cayman, Little
Cayman, and Cayman Brae, May, June, and July.

Culmen

Wing

Tail

Tarsus

from Cere

132

110

21

27

134

109

23

27

133

113

21

27

129

103

22

25

132

109

22

26

128

111

23

25

126

102

21

24

129

107

21

23

134

106

21

25

Little

Caymax.

131

ill

22

25

310 bulletin: museum of comparative zoology.

These specimens agree with Jamaican skins in size and proportions
and are a Uttle larger than the Bahaman form C. m. maynardi Ridg.
In the color of the under parts this series shows a wide range of indi-
vidual variation. The darkest ones are exactly like the paler speci-
mens from Jamaica and the palest ones like the darker examples of
maynardi. Thus as a whole the series averages a little paler below
than the average of a long series of nesiotes from Jamaica. All, how-
ever, were taken later in the season, than any skin we have from
Jamaica and are without doubt somewhat faded out.

Hybrididae.
Hybris perlata furcata (Temm.).

One adult 9 , Cayman Brae, June 27. This is a very pale and
gray individual, rather different from ordinary specimens from
Jamaica or Cuba. In a long series from Jamaica, however, we have
one skin that agrees with it exactly.

Caprimulgidae.
Chordeiles virginianus minor (Cab.).

One adult cf , Little Cayman, July 17.

Brown made no note on the abundance of the Little Nighthawk in
the Caymans, which I believe has not before been recorded from the
islands. It is possible that this individual may have been a migrant
from Cuba, where the bird breeds abundantl^r', but apparently does
not winter.

PiCIDAE.
t COLAPTES CHRYSOCAULOSUS GUNDLACHI Cory.

Thirteen specimens, both sexes, all adult, Grand Cayman, April,
May, and June.

This series shows in a marked degree the two characters pointed out
by Ridgway in his Birds of North and Middle America — smaller
size and smaller and narrower black malar patch in the males —
which distinguish it from the Cuban form. It is confined to Grand
Cavman.

bangs: birds from the cayman islands. 311

t Centurus caymanensis Cory.

Nine specimens, both sexes, all adult, Grand Cayman, April, May,
and June.

This strongly characterized island species, is one of the commoner
birds of Grand Cayman to which island it is confined.

Tyrannidae.
Tyrannus dominicensis dominicensis (Gmel.).

Four specimens, both sexes, all adult, Grand Cayman and Cayman
Brae, April and June.

TOLMARCHUS CAYMANENSIS (NicoU).

Twenty specimens, both sexes, young and adult, Grand Cayman,
Little Cayman, and Cayman Brae, April, May, June, and July.

This is a well-marked form peculiar to the Caymans, where, I be-
lieve, it does not differ either in color or size in the three islands of the
group. In fresh spring plumage (April specimens from Grand Cay-
man) the back is distinctly olivaceous, as compared with the gray
back in the Cuban form, T. caudifasciatiLS (D'Orbigny), in similar
plumage. In birds killed by May 25 and from then on through the
summer, the color of the back, by fading and wear, has changed to a
dirty grayish, quite the same as in Cuban skins in the same condition
of feather. In this plumage the Cayman bird can only be recognized
by its much duller, browner head, less contrasted with the gray of the
back — the head of the Cuban bird in worn plumage being very black
and sharply contrasted against the color of the back. The Cayman
bird also has a longer and more slender bill, this character being well
marked as an average one, but unfortunately failing in the case of
certain individuals. All the adults from Little Cayman and Cayman
Brae, except one, are in the worn and faded midsummer plumage
just referred to; the one exception is M. C. Z. 68248 Cayman Brae,
June 29, which, though taken on a date earlier than some others that
had not changed, has almost completed the postnuptial moult and
has again an olivaceous back. The color of the back in this skin is
quite the same as in the April specimens from Grand Cayman, while
the more faded Grand Cayman individuals killed May 25 are like the

312 bulletin: museum of comparative zoology.

ones from Little Cayman and Cayman Brae taken in June and July.
Brown took fully fledged young from July 10 to July 28. The wing
in the adults in this series ranges, in Grand Cayman skins, from 103-
107; in skins from Cayman Brae, 97-103; in the only adult from
Little Cayman (a cf ) it is 108. The Cayman Brae specimens have
the tips of the primaries a little more worn down than the Grand
CajTTian ones.

Myiarchus sagrae sagrae (Gundlach).

Myiarchus denigratus Cory, Auk, Oct. 1886, 3, p. 500, 502, Grand
Cayman.

Ten specimens, both sexes, all adult. Grand Cayman, April and May.
Apparently this bird is found in Grand Cayman only of the Caymans.
The specimens in the present series are indistinguishable in any way
from Cuban skins.

* Elaenia martinica caymanensis Berlepsch.

Elaenia martinica complexa Berlepsch, Proc. 4th International
ornith. congress, 1905, p. 395, Cayman Brae.

Twenty-six specimens, both sexes, all adult, Grand Cayman,
Little Cayman, and Cayman Brae, April, May, June, and July.

Specimens in exactly similar plumage from the three islands of the
Cayman group are absolutely alike, and no subdivision can be made.
I am sure Berlepsch was deceived by the artificial discoloration of
Maynard's Cayman Brae skins, upon examples of which he based
his E. vi. complexa. Two such skins are now before me and I do not
wonder at such a mistake being made.

The Cayman Elaenia fades and bleaches out late in summer,
losing all its colors. Two skins collected in Grand Cayman in
August, 188(3, by W. B. Richardson, have lost all traces of the colors
and markings of the form when in fresh plumage. The April speci-
mens in the present series from Grand Cayman are in beautiful fresh
unfaded plumage. Some of the late July skins from Cayman Brae
have nearly completed the postnuptial moult and are indistinguish-
able from these. Others taken at the same time had not commenced
to moult, and are nearly as faded out as the August examples just
referred to.

The Cayman bird appears an excellent form, but I cannot agree with

/^

bangs: birds from the cayman islands. 313

some other ornithologists that its nearest relation is E. m. riisii Sol. of
St. Thomas. Nor do I think it very closely related to any of the dis-
tinctly gray Lesser Antillean forms. It seems obviously much more
like E. m. subpagana Scl. and Salv. of the near by mainland, with which
it exactly agrees in size and markings and in color except in being j)aler
throughout. In good plumage the belly is uniformly yellow, the chest
dull yellowish gray, the throat grayish white, and the upper parts olive.
All these colors, however, are much paler than in the continental bird.
I think that this bird was derived not through any of the Lesser
Antillean forms, which on zoogeographical grounds would seem out of
reason, but like Vireosyha caymanensis and probably Vireo crassi-
rostris direct from the form occupying the adjacent mainland.

MiMIDAE.

^ MiMUS POLYGLOTTOS ORPHEUS (Linne).

One adult cf, Grand Cayman, May 14.

The Jamaican Mockingbird is abundant in Grand Cayman, but
apparently is wanting in the two smaller islands.

TURDIDAE.

MiMocicHLA RAViDA Cory.

Thirteen specimens, both sexes, all adult. Grand Cayman, April,
May, and June.

The Grand Cayman Thrush belongs in a group of the genus Mimo-
cichla by itself, and of all the birds peculiar to the Caj'mans is the only
one that is very distinct, having no representative elsewhere. In view
of the recent origin of the ornis of the Caymans, it is probable that
there was somewhere, possibly in Jamaica, where no member of the
genus now occurs, a related form which has disappeared.

The Thrush is now extremely rare and local in Grand Cayman.
Brown covered the whole island and found it only in two remote
patches of woodland. Each of these tracts of rather heavier forest
than is usual in the island now-a-days was inliabited by a few pairs of
thrushes, which Brown believes to be the entire population of the
island. In each of these woods Brown was careful to leave birds
enough to perpetuate the species, if it is not gradually becoming extinct
from some natural cause, as seems to be the case.

314 bulletin: museum of comparative zoology.

Brown noted the colors of the soft parts to be as follows: — "Iris,
brown; tarsus, bill, and bare skin of orbital region, coral red."

MiMOCiCHLA RUBRiPES CORYI Sharpe.

Twenty-three specimens, both sexes, all adult, Cayman Brae,
June and July.

Unlike its cousin of the larger island, the Cayman Brae Thrush is an
extremely abundant bird. It is a very well-marked form, with a large
light-colored bill.

ViREONIDAE.
ViREOSYLVA MAGISTER CAYTVLAJMENSIS (Cory).

Twelve specimens, both sexes, all adult. Grand Cayijian, April and
May.

Brown found this Vireo in the mangroves in Grand Cayman, where
it was not uncommon. It has been recorded from both Little Cayman
and Cayman Brae by Cory, but Brown did not find it in either of the
smaller islands, where its place seemed to be wholly taken by V.
calidris barbatula.

The Grand Cayman Vireo is very closely related to true V. magister
LawT. of the coast of British Honduras, from which it differs only by
its paler coloration.

ViREOSYLVA CALIDRIS BARBATULA (Cab.).

Fifteen specimens, both sexes, all adult. Little Cayman and Cayman
Brae, June and July. ^

The Black-whiskered Vireo was very common in the two smaller
islands of the group. The skins show no differences when compared
with Cuban examples.

ViREO CRASSiROSTRis CRASSiROSTRis (Bryant).

Vireo alleni Cory, Auk, Oct. 1886, 3, p. 500-501, Grand Cayman.

Seventeen specimens, both sexes, all adult, Grand Cayman, Little
Cayman, and Cayman Brae, April, May, June, and July.

This series critically compared with our sixty-fovu- skins from the
Bahamas proves beyond a doubt that the much discussed V. alleni

/

bangs: birds from the cayman islands. 315

is absolutely identical with the Bahama bird. All the Cayman
examples are in the yellow phase of plumage. They correspond
exactly with yellow specimens from the Bahamas from Inagua to
New Providence, the type locality of V. crassirostris. The three
characters that Ridgway in his Birds of North and Middle America
thought might distinguish V. alleni, all prove illusive. The browner
back in the specimens he examined was due entirely to discoloration
from the now famous chemical preservative used by Maynard; the
outermost primary is not smaller; and the pale wing-bands are not
broader.

Todd (Annals Carnegie mus., 1911, 7, p. 428^30) has discussed at
length the color-phases of V. crassirostris, and I wholly agree with him
that the gray and the yellow (the so-called Vireo crassirostris fiavescens
Ridg.) specimens, represent nothing but extremes of color-variation
in one and the same subspecies.

Examples from the different islands of the Caymans are all quite
alike.

Mniotiltidae.
Dendroica petechia petechia (Linne).

Dendroica auricapilla Ridg., Proc. U. S. N. M., Aug. 1888, 10, p. 572,
Grand Cayman.

Thirteen specimens, both sexes, adults and two young. Grand Cay-
man, Little Cayman, and Cayman Brae, April, May, and July.

This series together with four skins from the Cajnnans already in
the M. C. Z. I have compared most carefully with a fine set of
Jamaican speciniens, with the result that I find no way in which to
separate them. Ridgway in his Birds of North and Middle America
recognizes auricapilla as differing from petechia on the grounds of
"decidedly shorter wing and larger bill and feet." His own measure-
ments, however, which followed, show very trifling differences. My
measurements of eight adult males from the Caymans, the wing is : —
62-65, (63.81); exposed culmen, 10-11.5 (10.62). In eight adult
males from Jamaica, the wing is: — 62-67 (64.5); exposed culmen,
10-11 (10.68). I can see no differences at all in the feet.

There are no differences in specimens from the three islands of the
Caymans.

Dendroica petechia petechia can be separated from D. p. gundlachi
Baird of Cuba by slightly paler colors and more extensively ochraceous
€rown.

316 bulletin: museum of comparative zoology.

Dendroica vitellina vitellina Cory.

Ten specimens, both sexes, all adult. Grand Cayman, April and May.

This fine island form confined to Grand Cayman, was in Brown's
experience a very uncommon bird and he told me that it was with
difficulty that he got even the ten noted above.

Dendroica vitellina crawfordi Nicoll.

Thirty-seven specimens, both sexes, adults and young, Little Cay-
man and Cayman Brae, June and July.

This is a well-marked subspecies whose characters were accurately
noted by Nicoll, (Bull. B. O. C, 1904, 14, p. 95) who also figured it
(Ibis, 1904, ser. 8, 4, pi. 11, f. 1).

It is an abundant bird in the two smaller islands, and is quite the
same in both.

Coerebidae.

* COEREBA SHARPEI (Cory).

Twenty-eight specimens, both sexes, all adult, Grand Cayman,
Little Cayman, and Cayman Brae, April, May, June, and July.

Brown's specimens from Grand Cayman are unfortunately not
comparable with his series from Little Cayman and Cayman Brae,
and I am unable to say whether the differences shown by birds from
the two smaller islands, when compared with examples from Grand
Cayman, are seasonal or not. I am inclined, however, to regard these
differences as only seasonal. The Grand Cayman birds, all taken in
April and May, were in worn and somewhat faded breeding plumage,
while those from Little Cayman and Cayman Brae, taken in late
June and July, had completed or were just completing the post-
nuptial moult, and were therefore all in what might be called fresh
autumnal plumage. The upper parts in the Grand Cayman specimens
are dull brownish black; the yellow of the under parts is pale and dull.
The upper parts in the Little Cayman and Cayman Brae skins are
grayish black with a slight olivaceous cast; the yellow of the under
parts is richer and rather more orange. Brown noted that the " skin
at corners of mouth, red" in the Grand Cayman bird; " skin at corners
of mouth, flesh-color" in Little Cayman and Cayman Brae specimens.
This possibly also has to do with the breeding season.

bangs: birds from the cayman islands. 817

icteridae.
* holoquiscalus caymanensis caymanensis (cory).

Foiu" specimens, three males and a female, all adult. Grand Cayman,
May.

Brown had to spend so much time while in Grand Cayman search-
ing for the rare species, that he rather neglected the Grackle and some
of the other very common birds.

This is a very well-marked insular subspecies peculiar to Grand
Cayman.

HOLOQTJISCALUS CAYMANENSIS CARIBAETJS Todd.

Fourteen specimens, both sexes, adults and one young. Little
Cayman and Cayman Brae, June and July.

The Grackle of the two smaller islands which differs from true
H. caymanensis of Grand Cayman in its much larger size and stronger
bill, has always been referred to H. gundlachii (Cassin) of eastern Cuba.
I had in the present paper corrected this old error, and had named the
form as new, arriving at the same conclusions as Todd, except that he
did not know the bird of Cayman Brae and Little Cayman, which is
identical with that inhabiting the Isle of Pines and western Cuba.

Todd's paper. The Birds of the Isle of Pines, Annals of the Carnegie
museum, 10, nos. 1-2, (dated Jan. 1916, but received by M. C. Z.
Mar. 1, 1916), containing a description of the form, came just in time
to allow me to change the name while reading proof.

Icterus bairdi Cory.

Seventeen specimens, both sexes, adults, and five immature (one
year old?) birds still carrying a partly or wholly greenish yellow tail.
Grand Cayman, April, May, and June. A nest made of palm fibres
and attached to a hemp palm leaf about sixty feet from the ground
was found 28 May; the nest contained three young birds.

This splendid island species confined to Grand Cayman differs from
/. leiicopteryx (Wagler) of Jamaica, from which it obviously was de-
rived, in being bright golden yellow only slightly tinged with olive on
the head, and just a trifle darker on the back than it is below. It is
also a little smaller and has a slightly slenderer and more delicate bill.

318 bulletin: museum of comparative zoology.

Baird's Oriole has always been extremely rare in collections, in fact
besides our series there exist only Cory's original specimens and two
in the Tring Museum that were collected by Taylor when he visited
the island in 1896 for the Hon. Walter Rothschild.

The species seems to be on the verge of extinction. Why this is
I can offer no suggestion. Certainly I. leucopteryx is common enough
in Jamaica and adapts itself to all the changes man makes there.

Brown found this Oriole scattered here and there at wide intervals
in the island and told me he thought it was one of the rarest birds he
had ever himted for.

Tanagridae.
* Spindalis salvini Cory.

Fifty-five specimens, both sexes (only five females) all adult. Grand
Cayman, April and May.

This is a fine, large species peculiar to Grand Cayman. Its nearest
relative is clearly S. pretrei (Lesson) of Cuba. Its bill though of course
larger than in the Cuban species, the bird itself being much larger, is
very like it, and quite different from the heavy coarse bill of <S. bene-
dicti Ridg. of Cozumel Island.

The female, I believe, was previously unknown; in color it is some-
what like the female of S. pretrei, (it is of course much larger), the upper
parts are, however, paler and more grayish olive, the under parts are
more uniform, the belly and under tail coverts not whitish but dull,
pale yellowish olive, and the chest is slightly paler olive.

This is another of the Cayman birds that has been very rare in col-
lections; Brown, however, tells me that it is really not uncommon
in Grand Cayman, but that it keeps itself hidden away in the dense
scrubby woods where it is difficult to shoot, females being especially
hard to find.

Fringillidae.
* TiARis olivacea olivacea (Linne).

Euetheia coryi Ridg., Auk, Oct. 1898, 15, p. 322, Cayman Brae.

Nineteen specimens, both sexes, all adult. Grand Cayman and Cay-
man Brae, April, May, and July.

The species has been recorded from Little Cayman, but Brown
during his short stay in that island did not find it.

bangs: birds from the cayman islands. 319

Some years ago Ridgway separated the Cayman Brae form based on
specimens collected there by Maynard. Some of Maynard's skins
of this bird are in the M. C. Z. so discolored by his chemical preserva-
tive as to be practically unidentifiable, and I am afraid even Ridgway
was deceived by them. Specimens in the present collection from
Cayman Brae are absolutely identical in color as well as in size with
those from Grand Cayman. In adult males from Grand Cayman
the wing runs 49-51.5; in adult males from Cayman Brae the wing
runs 48-51, the tips of the primaries are slightly more worn down in the
Cayman Brae skins. Birds from the Caymans are as a whole like
Jamaican specimens, and are slightly different from the average of
Cuban examples.

We have now in the M. C. Z. upwards of 150 skins of T. olivacea
from the Greater Antilles, and after a very critical study of these
specimens, I think the species might by very close splitting be sub-
divided. Individual variation, however, is so great and the characters
that separate birds from the various islands so subtle that the wisdom
of so doing is very questionable. If subdivided, the forms of the
Greater Antilles would stand, probably, as follows: —

Tiaris olivacea olivacea (Linne).
Haiti and Santo Domingo.

Slightly browner olive-green above and on flanks; yellow of throat
often very pale (the color of the tlu-oat-patch is, however, subject to
much individual variation in all the forms).

Tiaris olivacea lepicla (Linne).

Cuba and Isle of Pines.

Inclined to be darker and duller, than are the other forms, the upper
parts often dull dusky olive-green; the flanks darker and encroaching
more on belly; belly seldom yellowish.

Tiaris olivacea adoxa (Gosse).

Jamaica and the Caymans.

Usually paler and more grayish olive-green above and on flanks;
belly paler and often washed with pale yellowish.

I have no doubt that the subject of Gosse's plate was a young indi-

320 bulletin: museum of comparative zoology.

vidual of this form. If, however, Gosse's bird is considered uxiidenti-
fiable, then the name coryi Ridgway becomes available for it.

Tiaris olivacea hryanti (Ridg.).

Porto Rico.

Averaging slightly smaller than the other races, and slightly brighter
olive-green above; belly more yellowish. Perhaps the best of the
Greater Antillean forms.

* Melopyrrha taylori Hartert.

Fifty-one specimens, both sexes, adults and immature (one year
old?) males. Grand Cayman, April, May, and June.

This is one of the very strongly characterized species of Grand
Cayman. Brown found it to be far from uncommon, though usually
keeping well concealed in the scrubby woods.

Bulletin of the Museum of Comparative Zoology
AT HARVARD COLLEGE.
Vol. LX. No. 8.

ANTS COLLECTED IN TRINIDAD BY PROFESSOR

ROLAND THAXTER, MR. F. W. URICH,

AND OTHERS.

By William Morton Wheeler.

CAMBRIDGE, MASS., U. S. A.

PRINTED FOR THE MUSEUM.

March, 1916.

No. 8. — Ants collected in Trinidad by Professor Roland Thaxter, Mr.

F. W. JJrich, and Others.

contributions from the entomological laboratory of the
bussey institution. harvard university. no. 108.

By William Morton Wheeler,
formicidae.

1. Ectatomma tuberculatum Olivier. ^ 9- — Port of Spain and
Sangre Grande (Thaxter).

2. Ectatomma ruidum Roger. ^. — Port of Spain (Thaxter);
Chaguanas (Urich).

3. Neoponera obscuricornis Emery var. latreillei Forel. S . — Caura
(Urich).

4. Neoponera unidentata Mayr. ^ . — Port of Spain (Thaxter).

5. Pachycondyla crassinoda Latreille. S 9 . — Port of Spain
(Thaxter).

6. Pachycondyla harpax Fabricius. y . — Port of Spain (Thaxter).

7. Pachycondyla iinpressa Roger. ^. — Port of Spain (Thaxter).

8. Euponera (Mesoponera) cojistricta Mayr. ^ 9 . — Port of Spain
' (Thaxter).

9. Euponera ( Trachymesopu^) stigma Fabricius. ^ 9 . — Port of
Spain (Thaxter).

10. Ponera opaciceps Mayr. ^ . — Aripa Savanna (Thaxter).

11. OdontomachtLS haematoda Linne. ^ 9. — Port of Spain, Gas-
paree Island, and Sangre Grande (Thaxter).

12. Odontomachus haematoda Linne subsp. insularis Guerin var.
hirsutiusculu^ F. Smith, y . — Port of Spain (Thaxter).

13. Odontomachus haematoda Linne subsp. meinerti Forel. ^ . —
Port of Spain (Thaxter).

14. Anochetus inermis Ern. Andre var. meinerti Forel. ^ 9 cf . —
Chaguanas (Urich); Port of Spain (Thaxter).

The worker and female of this variety differ from those of the
typical form in having the superior border of the petiole distinctly
excised and the inner border of the mandibles with three teeth.

324 bulletin: museum of comparative zoology.

15. Anochetu3 (Stenomyrmex) emarginatus Fabricius (typical). S . —
Port of Spain (Thaxter); Ariopita Valley (H. D. Chapman).

16. Eciton burchelli Westwood. ^. — Port of Spain (Thaxter).

17. Eciton burchelli Westwood var. urichi Forel. y . — Port of
Spain (Thaxter); Erin (Urich).

I doubt whether this variety will prove to be valid. Among a
large number of specimens from the same colony received from Urich,
there are numerous transitions in color to the tj'pical form.

18. Ecito)t (Labidus) crassicorne F. Smith. § . — Matura (Urich) ;
Port of Spain (Thaxter).

19. Eciton {Acamatus) pilosurn F. Smith. S . — Aripa Savanna
(Thaxter).

20. Pscudomyrma championi Forel var. paulina Forel. ^ . — Port
of Spain (Aug. Busck and Thaxter).

21. Pseudomyrma excavata MajT. ^ . — Port of Spain (Thaxter).

22. Pseudomyrma pallida F. Smith, g . — Aripa Savanna (Thaxter).

23. Pheidole {Macropheidole) fimbriata Roger. Ql. — Port of Spain
(Thaxter).

24. Crematogaster brasiliensis MayT. ^ . — Aripa Savanna and
Sangre Grande (Thaxter).

25. Crematogaster limata F. Smith subsp. parabiotica Forel. S . —
Port of Spain and Gaspari Island (Thaxter).

26. Monomorium floricola Jerdon. ^. — Port of Spain (Thaxter).

27. Megalomyrmex bituberculatu^ Forel. S. — Arima (Urich); Port
of Spain (Thaxter).

28. Tranopelta gilva MayT. 9 cf ■ — Port of Spain (Aug. Busck).

29. Solenopsis geminata Fabricius. ^ . — Port of Spain (Thaxter) ;
Chaguanas (Urich).

30. Solenopsis minutissivia Emery. S . — Trinidad (Thaxter) .

31. Solenopsis altinodis Forel. ^ . — Port of Spain (Thaxter).

This species, which is easily recognized by the peculiar high petiolar
node, rectangular in profile and laterally compressed, the absence of
distinct clypeal ridges and the distinctly marginate epinotum, has been
recorded from Trinidad by Forel. The types are from Zigzag, Vene-
zuela.

32. Soletiopsis tenuis MayT. S . — Port of Spain (Thaxter).

33. Wasmannia auropunctata Roger. ^ 9 . — Port of Spain (Thax-
ter).

wheeler: ants collected in trinidad. 325

34. Mycocepurus smithi Forel. S . — Diego Martin (Urich).

35. Apterostigma wasmanni Forel. S . — Four Roads, Port of Spain
(Thaxter) ; " from fungus garden under a log."

36. Apterostigma urichi Forel. U. — Caparo (Thaxter); "from
fungus garden under a log."

37. Trachymyrmex urichi Forel. ^ . — Ariopita Valley (H. D.
Chapman); Gasparee Island (Thaxter).

38. Trachymyrmex humilis, sp. nov.

Worker. Length 2.2-2.5 mm.

Mandibles rather long, with concave external borders, three large
apical and several smaller basal teeth. Head subrectangular, as
broad as long, with broadly and feebly excised posterior and rather
convex lateral borders and rounded posterior corners. Eyes moder-
ately convex. Clypeus short, with nearly straight, entire anterior
border. Expanded anterior lobes of frontal carinae moderately large,
rounded, not angular, posterior ridges diverging but not reaching the
posterior corners of the head. Praeorbital carinae straight, not curved
inward across the antennal scrobes, terminating a little behind the eyes.
Antennal scapes moderately stout, reaching a distance not exceeding
their greatest transverse diameter beyond the posterior corners of the
head. Joints 2-8 of the funiculi not longer than broad, two terminal
joints forming an indistinct club, the penultimate longer than broad
and half as long as the last joint. Thorax with the pro- and mesono-
tum rather convex and rounded in profile, the mesoepinotal constric-
tion short and deep. Inferior pronotal spines short, moderately
acute. Epinotum much higher than long, the base in profile very
convex anteriorly, sloping behind, broadly sulcate above, somewhat
longer than the declivity, the spines reduced to two teeth which are
only slightly longer than broad at their bases, directed upward, out-
ward, and backward. Petiole small, not longer than high, the node
feebly developed, acute in profile, with longer concave anterior and
short concave posterior slope. Postpetiole very large, more than twice
as broad as the petiole, as long as broad, broadest behind, with a
median semicircular impression at the posterior border; in profile
the node is very convex and high in front, the remaining dorsal surface
flattened. Gaster suboblong, with rounded anterior and posterior
corners, a little longer than broad, with straight, marginate sides, its
upper surface evenly and feebly convex, without any longitudinal
impressions. Legs moderately long.

Mandibles shining, with a few very coarse, elongate punctures.

326 bulletin: musetjm of comparative zoology.

Remainder of body opaque, very densely punctate-reticulate. The
tubercles on the head, thorax, petiole, postpetiole, and gaster are
small and rather uniformly distributed, noticeably so on the posterior
corners of the head and dorsal surface of the gaster. On the front and
vertex of the head they are somewhat elongate so that the general
effect is that of several frequently interrupted rugae. Tibiae and
femora covered with minute, uniformly distributed tubercles. What
correspond to the spines and projections on the head and thorax of
other species of Trachymyrmex are reduced to tubercles not much
smaller than the teeth on the epinotum.

Hairs yellowish, very short, hooked, moderately abundant but not
conspicuous. Pubescence of the same color, short, distinct only on
the antennal funiculi.

Uniformly brownish ferruginous; mandibles a little darker, legs a
little paler than the remainder of the body.

Two specimens; one from Gasparee Island and one from Port of
Spain (Thaxter).

This species is very peculiar in its small size, small petiole, Jarge
postpetiole, and the great reduction of the spines and tubercles on the
head and thorax.

39. Acromyrmex octospinosus Reich. ^ . — Gasparee Island (Thax-
ter); Ariopita Valley (H. D. Chapman).

40. Atta cephalotes Linne. S . — Port of Spain and Sewa Valley
(Thaxter).

41. Cryptocerus piisillus Klug. S . — Aripa Savanna (Thaxter).

42. Cryptocerus (Zacryptocerus) clypeatus Fabricius. ^ . — Sangre
Grande (Thaxter); Port of Spain (U. S. N. M.).

43. Cryptocerus (Cephalotes) atratus Linne. S . — Port of Spain
(Thaxter).

44. Strwnigenys saliens Mayr. U. — Port of Spain (Thaxter).

Codiomyrmex, gen. nov.

Worker. Monomorphic, closely related to Strumigenys F. Smith,
Epitritus Emery, and Glamyromyrmex Wheeler, but differing in the
shap>e of the head. Mandibles large, swollen, triangular, their apical
margins with numerous, regular, acute teeth. Clypeus well developed,
projecting over the extreme bases of the mandibles and not separated
behind by distinct sutures from the head. Frontal carinae widely
separated, expanded horizontally and continued backward to form
sharp lateral margins as far as the posterior corners of the head, over-

wheeler: ants collected in TRINIDAD.

327

arching broad scrobes for the antennae dorsal to the eyes, which are
small but otherwise well developed. Ocelli absent. There is a short,
sharp longitudinal carina ventral to the insertion of each antenna and a
small acute tooth at each lateral corner of the gula near the lateral
insertion of the mandible. Ocelli absent. Frontal area represented
by a smooth, transverse region in the sculpture of the head; frontal
groove represented by a raised line extending back to the vertex.
Antennae, robust, 6-jointed. Thorax, petiole, postpetiole, and gaster
much as in Strumigenys. Spongiform appendages well developed on
the petiole, postpetiole, and base of gaster. Head coarsely sculptured.
Squamiform or clavate hairs absent, but both the body and appendages
covered with long, soft, dense, pointed hairs.

45. Codiomyrmex thaxteri, sp. nov. (Fig. 1).

Worker. Length nearly 2 mm.

Mandibles very convex dorsally and laterally, somewhat narrowed

Fig. 1. Codiomyrmex thaxteri, sp. nov. Body of worker in profile; head of same
from above.

at their insertions, their straight apical margins furnished with numer-
ous, crowded, acute, and equal teeth. Head subtriangular, a little

328 bulletin: museum of comparative zoology.

longer than broad, decidedly broader behind than in front, with
straight sides, rounded posterior corners, and narrow, excised, and
marginate occipital border. In profile the head is very convex in the
region of the vertex above and posterior portion of the gula below,
flattened in front and on the sides to the sharp edge of the frontal
carinae and their backward continuations. Clypeus rather concave,
about as long as broad, with semicircular, entire anterior border.
Eyes convex, at about the middle of the head, but near the ventral
surface and not visible when the head is seen from above. Antennae
robust, scapes terete but distinctly swollen, first funicular joint as long
as the second and third together. These are subequal and scarcely
longer than broad. Third joint longer than broad and nearly ^ as long
as the terminal joint. Thorax slender, through the pronotum about
half as broad as the head, broadest through the humeri which are very
projecting and distinctly angular or conical. Pro- and mesonotum in
profile gently convex, not separated by a suture, the mesonotum
laterally marginate and with a median longitudinal ridge, the meso-
pleurae high and rather concave. The lateral marginations of the
mesonotum are continued back over the base of the epinotum into the
spines, which are well developed, straight, and acute, nearly as long
as the base of the epinotum and directed backward, upward, and out-
ward. Their bases are laterally compressed and translucent below.
Epinotal declivity concave, somewhat shorter than the base, its
inferior angles compressed but not acute. Petiole with a long pe-
duncle, as long as the node, which rises abruptly in front and has a
gently convex, backwardly sloping dorsal surface; seen from above
the node is as long as broad and evenly rounded and submarginate on
the front and sides. Attached to the ventral border of the peduncle
and node is a long, compressed, band-shaped, transparent, spongiform
appendage. Postpetiole from above, transversely elliptical, nearly
twice as broad as the petiolar node and twice as broad as long, in profile
a little longer than high, evenly convex above, with a large and promi-
nent ventral and two smaller, lateral spongiform appendages. Gaster
a little larger than the head, elliptical, with nearly straight anterior
and rather sharply marginate lateral borders, its ventral as convex as
its dorsal surface and nearly the entire surface formed by the first seg-
ment. Its anteroventral surface is furnished with a flat, squamiform
and pointed spongiform appendage. Legs rather long and stout.

Mandibles shining, evenly and sparsely punctate. Head sub-
opaque; clypeus and upper surface of head reticulate-rugose and
coarsely punctate; antennal scrobes and gula densely and evenly
punctate; region of the frontal area smooth and shining. Thorax

wheeler: ants collected in trinidad. 329

smooth and shining, except the epinotum, the base of which is coarsely,
the declivity and sides more finely reticulate-punctate. Petiole and
postpetiole shining, the node of the former opaque, coarsely and
somewhat longitudinally reticulate-rugose, the node of the latter
coarsely and sparsely punctate. Gaster smooth and shining, its
extreme base with short, longitudinal rugae. Antennal scapes coarsely
and densely punctate, opaque ; legs smooth and shining.

Hairs yellowish gray, very fine, long, dense, flexuous, and erect,
covering the whole body and legs, as long on the latter as on the former,
shorter and subappressed on the antennae, where they are if anything
even denser. Pubescence absent.

Castaneous brown; clypeus, head, and antennal scapes black;
mandibles, bases of antennal funiculi, neck, knees, tibiae, tarsi, ante-
rior portion of first gastric segment, and whole of terminal gastric
segments, deep red.

Described from three specimens taken by Professor Thaxter in the
neighborhood of Port of Spain.

This species is very easily recognized by the singular shape of the
head and peculiar fleece-like pilosity of the body. I have made it
the type of a distinct genus, though it is evidently much like a Strumi-
genys, except in the structure of the head, because I believe that this
latter genus is soon destined to suffer disintegration into a number of
subgenera or genera. This fate has already overtaken several other
ant-genera (Camponotus, Formica, Crematogaster, Monomorium,
Pheidole, etc.) that have become unwieldy through accumulation of
species which even a very conservative myrmecologist must regard as
heterogeneous.

46. Dolichoderus attelaboides Fabricius. y. — Arima (Urich); Port
of Spain (Thaxter).

47. Dolichoderus decollatus F. Smith. ^ . — Port of Spain (Thaxter).

48. Dolichoderus {Hypoclinea) hidens\j\im.€ . ^. — Tamana (Urich).

49. Dolichoderus {Hypoclinea) championi Forel var. taeniatus Forel.

y . — Port of Spain (Thaxter). A single worker of very small
size, but agreeing in color and structure with cotypes from
Colombia.

50. Dolichoderus (Hypoclinea) championi Forel subsp. trinidadensis
Forel. ^ . — Port of Spain (Thaxter).

A single worker differing from a cotype specimen received from
Professor Forel only in having the head, tibiae, and antennal scapes
dark brown.

330 bulletin: museum of comparative zoology.

51. Dolichodcrus (Monacis) bispinosus OMxiei. ^ 9 cf. — Port of
Spain (Thaxter); Erin (Urich).

52. Dolichoderus {Monads) dehilis Emery. U. — Matura (Urich);
Sangre Grande (Thaxter).

53. Iridomyrmcx dispertitus Forel subsp. micans Forel. S . — Port
of Spain (Thaxter).

54. Tapinoma mrlanocephalum Fabricius. S . — Aripa Savanna
(Thaxter).

55. Azteca chartifex Forel. ^ . — Arima (Urich).

56. Azteca chartifex Forel subsp. decipiens Forel var. lanians Forel.

^ 9 . — Arima (Urich).

57. Azteca barbifex Forel. ^ . — Port of Spain (Thaxter).

58. Azteca trigona Emery subsp. viediops Forel. S . — Port of Spain
(Thaxter); Ariopita Valley (H. D. Chapman).

59. Azteca for eli Emery subsp. ursina Forel. ^ ,- — Chatham, "on
cacao" (Urich).

60. Azteca velox Forel. ^ . — Arima (Urich).

61. Azteca velox Forel var. nigriventris Forel. S. — Port of Spain
(Thaxter).

62. Prenolepis (Nylanderia) longicornis Latreille. ^ . — Sangre
Grande (Thaxter).

63. Prenolepis (Nylanderia) vividitla Nylander. ^ . — Port of Spain
(Thaxter).

64. Camponotus {Dinomyrmex) agra F. Smith. ^ . — Platanal
(Urich).

65. Camponotus (Myrmothrix) abdominalis Fabricius. ^ . — Port
of Spain (Thaxter); Ariopita Valley (H. D. Chapman).

66. Camponotus (Myrmothrix) femoratus Fabricius. S . — Port of
Spain (Thaxter).

67. Camponotus (Myrmobrachys) excisus Ma>T. ^ . — Port of Spain
(Thaxter).

68. Camponotus {Myrmobrachys) senex F. Smith. ^ . — Port of
Spain (Thaxter).

69. Camponotus {Myrmobrachys) lindigi Mayr. S . — Gasparee Is-
land (Thaxter).

70. Camponotus {Myrmobrachys) zoc Forel. y . — Ariopita Valley
(H. D. Chapman).

71. Camponotus {Myrmorhachis) latangulus Roger. S . — Port of
Spain (Thaxter).

72. Camponotus {Myrmorhachis) bidens Mayr. S . — Port of Spain
(Thaxter) .

hY\\ ^U

iD\'^

Bulletin of tha Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LX. No. 9.

BATS OF THE GENUS CORYNORHINUS.

By Glover M. Allen.

With One Plate.

CAMBRIDGE, MASS., U. S. A.: ,
PRINTED FOR THE MUSEUM.
April, 1916.

No. 9. — Bats of the Genus Corynorhinus.
By Glover M. Allen.
Introduction.

Specimens of the Big-eared Bats of the genus Corynorhinus have,
until very lately, been few in museums, and most of those available
to previous writers have been preserved in alcohol, so that they were
of little value in determining color variation. The desirability of
bringing together a large series of skins for comparison was emphasized
twenty years ago by Miller (1897) in his review of the Vespertilionidae
of North America, but no later attempt at a revision of the genus has
been made. Thanks to the generous interest of Professor Theodore
Lyman, the Museum of Comparative Zoology has lately acquired a
small series of these bats from southeastern California, and the identi-
fication of these and other specimens in the Museum has induced me
to undertake a general review of the genus. I have been fortunate
in being able to assemble most of the skins and skulls available in
American museums, some 126 specimens in all, covering practically
the entire known range of these bats.

Acknowledgements.

My thanks are due to the officers of several institutions for the loan
of specimens under their charge, and particularly to Mr. H. W.
Henshaw, Chief of the Biological Survey of the U. S. Department of
Agriculture and to Mr. Gerrit S. Miller, Jr., of the U. S. National
Museum for use of the series in the unrivalled collections at Washing-
ton. For the loan of valuable material from California, including the
type of C. VI. intermedins, I am indebted to Dr. Joseph Grinnell and
Mr. H. S. Swarth of the Museum of Vertebrate Zoology of the Uni-
versity of California. Acknowledgements are also gratefully made to
the following persons and institutions: Dr. J. A. Allen and Mr. R. C.
Andrews of the American Museum of Natural History; Mr. W. H.
Osgood of the Field Museum of Natural History ; Mr. Junius Hender-
son of the University of Colorado Museum; Mr. CD. Bunker of the
University of Kansas Museum, Dr. H. L. Ward of the Public Museum

334 bulletin: museum of comparative zoology.

of Milwaukee, Prof. Z. P. Metcalf of the North CaroHna College of
Agriculture, Prof. F. Payne of Indiana University, and Mr. M. L.
Church of Marshall, N. C.

Generic Affinities.

The Big-eared Bats of the genus Corynorhinus take the place of the
Old World vespertilionine genus Plecotus in North America. In
essential characters the two genera are very similar, and are evidently
closely related. Miller (1907, p. 225) in his " Families and Genera of
Bats" considers that "the great development of the glandular masses
on muzzle, and the absence of the distinct lachrymal ridge, distinguish
this genus sufficiently from Plecotus," to which LeConte and con-
temporary writers at first referred specimens. The shape of the
nostrils is also diagnostic, but the presence of a distinct lachrymal
ridge in a new species from Mexico (see p. 352), invalidates that charac-
ter as distinctive of Plecotus. In 1864, Harrison Allen used Synotus
of Keyserling and Blasius for the American bat, but in the following
year he erected for it the new genus Corynorhinus, by which it con-
tinues to be known, although Dobson in his Catalogue of Chiroptera
in the British Museum (1878) took the more conservative course of
regarding it as a subgenus of Plecotus. The dental formula — i 3^
c Pi, 'ptn 3Z3, wi 333 = 36 — is the same in both genera and shows
but slight numerical reduction over that of Myotis, in the presence of
two in place of three upper premolars on each side. In view of this
somewhat primitive or unreduced tooth formula it is perhaps less
surprising to find among the many specimens examined a single one
with three upper incisors. This condition is perhaps to be considered
reversionary to the more primitive state in which the full number of
three upper incisors characteristic of placental mammals, is present.
It has been generally assumed that it is the innermost upper incisor
that was the first to be lost in all bats with two incisors, partly be-
cause of the " correspondence of the two upper teeth with th' two outer
of the lower jaw when the maximum set is present, and also, even more
strongly, by the general tendency throughoiit the group for the pre-
maxillaries to become reduced, particularly along the inner edge"
which would "inevitably result in eliminating that part of the bone
in which the first incisor grows" (Miller, 1907, p. 27). Andersen
(1912, p. xxiv) without going further into the matter, asserts, however.

ALLEN: BATS OF THE GENUS CORYNORHINUS. 335

that the outermost incisor (r^) is the one lost in all known Chiroptera.
In the specimen of Corynorhinus mentioned (C. m. toivnseyidii, Biol.
Surv. Coll. 150273, from Happy Camp, California) it is evidt nt from
the agreement in form, that the two inner incisors correspond with the
two normally present in the genus, and that the supernumerary one
has be.'n added at the outer side, — is in fact the i^ usually missing in
all living bats. In outline (Plate 1, fig. 1) this tooth is roughly a
right-angled triangle with its height a little less than its base. It is a
very little shorter than i^ but much stouter, and with a long base,
rather than with the terete form characteristic of the second incisor.
In crown view it has a broad cutting edge, as broad as the crowns of
the other incisors. The skull of the specimen is unfortunately in
fragments, and the corresponding teeth of the left side are lost, but
the remaining teeth are normal. The case is instructive as indicating
not only that it is P that has been lost in the Chiroptera, but that in
this case, it was probably a larger tooth than i^ which is retained.

Geographic Distribution.

The general limits of distribution for the genus are now fairly well
ascertained. In the East it has been reported from Micanopy,
Florida, in the northern part of the peninsula, but to the southward
of that point there are no records. Northward it occurs throughout
Georgia and South Carolina, to western North Carolina and Virginia.
West of the AUeghenies, the northward limit of the range includes
Kentucky, southern Indiana, and west of the Mississippi swings north
again to southwestern South Dakota and the Yellowstone Park in
northwestern Wyoming. Between the Rocky Mountains and the
Sierra Nevada the records are few, but the genus undoubtedly is found
in southern Idaho and in Nevada. On the west coast, Vancouver
Island, British Columbia, seems to be the northernmost limit, and
thence it ranges south in the Sonoran zones, to the tableland of
Mexico as far as Oaxaca and Vera Cruz. Apparently it has not yet
been discovered in the peninsula of Lower California. In general it is
characteristic of the Austral zones as defined by Merriam, though in
the northwest, the subspecies townscndii is mainly confined to the
Transition and even enters the Boreal zone. This more northward
range in the northwest is possibly indicative of a more extended
northward distribution in ancient times, when we may assume that

336 bulletin: museum of comparative zoology.

the ancestral stock was enabled to reach America from northeastern
Asia by follow ing land connections. In the particular accounts of the
forms recognized in this paper, I have indicated more precisely the
limits of distribution of each.

Habits.

The Big-eared Bats are essentially cave-dwellers. In the West
they frequently haunt the abandoned shafts and tunnels made by
miners. Numbers of them may inhabit a single such tunnel, but they
appear to rest singly, scattered along the rock walls, rather than in
clusters. J. K. Townsend (1839) in the journal of his expedition to
the Columbia River, Oregon, in 1834, relates that they often lived in
the storehouses at the forts, and were considered by the fur traders to
be beneficial in ridding such places of Dermestes.
■ There is no evidence to indicate that any of the forms are migratory.
In the northern part of their range they retire to suitable ca\'erns to
hibernate. Hahn (1909) records finding specimens in caves at
Mitchell, Indiana, during the winter of 1906-7 and Butler (1895)
obtained two from Greencastle, Ind., 23 December, 1894. Brimley
(1905) reports one taken 1 February, 1893, in Bertie Co., North
Carolina. The University of Colorado has two from Boulder Coimty
in that State, captured in mid-winter, one on 21 January, 1912, in a
mine-tunnel where the temperature was 48° F., the other on 23 Febru-
ary, 1910, in a tunnel at fifty feet from the surface (altitude 7760 feet).

The young are probably born in early July or even earlier in the
southern part of the range. Stephens (1906, p. 265) records a female
of C. m. imUesccns captured at San Diego, California, on 25 April,
that contained a single foetus. In the San Jacinto Mts., of Southern
California, Grinnell and Swarth (1913, p. 379) collected a female
containing a single large foetus about 5 June. They found that the
adult bats in a resting posture folded the long ears back against the
sides, close to the body, a habit which Hahn (1909) seems to ha\e been
the first to record in the case of specimens from Indiana. In a
freshly killed specimen, however, the ears project forward.

History and Nomenclature.

What is perhaps the first mention of a bat of this genus, is found in
Clapton's (1722, p. 594) account of the animals and other products of

ALLEN: BATS OF THE GENUS CORYNORHINUS. 337

Virginia. Among mammals, he lists " Bats, as I remember, at least
two sorts; one a large sort with Long Ears, and particularly long
straggling Hairs; the other much like the English, something larger,
I think, very Common." The long ears of the first sort may perhaps
identify it with Corynorhinus; the other was possibly an Eptesicus.
It was not until 1818, however, that the naturalist Rafinesque named
and briefly described Vcspertilio megalotis from a specimen captured
somewhere on "the lower parts of the Ohio" River, the Wabash, or
the Green River, perhaps in Indiana. His name, which I think must
hold for a bat of this genus, has been generally ignored in favor of
LeConte's later name Plecotus macrotis based probably on specimens
from Georgia. A study of the large number of skins which I have
been able to assemble from many parts of the range of the genus shows,
rather unexpectedly, that the bat of the eastern United States west of
the Alleghenies is quite different from the dark brown, white-bellied
animall of the south Atlantic and Gulf States, to which the name
macrotis strictly applies. In the species inhabiting the interior and
western parts of the United States, the contrast in color between the
tips and the bases of the hairs of both surfaces is not abrupt as in
macrotis but passes imperceptibly from a dark base to a differently
colored tip, nor are the hairs of the lower surface tipped with pure
white. To this species, Rafinesque's name must apply. West of the
Mississippi, this species gradually becomes paler, and over the Rocky
Mountain area and the Southwest is a dull buffy color. To this race
of megalotis, Miller's (1897) name imllescens applies. On the humid
northwest coast, a gradual darkening takes place, and a strongly
marked subspecies is again recognizable, to which the name townsendii
was given by Cooper in 1837. On the Mexican tableland, the same
type of bat is found, but of a dark smoky hue and slightly reduced
proportions, which I here describe as new. Apparently the white-
bellied Corynorhinus macrotis of the Atlantic slope and Gulf States
as far west at least as Louisiana, does not intergrade with the differ-
ently colored representatives to the West, and I am therefore provi-
sionally regarding it as a distinct species. Still a third species, with
very large and differently formed ears and peculiar skull is represented
by a single specimen from central Mexico, and has remained hitherto
undescribed. The brief synonymy given under each name in the
descriptions which follow, indicates sufficiently the opinions of previous
writers as to the nomenclature of this genus.

338 bulletin: museum of comparative zoology.

CoRYNORHiNUS MEGALOTis (Rafinesque).

Rafinesque's Big-eared Bat.

Vespertilio megalotis Rafinesque, Amer. monthly mag., 1818, 3, no. 6, p. 446.
Plecotus rafinesquii Lesson, Manuel de mammalogie, 1827, p. 96. (Renaming

of Rafinesque's V. megalotis).
Corynorhinus macrotis Miller, N. Amer. fauna, 1897, no. 13, p. 51 (? in part,

Kentucky specimen cited).

Type. — None specified, and original specimens not known to be
extant.

Type Locality. — "The lower parts of the Ohio" River, probably in
southern Indiana and Illinois or western Kentucky in the region
between the ^Yabash and the Green Rivers.

Distribidion. — Central eastern United States from extreme western
Virginia, through Kentucky, southern Indiana and Illinois, to Kansas,
intergrading wdth the race pallcscens to the westward.

General Characters. — Largest of the megalotis-macrotis group;
bases and tips of hairs, above and below, not strongly and sharply
contrasted in color.

Color. — Adults: bases of the hairs, on dorsal surfaces of the body,
gray or slaty gray shading by imperceptible degrees into a 'wood
brown' (Ridgway, 1912) at the sides, and a 'clove brown' over the
median area of the back. The amount of ' clove brown ' wash over
the back varies slightly in individuals, but conduces to a much darker,
more drabby appearance than is found in typical pallcscens. Downy
hairs at the bases of the ears posteriorly are whitish. Ventral sur-
faces 'pale pinkish buff,' the bases of the hairs shading into grayish.

Immature specimens (No. 157075 Biol. Surv. C oil., Burke's Garden,
Va., 7 August, 1908), are uniform dark 'hair brown' to 'fuscous'
above to the bases of the hairs; below, pale 'hair brown' the hair
along the sides and on the belly paling at the tips to a dirty whitish.
Compared with immature pallescens, it is more uniformly dark, lacking
the light buffy admixture. It is also darker and larger than the
Mexican race.

Skull. — The skull is largest of all the viegalotis-macrotis group, with
broad depressed rostrum, and large brain case. The intermaxillary
notch viewed from above is rather larger with wider-bowing sides
than that of macrotis. The inner upper incisor is usually without
trace of a secondary outer cusp, though in one of four specimens from

ALLEN: BATS OF THE GENUS CORYNORHINUS. 339

Burke's Garden, Virginia, a cusp is barely indicated by a minute
shoulder. In macrotis this cusp is normally well developed.

Measurements. — No. 157076, Biol. Surv. Coll., from Burke's Garden,
Bland County, Virginia: forearm 45 mm. (average of three adults
44.1); digit III, metacarpal 39.8; first phalanx 13.8; second phalanx
19; tibia 19. Collector's measurements: total length 107 mm., tail
49; hind foot 12; extent of wings 320.

Skull: greatest length 17.4 mm.; basal length 14; palatal length 8;
zygomatic breadth 9; interorbital constriction 4 ; mastoid breadth 9.2 ;
width of braincase 8.6; upper tooth row 6.7; lower tooth-row 7.3.

Remarks. — The discovery of a Corynorhinus distinct from C. ma-
erotis, from extreme western \'irginia, westward, in the eastern United
States was wholly unexpected. It is the eastern representative of the
desert-colored paUescens of west-central United States, from which it
chiefly differs in its somewhat darker, more drab, coloration. The
skull is a trifle larger as well. I have applied to it Rafinesque's name
megalotis, based on specimens which he collected from " the lower parts
of the Ohio" River probably in southern Indiana or western Ken-
tucky, where the genus is known to occur at the present time (see
Cory, 1912, p. 476). Rafinesque's description is brief, yet I think
unmistakable in the light of our present knowledge. Nevertheless,
Miller, in 1897, rejected his name as unidentifiable and applied Le
Conte's later name, macrotis, to the species of Corynorhinus in eastern
North America. The discovery of a species in the interior distinct from
macrotis makes it necessary either to erect a new name or to recognize
megalotis as applicable to it, and this latter course I propose to adopt.
In 1818, Rafinesque sent to the editors of the American Monthly
Magazine brief accounts of the animals he discovered in the course of
his journey "through the western region of the United States," and
gave new names to many of these. In one of these communications
written in October, he gave a brief statement of certain supposed new
species obtained since July, during which time he says, he had visited
" the lower parts of the Ohio, the Wabash, Green River, Barrens,
Prairies, and the states of Indiana, Illinois, &c." This is the general
locality whence he obtained the new bat which he describes as follows :

"9. VespertiUo megalotis R. (Big-eared Bat.) Tail three-eighths
of total length, body dark gray above, pale gray beneath, ears very
large, duplicated, auricles nearly as long. Length 4 inches, breadth
12 inches."

The evident similarity to the Old World Plecotus, led Lesson in
1827 to change the genus and (as the custom then was) the specific

340 bulletin: museum of comparative zoology.

name as well, calling it Plecotus rafinesquii. The description of the
color ("dark gray above, pale gray beneath") though inexact, is
certainly applicable to the present form and not at all to macrotis.
The statement that the ears are "duplicated" is descriptive of the
manner in which the inner rim folds upon the rest of the conch, and
will apply to no other of the eastern genera with which he could have
met, except possibly the very different Nyctinomus; in which the
" auricle " [i. e. tragus] is not " nearly as long." Even in Corynorhinus
the tragus is hardly more than half the length of the ear. The meas-
urements given, — "length 4 inches [= 101.5 mm.], breadth 12 inches
[= 304.5 mm.]" are not far from those of the Virginia specimens
(lengths 107, 108, 110; extent 320, 313, 313) allowing for differences
in manner of taking these dimensions. The tail is nearer one half
than three eighths of the total length. In spite of slight discrepancies,
I think the description can apply to no other bat of the eastern United
States. The very name is diagnostic. When LeConte proposed the
name Plecotus macrotis for the Big-eared Bat of the coast States, he
acknowledges its similarity to the species of the interior, by his
remark in a footnote: "There is another species with equally long
ears, which are not united on the cranium; which of these is the
megalotis of Raffin., it is impossible to say." In view of these facts,
I think the propriety of using Rafinesque's name is no longer open to
question.

West of the Mississippi, typical megalotis (of which in lack of topo-
types I have assumed the Virginia specimens to be representative)
grades by insensible degrees into the more buff-colored subspecies
jjallescens, and the latter again shades rather abruptly into the dark-
colored townsendii of the humid Pacific coast area. Two specimens
from Sun City, south central Kansas, in the Biological Survey Collec-
tion, though not as dark as the Virginia megalotis are better referable
to it than to pallescens. Specimens from eastern and central Colorado
are intermediate, but on the whole, nearer pallescens. To the south-
ward as well as to the north, the limits of the range remain to be more
carefully worked out.

Many years since. Dr. Harrison Allen (1864, p. 64) recorded on the
authority of Professor Baird, " that specimens of a Synotus, probably
of this species [i. e., viacrofis], were received some years ago by the
Smithsonian Institution, from Meadville," Crawford County, north-
western Pennsylvania. The specimens have been lost and no sub-
sequent captures of this bat have been made in the State. The record,
though discredited by Rhoads (1903, p. 226) may nevertheless be

ALLEN: BATS OF THE GENUS CORYNORHINUS. 341

valid, and if substantiated, would fix nearly the northern limit for
the species in the east. It is not known that the genus ranges farther
north than Virginia on the eastern side of the Alleghanies, though to
the westward of that range its presence at a more northerly latitude
is well known.

Specimens examined. — Eight from the following localities :

Virginia: Burke's Garden, 4 (Biol. Surv.).

Kansas: Sun City, 2 (Biol. Surv.).

Colorado: 12 miles south of Lyons, Boulder Co., 1 intermediate
(Univ. of Colo.); Crisman, Boulder Co., 1 intermediate (Univ. of
Colo.).

Additional locality records, probably referring to this bat, are:
Indiana, Greencastle (Putnam Co.) (see Cory, 1912, p. 476); Ken-
tucky, Bowling Green (Miller, 1897).

CORYNORHINUS MEGALOTIS PALLESCEN8 Miller.

Pallid Big-eared Bat.

Synotus townsendi H. Allen, Smithsonian misc. coll., 1864, 7, p. 65 (not of

Cooper, 1837).
C[orynorhinus\ townsendi H. Allen, Proc. Acad. Nat. Sci. Phila., 1865, p. 175

(? not of Cooper, 1837).
Corynorhinus townsendii H. Allen, Bull. 43, U. S. N. M., 1893, p. 58, (not of

Cooper, 1837).
Plecotus {Corinorhinus) townsendi Trouessart, Cat. Mamin., 1897, fasc. 1,

p. 105 (in part).
Corynorhinus macrotis pallescens Miller, N. Amer. fauna, 1897, no. 13, p. 52,

fig. 10.

Type. — Skin and skull 65534, U. S. N. M. (Biological Survey Col-
lection), adult female, collected 3 August, 1894, by A. K. Fisher.

Type Locality. — Arizona: Navajo County, Keam Canyon.

Distribution. — Western United States from western Texas, Colo-
rado, and southwestern South Dakota, to the Pacific coast of southern
California. Typical pallescens may yet be found to occur in northern
Mexico, but none have been examined from there.

General Characters. — Similar to typical megalotis but slightly
smaller; colors paler, more buffy throughout.

Color. — Adult: no sharp contrast in color between bases and tips
of hair. General effect of an average specimen (Prescott, Ariz.)

342 bulletin: museum of comparative ztoology.

'pinkish buff,' the hairs paUng on their middle third and darkening by
imperceptible degrees to gray or slate gray at their extreme bases.
DoAvny hairs at the posterior bases of the ears whitish. Below, the
general appearance is 'pale ochraceous buff,' the hairs darkening
gradually toward their bases to a neutral gray, except in the middle
region of the throat, where usually the hairs are not perceptibly darker
at base.

Immature specimens have the pelage more dusky throughout than
the adults. The basal two thirds of the hairs above is nearly ' neutral
gray' with short pale tips of buffy, nearly 'vinaceous buff.' Below,
the color is paler, the light tips of the hairs a soiled whitish.

Color Variation. — In a large series of skins that may fairly be taken
to represent pallescens, there is much individual variation in the
intensity of coloring among adults. A specimen from Ash Creek,
Graham IMts., Arizona (204375 Biol. Survey Coll.) has the basal halves
of the hairs 'slate color' producing an effect much darker than usual.
It may be considered a step in approach to the dark subspecies of the
Mexican highlands. Two specimens (10694, 10695 Univ. of Cali-
fornia Coll.) obtained by Dr. Joseph Grinnell at Riverside Mountain,
Colorado River, southeastern California, are the brightest colored
individuals I have seen, with a distinct reddish cast to the upper
surface, nearly 'vinaceous cinnamon,' shading into a 'buff pink'
below, the bases of the hairs only slightly darkened. The palest
specimen of all is one collected in the hills back of Lone Pine, California,
by Dr. Theodore Lyman's expedition of 1915. It is 'pale pinkish
buff' above and nearly white below to the roots of the hairs. Apart
from these slight variations pallescens is remarkably uniform in tint
over a wide range of territory.

Shnll. — The skull of this race is hardly to be distinguished from
that of true megalotis. To the eye, it seems a trifle narrower across
the rosti-um but the difference is not clearly brought out by measure-
ments. The upper inner incisor is normally without the lateral
cusp characteristic of macrotis.

Measurements. — No. ||-|f, Coll. Amer. Mus. Nat. Hist., from
Prescott, Arizona: forearm 43 mm. (average of four Arizona speci-
mens 42.6); digit III, metacarpal 37.3 (average of four 37.5); first
phalanx 13 (average of four 13); second phalanx 16 (average of four
17.2); tibia 18. Collector's measurements, No. 204375, Biol. Sur-
vey Coll., Graham Mts., Arizona, total length 1182 mm.; tail 50;
foot 9; extent of wings 300.

Skull: No. 204375 Biol. Survey Coll., Graham Mts., Arizona:

ALLEN: BATS OF THE GENUS CORYNORHINUS. 343

total length 16 mm.; basal length 13; palatal length 7.3; zygomatic
breadth 8.5; interorbital constriction 4; mastoid breadth 9; width
of braincase 8 ; upper tooth-row 6; lower tooth-row 6.5.

Remarks. — The subspecies pallescens is characteristic of the arid
and desert country of western United States as far north at least as
southern South Dakota. Its pallid buffy coloration recalls that of
other mammals that dwell in a dry open country. Its intergradation
with true inegalotis to the eastward seems to be very gradual, but in
the northwest as it enters the humid coastal area from western and
northern California, to southern British Columbia it merges rather
abruptly into the darker townsendii. In the southwest, it appears to
range as far east as the Pecos River in Texas and probably intergradas
in northern Chihuahua with the darker race of the Mexican plateau.
All the specimens that I have seen from Arizona and southern Cali-
fornia, however, seem referable to pallescens.

The line of intergradation with townsendii seems to follow the Sacra-
mento Valley of California back of the coast range from a short
distance to the south of San Francisco about as far north as Placer
County, where the change comes rather abruptly. The name inter-
medius H. W. Grinnell was based on specimens from iVuburn, in this
County, but to my mind there is hardly room for the recognition of an
additional race. The type is well within the range of variation of
townsendii, while other specimens from the same locality are quite as
pale as some specimens of pallescens. Further records for central and
eastern California and for the states to the eastward are to be desired.
Specimens examined. — Fifty -seven, from the following localities :
Arizona: Fort Verde, 1 (Amer. Mus. Nat. Hist.).
Graham Mts., 1 (Biol. Surv.).
Pinal Co., 1 (Amer. Mus. Nat. Hist.).
Fresco tt, 1 (Amer. Mus. Nat. Hist.).
California: Auburn, Placer Co., 2 not typical (Univ. of Calif.).
Julian, San Diego Co., 1 (Univ. of Calif.).
Kenworthy, San Diego Co., 12 (Univ. of Calif.).
Lone Pine, Inyo Co., 4 (M. C. Z.).
Los Angeles, 1 (Pub. Mus. Milwaukee).
Oro Grande, San Bernardino Co., 2 (Biol. Surv.).
Riverside Mt., Colorado River, 2 (Univ. of Calif.).
Vallecito, San Diego Co., 2 (Univ. of Calif.).
Whitewater, Riverside Co., 1 (Univ. of Calif.).
Colorado: Boulder Co., 2, not typical (Univ. of Colo.).
Wyoming: Mammoth Hot Springs, 4 (Biol. Surv.).

344 bulletin: museum of comparative zoology.

Wyoming: Sand Creek, 10 miles east of Sundance, 10 (Biol. Surv.).

South Dakota: Cheyenne River, 3 (Amer. Mus. Nat. Hist.).
Custer, 7 (Biol. Surv.).

Miller and others have recorded pallescens from the following addi-
tional localities:

Arizona: Fort Huachuca; Keam Canyon, Navajo Co. (type
locality).

California: Dulzura; Owens Lake; Owens Valley; San Diego.

Colorado: Larimer County.

Texas: East Painted Cave, near mouth of Pecos River.

Utah.

\

CORYNORHINUS MEGALOTIS TOWNSENDII (Cooper).

Townsend's Big-eared Bat.

Plecolus townsendii Cooper, Ann. Lye. nat. hist. N. Y., 1837, 4, p. 73, pi. 3, f . 6.
Plecotus macrolis Dobson, Cat. Chiroptera Brit, mus., 1878, p. 180 (not of

LeConte, 1831).
Plecotus {Corinorhinus) macrotis Trouessart, Cat. Mamm., 1897, fasc. 1, p. 105

(in part).
Corynorhinus macrotis townsendii Miller, N. Amer. fauna, 1897, no. 13, p. 53,

f. 8, a, a'; 9, a, a'.
Corynorhinus macrotis intermedius H. W. Grinnell, Univ. Calif, publ. Zool.,

1914, 12, p. 320.

Type. — None specified. The original three specimens are not
known to be still in existence.

Type Locality. — Oregon, on the lower Columbia River. Townsend,
(1839, p. 325) who collected the t>T)es, says that they frequent "the
store houses attached to the forts" hence it is probable that since The
Dalles, Fort Walla Walla, Vancouver, Fort George, and Astoria,
were the forts visited, one of these furnished the three skins he col-
lected.

Distribution. — The humid coast region from Vancouver Island,
British Columbia, southward to San Francisco, California, intergrad-
ing with pallescens here, as well as in north central California. Inland
it extends over most of (?) Washington, Oregon, and the western half
of northern California.

General Characters. — A dark-colored race, characterized by the
blackish bases of the hairs, with contrasted brown tips above, and pale
brown wash below.

ALLEN: BATS OF THE GENUS CORYNORHINUS. 345

Color. — Adults: general effect above a uniform 'warm sepia.'
The basal half of the hairs is dark slaty in strong contrast to the
terminal half or third which is nearly 'snuff brown.' Downy hairs
at the posterior bases of the ears and on their anterior rim, whitish.
Below, the rich brownish of the back passes gradually into a wash of
pale wood brown, nearly ' avellaneous ' (Ridgway, 1912). All the
hairs of the ventral surface are 'blackish plumbeous' in the basal
half.

Immature specimens (76250 Biol. Survey Coll., from Comox, B. C.)
are darker, the hairs nearly uniform blackish plumbeous, with a faint
brown tipping, above. Below, the pale tips of the hairs are more
noticeable, and the coloring is much as in the adult, though lacking
the warm brownish or russet wash.

Skull. — No tangible differences can be made out that will distin-
guish the skull of townsendii from that of pallescens. As in the latter
the second lower premolar is frequently drawn in from the axis of the
tooth-row and the inner upper incisor is normally unicuspidate. In
two specimens examined, however, the latter tooth has a distinct
shoulder or incipient cusp (204435, 76250 Biol. Survey Coll.). Com-
pared with C. macrotis the profile of the skull is more abruptly ele-
vated from the rostrum.

Measurements. — No. 9744 Field Mus. Nat. Hist, from Goldbeach,
Oregon: forearm 43 mm. (average of ten Oregon specimens 42.0);
digit III, metacarpal 38 (average of ten Oregon specimens 38.4);
first phalanx 12.8 (average of same ten 12.7); second phalanx 16
(average of same ten 16.8). Collector's measurements: total length
111mm.; tail 47; hindfootl2; ear 36.

Skull: greatest length 16.2 mm.; basal length 13.5; palatal length
7.5; zygomatic breadth 8; interorbital constriction 4; mastoid
breadth 9; width of braincase 8; upper toothrow 6.3; lower tooth-
row 7.

Remarks. — This dark brownish race is characteristic of the humid
coastal area of western North America from southern British Columbia
southward to the region of San Francisco, California. Inland from
the coast ranges and to the south and east of San Francisco the
increasing aridity causes a progressive decrease in the amount of dark
pigment so that complete intergradation by imperceptible degrees
takes place with the interior subspecies pallescens. Specimens from
intermediate localities can usually be referred to one or the other,
however, though occasional individuals are strictly intermediate. A
skin from Mt. Veeder, Napa County, just north of San Francisco, is

346 BULLZTIX: MTSEUil OF COMPARATR'E ZOOLOGY.

indistinguishable from t\-pical toicnsendii. Another from Bear Valley,
San Benito County, to the south of that place, is nearly as dark, yet a
shade paler. Through the kindness of Dr. Joseph Grinnell and Mr.
H. S. Swarth of the University of California, I have had for examina-
tion a very interesting series of nine skins from Auburn, Placer Coimty,
north central California, at the semi-arid western foot of the Sierra
Nevada. Two of these are in dark immature pelage. Two others
(7755, 19214) agree perfectly with specimens of pallescens from
Arizona or Wyoming, though the bases of the hairs are a trifle darker
than the average of that subspecies. Four, though of a richer brown
than pallescens are yet not quite so dark as topical toicnsendii of the
humid coastal area. Nevertheless they are nearer to the latter than to
pallescens. The remaining specimen is indistinguishable in any essen-
tial particular from toicnsendii of the Oregon coast. This last example
ser\'ed as the t\"pe of C. macrotis intermedius Hilda W. Grinnell.
\Yith the advantage of more abundant material and after careful
consideration, I feel unable to concur with Mrs. Grinnell in regarding
these specimens as representing a recognizable race. They are clearly
intermediate between pallescens and toicnsendii; the t}'pe can be
absolutely matched by Oregon specimens of the latter, while others
again, from the same locality, might without \-iolence be referred
to the former. In other words, a series of topot\-pes shows no char-
acters by which they may constantly be distinguished from the two
neighboring races over any considerable area. The same series was
originally referred . to toicnsendii by Dr. Joseph Grinnell, and the
si)ecLmen3 are on the whole best considered as representatives of that
subspecies, with a tendency toward the pallid form of the interior.
The same is true of specimens from Happy Camp (Siskiyou Co.)
and Bear Valley (San Benito Co.), referred by Mrs. Grinnell to
'intermedins.' A single skin (6957 Univ. of Cal., Mus. Vert. Zool.)
from Johnson's Harbor, Santa CatalLna Island, California, though
much too dark to be t^-pical of pallescens is not so dark as t\'pical
toicnsendii. Though an intermediate specimen in color, it may
for the present be considered nearer the latter. The record is of
interest in connection with the occurrence of other small land mam-
mals on this island, some of them distinct insular representatives of
continental species.

Specimens examined. — Including intermediate sjjecimens, which are
nearer toicnsendii than pallescens, twenty-three from the following
localities :

British Columbia: Comox, 1 (Biol. Sur\'.).

ALLEN: BATS OF THE GEXUS CORTNORHEN'US.

347

Oregon: Gold Beach, 6 (Field Mus. Xat. Hist.).
McKenzie Bridge, 3 (Biol. Surv.).
Vida 1 (Biol. Surv.).
California : IMt. Veeder, Napa Co., 1 (Biol. Surv.).

Bear Valley, San Benito Co., 2 (Biol. Surv.).
Happy Camp, Siskiyou Co. (Biol. Surv.).
Auburn, Placer Co., 7, not all typical (Univ. of Calif.).
Santa Catalina Id., 1, not tx-pical (Univ. of Calif.).
^Miller has also recorded it from Creswell, Oregon.

CORYNORHIXUS MEGALOTIS MEXIC.Os'US, subsp. nov.

Mexican Big-eared Bat.

Plecotus (Oorinorhinus) townsendi J. A. Allen, Bull. Amer. mus. nat. hist.,

1890, 3, p. 176 (not of Cooper, 1837 j.
Corynorhinus macrotis pallesceris Miller, X. Amer. fauna, 1897, no. 13, p. 52,

f. 10 (in part — Mexican specimens cited).

Type. — Skin and skull, 9S2S5, Biological Survey collection, adult
female, collected by E. W. Xelson and E. A. Goldman, 25 Aug., 1S99.

Type Locality. — Mexico: Chihuahua, near Pacheco.

Distribution. — The Mexican tableland, from central and western
Chihuahua, southward to Oaxaca and Santa Cruz; the precise limits
are not yet fully ascertained.

General Characters. — Smallest of the megalotis-macrotis group, the
skull small w-ith weak canines, a short and contracted rostrum with
evenly tapering lateral outlines as seen from above; color dark, the
hairs nearly tmiform drab throughout.

Color. — Adult in summer; above, a nearly uniform 'drab,' the
bases of the hairs hardly at all darker than their tips; below, the
terminal third of the hairs is soiled whitish, the bases becoming
gradually darker, nearly fuscous or 'benzo brown.' At the throat
the dark bases show through more than elsewhere. In fall, the j)elage
is longer and more silky, with slightly more contrast between the tips
and the bases of the hairs, the latter now decidedly darker, shading
into a pale 'hair brown.' The hairs of the lower surface are tipped
with a clearer whitish, washed with pinkish buff. In this pelage they
approach the coloration of t\"pical megalotis of the eastern United
States.

348 bulletin: museitm of comparative zoology.

Skull. — The small delicate skull is notable for its weak canines, the
short and contracted rostrum. In the other races of this group the
roots of the upper canines cause a distinct bulge in the outline of
the snout as viewed from above, but in these small Mexican bats the
gently convex and tapering outline is not noticeably interrupted. In
contrast with the other races of mcgalotis the inner upper incisor is
normally provided with a distinct pointed cusp at its outer side. This
cusp is wanting in but one (91930, Biol. Surv. Coll.) among eighteen
skulls from Mexico that I have examined. Mr. G. S. jNIiller, Jr.,
(1897, p. 53, fig. 10) recoi'ds variation in respect to this cusp in a series
from Guanajuato, in which both extremes of development are repre-
sented. The small size of the skull, and the short tapering rostrimi
distinguish it at a glance from macrotis.

Measurements. — The tj-pe measures: forearm 39.4 mm. (average
of ten topot\-pes 41.1); digit III, metacarpal 37 (average of ten topo-
types 37.9); first phalanx 12 (average of ten topotypes 12.1); second
phalanx 16 (average of ten topotypes 16.8); tibia 18 (average of ten
topot^-pes 18.6).

Skull: greatest length 15.6 mm.; basal length 12.5; palatal length
7; zygomatic breadth 8; interorbital constriction 3.4; mastoid
breadth 9; width of brain case 7.6; upper tooth-row 6; lower tooth-
row 6.2.

Remarks. — This small dark form from the Mexican highlands has
till now been confused with pallescens of western United States. The
series of skins at present available, however, shows that it is quite
different in color, a very smoky appearing bat, nearly uniform in tint,
with none of the buff or brown tones of pallescens or toivnsendii, nor
the white belly of macrotis. In fall and winter pelage it seems to
resemble pallescens more closely but is darker. It is odd that the
accessory cusp of the inner upper incisor, usually wanting in other
races of megalotis, should be normally present in me.vicanus. In this
respect it resembles macrotis of the southeastern United States, but
otherwise shows no near approach to that species. It is further re-
markable that the adults are indistinguishable in color from the
immature individuals, which in other races are darker than the fully
grown specimens.

I have provisionally considered all records of Corynorhinus from
south of Chihuahua as referring to the present race, but the possibility
that those recorded from Oaxaca or Vera Cruz may be still different
is not to be overlooked. Through the kindness of Mr. C. D. Bunker
of ihd University of Kansas Museum, I have before me a single speci-

ALLEN: BATS OF THE GENUS CORYNORHINUS. 349

men said to have been collected by George F. Gaumer on the Island
of Cozumel, Yucatan. The skin had no original label and is appar-
ently quite typical of pallesceiis so that it is more than likely the lo-
cality is erroneous.

For the privilege of describing this new form I am indebted to Mr.
H. W. Henshaw, Chief of the Biological Survey.

Specimens examined. — Total number eighteen, from the following
localities : —

Chihuahua: near Pacheco, 14 (Biol. Surv.).

Zacatecas: Valparaiso Mts., 2 (Biol. Surv.).

Jalisco: Guadalajara, 1 (Amer. Mus. Nat. Hist.).

Guanajuato: Sta. Rosa, 1 skull (Biol. Surv.).

The following additional locality records probably refer to this
subspecies: San Luis Potosi, Hacienda la Parada; Michoacan,
Patzcuaro; Oaxaca, Oaxaca; Vera Cruz, Jico (see Miller, 1897, p. 53).

CORYNORHINUS MACROTIS (LcConte).

LeConte's Big-eared Bat.

Plecotus viacrotis LeConte, Cuvier's Animal kingdom, ed. McMurtie, 1831,

1, appendix, p. 431.
Plecotus leconlii Cooper, Ann. Lye. nat. hist. N. Y., 1837, 4, p. 72, pi. 3, f. 5.
Sijnotus leconlii Wagner, Schreber's Saugethiere, suppl., 1855, 5, p. 720.
Vespertilio macrotis LeConte, Proc. Acad. nat. sci. Phila., 1855, p. 436

("Georgia").
Synotus macrotis H. Allen, Smithsonian misc. coll., 1864, 7, no. 165, p. 63.
Corynorhinus macrotis H. Allen, Proc. Acad. nat. sci. Phila., 1865, p. 174.
Plecotus {Corinorhinus) macroiis Trouessart, Cat. Mamm., 1897, fasc. 1, p. 105.

Txjpe. — None specified. A "dry" specimen, 4727 U. S. N. M.,
presented by Major LeConte, with locality entered as "United
States," is listed by Harrison Allen in his Monograph of the Bats of
North America, 1893. This specimen may well be the actual one on
which LeConte based his description.

Type Locality. — No locality is mentioned in the original description.
In a later paper, however, LeConte states (1855, p. 436) that it
inhabits "Georgia," whence it is inferred that the type locality is
" probably near the LeConte Plantation, 5 miles south of Riceboro,"
Liberty County, in that State (INIiller, 1897, p. 51).

Distribution. — Southeastern Lnited States, from North Carolina,

350 bulletin: museum of comparative zoology.

Georgia and (? northern) Florida, westward through the Southern
and Gulf States, into Louisiana, and probably eastern Texas.

General Characters. — Distinguished from megaJotis and its sub-
species by the clear white tips to the belly hairs, and the contrasted
brown tips and blackish bases of the hair of the back; inner upper
incisor bicuspidate.

Color. — Adults : the basal two thirds of the hairs of the dorsal
surfaces is ' plumbeous black,' the terminal third a uniform cinnamon-
brown, practically ' sayal brown, ' in sharp contrast ; beneath, the hairs
are 'plumbeous black' basally, their tips clear white, again in sharp
contrast. Specimens in thin pelage show much less of the cinnamon-
brown above and the plumbeous bases of the hairs give a predominat-
ing dark appearance with a streaking of cinnamon; below, the white
tipped hairs are less numerous particularly on throat and chest. The
region at the posterior base of the ears is usually dark like the rest of
the back, but in one specimen is whitish.

Immature individuals, though similar to adults in the contrasting
dark bases and white tips of the hairs of the ventral surfaces, are much
darker above, owing in part to the thinness of the pelage which allows
the plumbeous bases of the hairs to show through, and in part to the
paler (almost buffy) tipping of the contour hairs.

Skull. — Though essentially similar, the skull of macrotis differs
from that of true viegalotis and its race pallescens in being slightly
smaller, and with a flatter profile. The intermaxillary notch is in
general a trifle more contracted in dorsal view. The inner upper
incisor of viacrotis seems invariably to bear a small cusp on its exterior
side, whereas in megaloiis and m. pallescens this cusp is normally
wanting, though in two out of twenty-eight specimens it was indicated,
and in the race mexicanus is usually present.

Measurements. — No. 159413, Biol. Survey Coll., from Young Harris,
Union County, Georgia: forearm 43.5 mm. (average of nine speci-
mens 41.7); digit III, metacarpal 39.6 (average of nine 37.7); first
phalanx 13.6 (average of nine 12.9); second phalanx IS (average of
nine 16.6); tibia 21.

Skull: greatest length 16.6 mm.; basal length 13.2; palatal length
7.3 ; zygomatic breadth — ; interorbital constriction 4 ; mastoid
breadth 9.5; width of braincase 8.5; upper tooth-row 6; lower tooth-
row 7.

Remarks. — In its style of coloraton this bat differs notably from
megalotis and its races, though in structure it is very similar. Its
smaller skull, and the constantly bifid inner upper incisor distinguish

ALLEN: BATS OF THE GENUS CORYNORHINUS. 351

it further from any of the forms occurring in the United States. I
have found no evidence that it intergrades at any point with true
megalofis or with m. paUescens. In North Carolina, a typical specimen
in the collection of Mr. Morton L. Church, was captured at Marshall,
in the extreme western end of the State, while from extreme western
Virginia comes megalotis, without any sign of intergradation, though
the localities are not far distant on opposite sides of the AUeghenies.
Since writing the above, I have also examined a skin of macrotis from
Mitchell, Ind., which is of interest as indicating not only the north-
ward limit of the species' range in east central United States, but also
that it keeps distinct from megalotis where the two occur together.
Alcoholic specimens, if in good condition, show the white-tipped hair
with its dark bases on the belly, and can usually be distinguished by
this character. How far to the westward this bat ranges is as yet
unknown. It is found in Louisiana and northward into Arkansas
and Indiana but has not yet been discovered in eastern Texas al-
though paUescens is recorded from western Texas. If this apparent
hiatus shall prove to be real, it would indicate that the range of
macrotis is fairly distinct from that of megalotis and its races. The
present evidence therefore shows that macrotis constitutes a species
distinct from the latter, though closely allied and of similar struc-
ture. It is characteristic of the Lower Austral life zone.

Specimens examined. — The following specimens have been studied,
a total of nineteen.

North Carolina: Marshall, 1 (M. L. Church Coll.); ten miles
northwest of Taylorsville, 2 (N. C. Coll. Agric).
South Carolina: Society Hill, 2 (U. S. N. M.).
No locality, 2 (U. S. N. M.).
Georgia: Kesler, Early Co., 1 (M. C. Z.).

Young Harris, Union Co., 1 (Biol. Surv.).
? Riceboro, Liberty Co., 2 skulls (U. S. N. M.). These
specimens were collected by LeConte, and though with-
out record of locality, may have come from his planta-
tion.
Alabama: Huntsville, 1 (Biol. Surv.).

Leighton, 1 (Biol. Surv.).
Louisiana: Houma, 4 (Biol. Surv.).
Arkansas: Osage River, 1 (M. C. Z.).
Indiana: Mitchell, 1 (Ind. Univ.).

In addition, it has been recorded from
Virginia: Dismal Swamp.

352 bulletin: museum of comparative zoology.

North Carolina: Bertie County; Goldsboro; Weaverville (Brim-
ley, 1905, p. 22) ; Pisgah Forest, 3300 ft. (Ober-
holser, 1905, p. 9).

South Carolina: Hardeeville.

Florida: Micanopy.

Alabama: Greensboro.

Mississippi: Bay St. Louis.

CORYNORHINUS PHYLLOTIS, sp. nov.

Leaf-eared Bat.
Plecotus ouritus J. A. Allen, Bull. M. C. Z., 1881, 8, p. 184 (not of Linne, 1758).

Tijpe.— Skin and skull 5943, M. C. Z., collected by Dr. Edward
Palmer, 24 March, 1878.

Type Locality. — Mexico: San Luis Potosi (probably near the city
of the same name).

Distribution. — At present known from the type locality only.

General Characters. — Ears larger than in megalotis, the transverse
ribs on the middle third of the outer edge subdividing and extending
quite to the border; skull larger, the braincase inflated, and a distinct
lachrymal ridge present; tips of hairs in sharp contrast with the dark
bases; upper surface tawny olive; calcaneum with a well-developed
keel.

Color. — Hair above long and silky; basal half dark, ' fuscous black,'
the tips pale ' tawny olive ' ; a band of downy hair at the posterior base
of the ears whitish. Below, the basal two thirds of the hairs is 'fus-
cous black,' the tips white, washed with 'pale ochraceous buff.' The
distinct olive tone above is in marked contrast with the buffy or dark
fur of the viacrotis and megalotis coloration.

Skull. — This species is at once distinguished from the other species
of the genus by its larger and differently shaped skull (Plate 1, fig. 6).

The braincase is flattened and broad, the rostrum broader and
sharply depressed, with a more marked excavation medially ; there is a
distinct lachrymal ridge, as in Plecotus; the audital bullae are also
larger. The teeth differ mainly in their stouter proportions, but the
two upper incisors instead of being nearly side by side are one behind
the other in the line of the tooth-row; there is also a greater disparity
in size between the two anterior lower premolars, the first of which is
much larger, in lateral view, than the second.

ALLEN: BATS OF THE GENUS CORYNORHINUS. 353

Measurements. — The type measures: forearm 44 mm.; digit III,
metacarpal 42.4; first phalanx 14; second phalanx 16.3; digit IV,
metacarpal 41.6; first phalanx 11.5; second phalanx 13.5; digit V,
metacarpal 42; first phalanx 11; second phalanx 7; tibia 17; hind
foot 10; ears (dry) from meatus 31 ; greatest breadth 21.

Skull: greatest length 17.5 mm.; basal length 14.9; palatal length
8.5; zygomatic breadth 10; interorbital constriction 4.8; mastoid
breadth 10; width of braincase 9.6; upper tooth-row 7; lower tooth-
row 7.1.

Remarks. — Though the type specimen was recorded thirty-five
years ago, as Plecotus auritus, this remarkable species has remained
till now unknown. The peculiar olive tone to the fur of the back is
very different from the buffy or brown of the other known species.
The very large and stiff ears are much like those of Antrozous, except
for the breaking up of the transverse ribs at the middle of the outer
margin. In this latter respect the ears recall those of Plecotus, in
which also, the ribs run quite to the margin of the ear, instead of to a
line parallel with the rim, as in other species of Corynorhinus. In
Plecotus, however, the number of these ribs is about double that in
C. phyllotis. The latter further resembles Plecotus in the possession
of distinct lachrymal ridges on the skull, so that this character can no
longer be considered of generic value in distinguishing between the two.
A long narrow keel on the calcaneum is likewise diagnostic of this new
bat, for in Plecotus as in other forms of Corynorhinus the calcaneum
is without keel. In the squarish outline of the nostrils and the devel-
opment of the excrescences on the muzzle it is typical of its genus.
The peculiar bulging of the anterior part of the braincase iS not seen
in other members of this group. The tibia is proportionally shorter
than in the smaller C. megalotis mexicanus which inhabits the same
region.

Specimen examined. — The type.

354 bulletin: museum of comparative zoology.

REFERENCES.

Allen, Harrison.

1864. Monograph of the bats of North America. Smithson. misc. coll.,
7, no. 165, xxiii + 85 pp., 68 text-figs.

(Synotus [= Corynorhinus] treated on p. 62-66).

1865. On a new genus of Vespertilionidae. Proc. Acad. nat. sci.
Phila., p. 173-175.

(The genus Corynorhinus defined, p. 173).
1 893 . A monograph of the bats of North America . Bull . 43 U . S . N . M . ,
X + 198 pp., 38 pis.

(Corynorhinus, p. 53-60, pi. 6, 7).
Allen, J. A.

1881. List of mammals collected by Dr. Edward Palmer in northeastern
Mexico, with field-notes by the collector. Bull. M. C. Z., 8, p. 183-189.
(The type of C. phyllotis recorded as Plecotus auritus, p. 184.)
1890. Notes on collections of mammals made in central and southern
Mexico, by Dr. Audley C. Buller, with descriptions of new species
of the genera Vespertilio, Sciurus, and Lepus. Bull. Amer. mus.
nat. hist., 3, p. 175-194.

(A Corynorhinus recorded from Guadalajara, p. 176).
1895. List of mammals collected in the Black Hills region of South
Dakota and in western Kansas by Mr. Walter W. Granger, with field
notes by the collector. Bull. Amer. mus. nat. hist., 7, p. 259-274.
(Corynorhinus from Cheyenne River referred to townsendii, p. 272).
Andersen, Knud.

1912. Catalogue of the Chiroptera in the collection of the British Mu-
seum. Second edition. 1. Megachiroptera. London, 8vo, cii +
854 pp., 79 text-figs.
Brimley, C. S.

1905. A descriptive catalogue of the mammals of North Carolina, exclu-
sive of the Cetacea. Journ. Elisha Mitchell sci. soc, 21, p. 1-32.
(Records of C. macrotis, p. 22).
Butler, A. W.

1895. The mammals of Indiana. Proc. Indiana acad. sci. for 1894,
p. 81-86.

(Two Corynorhinus recorded from Greencastle, Ind., 23 Dec. 1894,
p. 86).
Clapton, J.

1722. A voyage to Virginia; and an account of that country. Phil,
trans. Roy. soc. London, Abridgement, 3, p. 575-600.

ALLEN: BATS OF THE GENUS CORYNORHINUS. 355

Cooper, William.

1837. On two species of Plecotus inhabiting the United States Territory.
Ann. Lye. nat. hist. N. Y., 4, p. 71-75, pi. 3, figs. 5, 6.

(Renames Plecotus [■= Corynorhinus] macrotis and describes town-
sendii) .
Cory, C. B.

1912. The mammals of lUinois and Wisconsin. Field mus. nat. hist.,
Zool. ser., 11, 505 pp., illustr.

(Corynorhinus, p. 476).
Grinnell, Hilda W.

1914. Three new races of vespertilionid bats from Cahfornia. Univ.
Cahf. publ. Zool., 12, p. 317-320.

(Describes C. macrotis intermedius, subsp. nov., p. 320).
Grinnell, Joseph.

1914. An account of the mammals and birds of the lower Colorado Valley
with especial reference to the distributional problems presented. Univ.
Calif, publ. Zool., 12, p. 51-294, pi. 3-13, 9 text-figs.
(Records C. m. pallescens, p. 263).
Grinnell, Joseph, and Swarth, H. S.

1913. An account of the birds and mammals of the San Jacinto area of
southern California with remarks upon the behavior of geographic
races on the margins of their habitats. Univ. Cahf. publ. Zool., 10,
p. 197-406, pi. 6-10, 3 text-figs.

(Notes on C. m. pallescens, p. 379).
Hahn, W. L.

1909. The mammals of Indiana. A descriptive catalogue of the mam-
mals occurring in Indiana in recent times. 33d Ann. rept. Ind. dept.
geol. and nat. resources, for 1908, p. 417-654, 659-663, 33 text-figs.
(Corynorhinus recorded from Indiana, p. 619).
LeConte, John.

1831. Appendix of the American Editor, in the Animal Kingdom
arranged in conformity with its organization, by the Baron Cuvier.
M'Murtrie's edition, 1, p. 431-448.

(Plecotus [= Corynorhinus] macrotis described, p. 431).
1855. Observations on the North American species of bats. Proc.
Acad. nat. sci. Phila., p. 431-438.

(Describes V. [= C] macrotis in detail, p. 436).
Lesson, R.

1827. Manuel de mammalogie, ou histoire naturelle des mammif^res.
Paris, xvi + 442 pp.

(Plecotus rafinesquii substituted for Vespertilio megalotis, p. 96).
McAtee, W. L.

1907. A hst of the mammals, reptiles and batrachians of Monroe County,
Indiana. Proc. Biol. soc. Washington, 20, p. 1-16.

(Corynorhinus recorded from Lawrence Co., Ind., p. 8).

356 bulletin: museum of comparative zoology.

Miller, G. S., Jr.

1897. Revision of the North American bats of the family Vespertihoni-
dae. N. Amer. fauna, no. 13, 140 pp., 3 pis., 40 text-figs.
(Genus Corynorhinus reviewed, p. 49-54, figs. 76-10).
1907. The families and genera of bats. Bull. 57, U. S. N. M., xviii +
282 pp., 14 pis., 49 text-figs.
Oberholser, H. C.

1905. Notes on the mammals and summer birds of western North Caro-
lina. Biltmore, N. C, 24 pp.

(C. macrolis from Pisgah Forest, p. 9).
Rafinesque, C. S.

1818. Further discoveries in natural history, made during a journey
through the western region of the United States. Amer. monthly mag.
and crit. review, 3, p. 445-447.

(Describes V[espertilio\ megalotis, p. 446).
Rhoads, S. N.

1903. Themammalsof Pennsylvania and New Jersey, etc. Philadelphia,
266 pp., 9 pis., map.

(Doubts validity of Meadville, Pa., record of Corynorhinus, p. 226).
Ridgway, Robert.

1912. Color standards and color nomenclature. Wasliington, iv + 43
pp., 53 pis.
Stephens, Frank.

1906. California mammals. San Diego, 351 pp., illustr.
(Notes on C. m. pallescens, p. 265).

Townsend, J. K.

1839. Narrative of a journey across the Rocky Mountains, to the
Columbia River, and a visit to the Sandwich Islands, Chili, &c. with a
scientific appendix. Philadelphia, 352 pp.

(Cooper's description of Plecotus townsendii reproduced, with added
notes, p. 324).

EXPLANATION OF THE PLATE.

Allen. — Bats of the Genus Corynorhinus.

EXPLANATION OF THE PLATE.

Fig. L — Corynorhinus megalotis toxvnsendii. 150273 Biol. Surv., upper in-
cisors and canine of right side, with supernumerary incisor (i').

Fig. 2. — C. m. townsendii. 204437 Biol. Surv. from McKenzie Bridge,
Oregon, showing normal incisors (i^, i"^) and canine.

Fig. 3. — C. VI. mexicanus. 98285 Biol. Surv. Type, to show bicuspidate
first incisor and weak canine.

Fig. 4. — Ear of C. megalotis. 170933 Biol. Surv. from Sun City, Kansas.

Fig. 5. — Ear of C. phyllotis. 5943 M. C. Z. Type, from San Luis Potosi,
Mexico.

Fig. 6.— Skull outline of C. phyllotis. 5943 M. C. Z. Type, from above.

Fig. 7. — Skull outline of C. megalotis mexicanus. 98285 Biol. Surv. Type.

BULL. MUS. COMP. ZOOL.

CORYNORHINUS.

4

Bulletin of the Museum of Comparative Zoblogy

AT HARVARD COLLEGE.

Vol. LX. No. 10.

THE RESIDENT BIRDS OF GUADELOUPE.

By G. K. Noble.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

August, 1916.

No. 10.— The Resident Birds of Guadeloupe.

By G. K. Noble.

Introduction.

During the summer and early fall of 1914 I visited, in the interest
of the Museum of Comparative Zoology, Guadeloupe in the French
West Indies.

On the way and for the first few days while on Guadeloupe I was
fortunate in having the company of Mr. F. R. Wulsin. We had
several opportunities to make short ornithological excursions on some
of the islands to the north of Guadeloupe. On the homeward voyage
the ship was detained at St. Croix and several days were spent there
in the field.

I remained on Guadeloupe from June 22nd to September 12th.
In order to cover all the main regions of Guadeloupe and Grande
Terre I spent four days or more in each of the following localities: —
Pointe a Pitre, Goyave, Basse Terre, Ste. Claude, the Soufriere, Vieux
Habitants, Cluny, Ste. Rose, Soffire, and St. Francois. My interest
was chiefly in the land-birds and I was fortunate in finding all of the
existing resident species.

Guadeloupe possesses a local museum, the Musee L'Herminier,
in the town of Pointe a Pitre. Dr. F. L'Herminier was undoubtedly
the greatest naturalist who ever lived on Guadeloupe; a student of
the famous Blainville, he applied the best methods of his time to the
study of Antillean bird life. Unfortunately in the great Pointe a Pitre
fire of 1844, all his work on Guadeloupe birds was swept away. Lafres-
naye (Rev. zool., 1844, p. 168) in describing some of L'Herminier's
specimens comments briefly on this disaster: —

"Quoique nous eussions deja ces trois especes, cjue nous devions a
I'obligeance de M. L'Herminier, avant I'affreuse catastrophe de la
Pointe-a-Pitre, nous ne les avions pas publiees parce qu'il nous avait
confie son projet de publier une Faune ornithologique de la Guade-
loupe. Depuis lors, il nous y a autorise, tout en nous apprenant que
loin de renoncer a son projet, et malgre la perte immense pour lui de
tons ses oiseaux, soit montes, soit dans I'alcool, de tous ses livres et de
toutes ses notes, il s'occupe avec une nouvelle ardeur de le mettre a
execution."

360 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY.

A short time after the fire, the Musee L'Herminier was built. In
1876 Ober visited the island and remarked, (cf. LawTence, Proc.
U. S.N. M., 1878, 1, p. 452):—

" Few birds are, as yet, in the museum but there are very excellent
and complete collections of Crustacea, etc., and many fine specimens
of aboriginal implements."

I found the bird collection had so increased that it was larger than
all of the other collections. Three well-mounted specimens of Aestre-
lata diabolica were included in it. Unfortunately most of the speci-
mens were without data or records of their receipt.

I have omitted all reference to the birds noted at St. Croix except
when they have direct bearing on the Guadeloupean birds. The
West Indian avian fauna is rapidly being destroyed and for that
reason it seems advisable to include here a complete record of the
birds collected rather than to give simply an account of the more
important studies, such as those relating to Aestrclata haesitaia, Cich-
Iherminia herminieri, and Cocrcba dominicana. Clark (Proc. Bost.
soc. nat. hist., 1905, 32, p. 203-312) has already summarized the
general conditions of bird distribution in the Lesser Antilles.

Topography.

Guadeloupe consists physiographically of two distinct parts: —
Guadeloupe proper, a rugged mass of volcanic formation and
Grande Terre, a flat limestone island separated from the former
by a sluggish tidal water-way, La Riviere Salee. The adjacent
islands of Desirade and Marie Galante are similar in structure to
Grande Terre while the small archipelago called Les Saintes is vol-
canic like Guadeloupe proper. Guadeloupe and Grande Terre taken
together are about forty miles in their greatest length and the same
in their greatest width.

The whole surface of Guadeloupe is broken up into peaks and is
cut by deep valleys making walking for any distance very difficult.
Many of the hills reach to four thousand feet or over. The Soufriere
for instance, is 4863 feet high. Its summit is wreathed with rain
clouds throughout most of the year. The slopes on the mountainous
core of this island exhibit many contrasts. There are at least four
distinct life-zones: —

(1) Lowland savannahs. Great areas of grass- and scrub-land skirt

noble: the resident birds of Guadeloupe. 361

the coast of Guadeloupe and extend over the whole of Grande Terre.
These regions are now for the most part under cane cultivation. A
few swift flowing streams cross the plantations, but only the Grande
Riviere in the north is navigable for more than a few hundred yards.
The monotony of the grasslands is broken on the east coast by small
stands of timber and on the west coast by broad outcrops of volcanic
rocks. Marshes are rare in this belt and ponds even more so. On
Grande Terre the soil is sandy and calcareous and the absence of
water makes a hot and dusty landscape.

(2) Uplands of hardwood forest. The savannahs merge gradually
into the uplands which are generally coAered with medium sized
deciduous trees. This belt varies in width from one to four miles and
in several places such as at Ste. Marie, Ste. Rose, and Trois Rivieres
it encroaches through the grassland area and extends to the ocean.

(3) Rain forest. The dense tropical forest starts abruptly at about
1500 ft. and covers the greater part of the central region of Guadeloupe
proper. It is almost impenetrable. While its bird fauna is character-
istic, the number of species is small.

(4) Mountain barrens. Lastly the wind swept country above
the tree line begins at about 4000 feet. It covers relatively a very
small proportion of the island and is almost devoid of life.

' The Vertebrate Fauna.

The vertebrates are represented in Guadeloupe by a very small
number of species. An agouti (Dasyproda nohlei G. M. Allen),
a racoon (Procyon minor Miller) and four or five species of bats com-
prise the entire indigenous mammalian fauna. The introduced
Mongoose {Mungos hirmanicus (Thomas)) is everywhere abundant
and it is due to its depredations that the land vertebrates are so
rapidly disappearing. The Anolis ferreus Cope is the commonest
reptile. Iguana delicatissima Laurenti, and the Ground Lizard
(Ameiva cincracea Barbour & Noble) are found today only on the small
islands off the coast. A skink (Mabuya maboia Dumeril & Bibron)
may likewise have been completely extirpated from the mainland of
Guadeloupe by the Mongoose. Two geckos, Sphaerodactylus fan-
tasticus (Dumeril & Bibron) and Thccodactylus rapicaudus (Houttuyn)
because of their secretive habits are not commonly found, but they
may be locally abundant. The two snakes, Typhlops lumbricalis

362 bulletin: museum of comparative zoology.

(Linne) and Alsophis leucomelas (Dumeril & Bibron) formerly occur-
ring on the island are now both extinct, but a species of Couleuvre
(Alsophis sanctorum Barbour) closely related to the latter is peculiar
to Les Saintes and is still found abundantly because the Mon-
goose has not been introduced into these islands. The Grenouille
(EleutJierodactylus martimccnsis Tschudi) and the introduced Crapaud
{Bujo marinus Linne) are locally numerous on Guadeloupe. As in
other Lesser Antilles the fresh-water fishes are few. The number of
species of resident birds is noteworthy because the list is so short.
Migrants are also few but deserve special mention.

Present Status of Bird Life.

There are thi^ee classes of birds which are not fully considered under
the annotated list and which I shall discuss here: (1) extinct species,
(2) probable resident birds, mostl3^ water-birds, not collected during
the trip, and (3) migratory species.

The parrots (Ara guadehupensis, Aiiodorhynchiis purpurascens, Ama-
zona violaccus, and Connrus labati) were the earliest Guadeloupe land-
birds to be completely exterminated. The old French accounts show
that the natives killed them in numbers, but even so it is hardly pos-
sible that their annihilation was due wholly to human agency. Fidica
caribaea, and Ralhis crepitans have probably been extirpated by the
Mongoose, at least they have not been taken by the native chasseurs
for a long time. I saw at Cluny, on several occasions, a Mongoose
far out in the middle of the swamp jumping from log to log in its eager
hunting. The Diablotins, or Black-capped Petrels, have been extir-
pated from Guadeloupe, probably tlii'ough several causes.

There are several birds not noted during my stay which may nest
on Guadeloupe. The Black Hawk (Urubitinga anthracina) and the
Chicken Hawk (Buteo antillarum) have both been observed by Pointe
a Pitre sportsmen but it is doubtful if either of these were more than
stragglers. I was informed that the Redstart (Setophaga ruticilla)
occurred throughout the year but my own observations did not
confirm this rather improbable statement. If the fishermen may be
relied upon, the following sea-birds breed on the outlying islets of Les
Saintes and Tete Anglais : — Phaethon acthercus, Sida leucogastra,
Sula piscator, Sterna maxima, and Anous stolid us. The Yellow-
crowned Night Heron (Nyctanassa violacea) breeds regularly just east
of Ste. Rose.

noble: the resident birds of Guadeloupe. 363

There are many migrants which are either erratic wanderers or
reguhir visitants and which pass over Guadeloupe, but the information
regarding them is usually quite unreliable. The list of birds observed
by L'Herminier (Proc. U. S. N. M., 1879, 1, p. 450-451), between 1827
and 1844, is very large, and I doubt if some of the species recorded
have been found on Guadeloupe in recent times. Winch (see Cory,
Auk, 1891, 8, p. 48-49) and Ober (see Lawrence, Proc. U. S. N. M.,
1879, 1, p. 452-462) both collected several of the migrants. Their
lists, plus my own observations, include the following species which
though not mentioned in the annotated list are nevertheless probably
of regular occurrence: — Sterna antillarum, Ereunetes jnisillus, Pisobia
minutiUa, P. maculata, Actitis macularia, Helodromas solitarius,
Aegialifis semipahnata, Ceryle alcyon, Seiurus novcboracensis,
Wilsonia canadensis, Sctophaga ndiciUa. Probably other species,
chiefly herons and sea-birds visit the island as stragglers. I saw on
July 22nd in a pond near Cluny a duck which I believe was Dendro-
cygna discolor. Many other species of ducks visit the island on migra-
tion. Mr. Delphin Duchamp, a prominent planter, informed me
that the Ani {Crotophaga ani) and the White-crowned Pigeon {C'olumba
leucocephala) have been occasionally seen on Guadeloupe after a
hurricane. Other species very probably reach the island under similar
circumstances.

Acknowledgements.

It gives me pleasure to acknowledge with many thanks the kind
assistance of His Excellency, Monsieur Lauret, the Governor of
Guadeloupe. Monsieur C. Thionville of his staff was also especially
considerate. I am indebted to Monsieur D. Duchamp and Monsieur
Riese for many favors while visiting their plantations.

During the preparation of this paper I have received assistance
from Messrs. W. De W. Miller, John T. Nichols, and Charles H.
Rogers of the American Museum of Natural History, who have exam-
ined specimens in that museum at my request ; to the officers of the
United States National Museum, the Philadelphia Academy of Natu-
ral Sciences, and the Field Museum of Natural History I am also
indebted for their help in straightening out the status of Tiaris bicolor
omissa; and I take great pleasure in thanking Mr. Outram Bangs and
Dr. Thomas Barbour for explaining many technicalities of descrip-
tion and literature.

364 bulletin: museum of comparative zoology.

Annotated List of Species.

1. PODILYMBUS PODICEPS ANTILLARUM Bangs.

Chien d'eau.

One adult male from Cluny, near Ste. Rose, July 25th, and two
young birds in the down, from the same locality, July 20th and 24th.

The series of Antillean Pied-billed Grebes in the Museum of Com-
parative Zoology is much too small to determine satisfactorily the
status of this race. I do not believe the evidence brought forward
by Todd (Ann. Carnegie mus., 1916, 10, p. 170) is sufficient to consider
that the Pied-billed Grebe breeds in the Antilles the same as the typical
mainland form. Wetmore (Bull. 326, U. S. dept. agr., 1916, p. 17)
has recently recognized PodUymhxis podiccps antillarum as a valid
race and I am inclined to regard it such until sufficient material has
been brought together to allow an intensive study.

The Pied-billed Grebe still breeds on Guadeloupe although nearly
extirpated by the Mongoose. Like the gallinules and rails, this
species, formerly common, is now very rare. I found it breeding in
only one locality. Grand Etang, Cluny. Several sportsmen of Pointe
a Pitre told me that it only occurs in those lakes where there is an
island to protect it from the stealthy approach of the Mongoose.

2. Frigata magnificens Mathews.
Mansfeny.

One adult male from Ste. Rose, July 21st.

The only Frigate-bird I succeeded in shooting was flying over a
fresh-water pond and diving at intervals for fish. I saw many others
off the coast. The fishermen told me that the Mansfeny breeds regu-
larly on Tete x\nglais.

3. Butorides virescens maculatus (Boddaert).

Qui-o. Crabier Vert tachete.

One adult from Ste. Rose, July 20th, seven adult and half-grown
specimens from Goyave, taken during the end of August; and six

noble: the resident birds of Guadeloupe. 365

adult and half-grown birds from Les Saintes taken during the first
week in September.

Oberholser (Proc. U. S. N. M., 1912, 42, p. 529-577) has recently
revised the subspecies of Biitorides virescens and has described several
new Antillean races. He refers the Guadeloupe bird to B. v. cubamis
(Oberholser, Loc. cit., p. 559-561) and erects another new race, B. v.
christophorensis between the Guadeloupe bird and the northern race,
B. V. cuhamis.

I have compared a large series of specimens from nearly every
island in the Antilles with the eighteen specimens taken on Guadeloupe
and St. Croix and am convinced that the green herons from Cuba
to Grenada all belong to one subspecies. Messrs. Bangs and Barbour
have studied this series with me and have called my attention to the
fact that the series from Guadeloupe includes within its range of
variation, in color and measurements, the distinguishing characters
of four of Oberholser's new subspecies: — namely, B. v. christophoren-
sis, B. V. dominicanus, B. v. lucianus, and B. v. grenadcnsis. I saw
in early July a green heron flying a considerable distance off the north
shore of Guadeloupe and from the bird's position high in the air, both
Antigua and Dominica must have been plainly visible. It seems very
probable that green herons wander from island to island throughout
the Greater and Lesser Antilles.

The Antillean race of green heron feeds in both fresh- and salt-
water swamps but for some reason it is only locally abundant through-
out Guadeloupe and Grande Terre. In general its habits are like
those of our own Green Heron {B. v. virescens) except that the Antil-
lean race has adapted itself to lizard hunting. On both St. Croix
and Guadeloupe I have observed the green heron standing motion-
less in the center of a dry field watching for an Anolis. It is surpris-
ing to see this bird of the twilight and the swamps apparently dozing
in the middle of an open field while the tropical sun glares down from
directly overhead.

4. Cerchneis sparveria caribbaearum (Gmelin).

Gli-Gli.

Seven adults from various localities on Guadeloupe and Les Saintes :
two from Ste. Rose July 13th, one from Ste. Claude June 26th, one
from Goyave September 6th and three from Les Saintes September
15th.

366 bulletin: museum of comparative zoology.

The Sparrow Hawk is a common bird throughout the whole of the
lowlands of both Guadeloupe and Grande Terre. It is the only
resident hawk although several others occur as stragglers. In habits
it resembles Cerchneis sparverius sparvcrius but seems more sluggish
in flight. During the heat of the day it glides almost wearily over the
plantations or dozes on some palmiste in the full glare of the sun.

5. loNORNis MARTiNicus (Linne).
Poule d'eau a Cachet Vert.

One adult purchased at La Moule.

This specimen was taken sometime during 1913. It was the last
Purple Gallinule that had been seen on Grande Terre. As with the
rails, this species has probably disappeared from Guadeloupe be-
cause of the Mongoose.

6. Gallinula chloropus cachinnans Bangs.
Poule d'eau a Cachet Rouge.

Two adults, a male and a female from Grand Etang, Cluny, near
Ste. Rose taken July 24th.

As already mentioned the gallinules are almost extinct in Guade-
loupe. I do not believe there were more than three pair breeding in
the Grand Etang. Bangs (Proc. N. E. zool. club, 1915, 5, p. 93-99)
has recently' revised the American forms of Gallimila chloropus and
has referred the Guadeloupe bird to the northern race. The measure-
ments of the two specimens noted above are included in his revision.

7. AcTiTis macularia (Linne).

L'Alouette de Mer.

One adult, a female, from Ste. Rose, July 15th.
On the north shore of Guadeloupe near Ste. Rose I found the
Spotted Sandpiper the first week in July. It is very likely that the

noble: the resident birds of Guadeloupe. 3G7

species is a resident. The bird seeks out the lowland streams which
are not too overgrown with foliage. In such regions it is common.

8. Larus atricilla atricilla Linne.
Mauve a Tete Noir.

A few birds observed near Goyave, the latter part of August, but
no specimens taken until September 16th at the island of St. Croix.

A comparison of a fair series of Laughing Gulls from the Lesser
Antilles, Greater Antilles, and Bahamas with a large series from the
mainland of North America shows conclusively that the island birds
are smaller than the mainland ones. No color difference is apparent
but the decided difference in size warrants the referring of the Laugh-
ing Gulls inhabiting the coast of North America, to a distinct race^
which may be known as Lams atricilla megahpterus (Bruch).

Although there is some uncertainty as to the priority of this name,
it seems most probable that Bruch (Journ. f. ornith., 1855, p. 287)
was the first to describe the North American Laughing Gull. Linne's
description was based upon Catesby (Nat. hist. Carol., 1733, 1, p. 89)
and Catesby only mentions the Bahaman bird. It seems probable
that Bruch when describing .4. catesby i and A. micropterus was dis-
cussing the Bahaman bird. To be sure Bruch quotes Bonaparte as
the authority for these names but when Bonaparte a year later (Comp.
rend. Acad, sci., 1856, 42, p. 771) published upon the birds he gave
Bruch as the authority. In setting up Atricilla as a generic name
Bruch was compelled by the custom of the time to use a new name
for Larus atricilla Linne, to avoid duplicating names. Atricilla
cateshyi refers, tJien, to the bird of Linne, in other words to the Ba-
haman-Antillean race. The next bird described by Bruch was A.
megahpterus and although the description (Bruch, Loc. cit., p. 287) is
not good, he gives the type-locality as "Peru und der Mexicanische
Meerbusen." It is at least conservative to refer this name to the
North American race. The name A. micropterus cannot refer to this
race because Bruch {Ibid, p. 287) gave it to a species smaller than
A. catesbyi, in other words smaller than the Bahaman- Antillean race.

The difference in size between L. a. atricilla and L. a. megahpterus
is shown in the following table.

3G8

bulletin: museum of comparative zoology.

Larus africilla atricilla.

■

Exposed

M.C.Z.

Sex

Locality

Wing

Tail

Culmen

Tarsus

14704

0^

Antigua

302

121

41

45

14699

d"

u

303

123

39

43

14702

d"

«

294

122

38.5

42 ■

112927

d

Union Isl.

301

116

37

41

112926

&

(1 u

295

118

38

40

66611

d

St. Croix

311

122

41.5

43

40181

&

Bahamas

310

119

42

45

11842

d

u

315

118

41.5

46

67424

9

Cuba

303

116.5

38.5

42

Larus atricilla megalopterus.

13832

0^

Florida

337

121

42

49

13839

d

a

343

124

41

45

30571

&

ti

334

130

43

48

13837

d

u

351

128

40

47

13836

&

u

339

123

40

49

13834

&

u

338

122

41

48

31691

d

u

348

131

43

50

30810

d

New Jersey

349

131

42

48

30631

d

Georgia

337

125

42

49

42175

d

a

356

132

40

46

3062 1

9

u

332

123

42

44.5

13835

9

Florida

322.5

118.5

39

48.5

30691

9

u

326

121

40

47

9. Sterna dougalli dougalli Montagu.
Petite Mauve.

One adult female from Les Saintes, September 10th.

The Roseate Tern is seen rather rarely about the coast of Guade-
loupe, but on the outlying islands it is common. The species is re-
ported by the natives to breed on Les Saintes to the south and on
Tete Anglais to the north of Guadeloupe.

1 Coll. E. a. and O. Bangs.

noble: the resident birds of Guadeloupe. 369

10. Sterna fuscata fuscata Linne.

Mauve a Manteau Noir.

One adult female from Les Saintes, September 10th.
The fishermen report that this species also breeds on some of the
outlying islets of Guadeloupe.

11. Anous stolidus stolidus (Linne).
Mwen. Noddi.

One adult male from Les Saintes, September 10th.

Like the Roseate and Sooty Terns, the Noddy is rarely seen about
the mainland of Guadeloupe but is reported to breed on Les Saintes,
Tete Anglais, and possibly elsewhere nearby.

12. Aestrelata il\esitata (Kuhl).

Diablotin.

One of the chief reasons of my visit to Guadeloupe was to obtain
information about the Black-capped Petrels. A few days after land-
ing I had the good fortune to meet Monsieur C. Thionville, President
of the Club des Montagnards. The name Diablotin was associated
in his mind with the past history and early colonization of the French
in Guadeloupe. He immediately began to make inquiries about
Basse Terre but without much success. Finally we made a trip
together high up into the hills of Matouba to visit an old negro called
Pere Lownisky living on the slopes of the Soufriere. This old man
in his early youth had often hunted Diablotins and had joined several
of the large parties which had camped on the Nez Casse to dig out
the Diablotins from their burrows. Since Pere Lownisky had spent
his entire life in Matouba he knew all the old breeding grounds of the
Black-capped Petrels. He told us that the Diablotins formerly bred
on the north and northeast slope of Nez Casse. The birds arrived
in late September and the period of incubation for the colony as a

370 bulletin: museum of comparative zoology.

whole extended through November and December. The young
birds remained in the nest until March. He asserted positively,
however, that no Diablotins had been heard or seen since the great
earthquake of 1847. The old negro remembered that earthquake for
during it the whole side of Nez Casse, on which the Petrels bred, had
collapsed and fallen into the valley. Pere Lowinsky ended his exposi-
tion by dramatically raising his withered hand, exclaiming again in
his " Creole" French that the Diablotins had not been heard of for
nearly seventy years, — "Jamais! — Jamais!"

A few days later I penetrated the "Grand Bois" as far as the old
breeding grounds of the Diablotins. A sheer wall of basalt arose
for several hundred feet finally losing itself in a bank of rain clouds.
Very little vegetation clung to the steep sides of the cliff. My guides
seemed to think it possible to scale the cliffs by the help of ropes.
But remembering the old negro's statements in regard to the breeding
season I did not make the attempt.

During the rest of my stay on the island I could get no more accurate
information about the Diablotins. Several of the natives believed
queer noises which they had heard nightly some years ago to be the
call of the Diablotin.

The vast jungle covering the mountainous core of Guadeloupe is
nearly impenetrable and entirely unexplored. While it is possible
that a Black-capped Petrel may still breed on some isolated peak in
the heart of Guadeloupe it is significant that not a single bird has
actually been seen in the vicinity of the island in all these years.

13. x\estrelata diabolica (Lafresnaye).

Diablotin.

Upon my return to Cambridge I learned that the Lafresnaye col-
lection (transferred from the Boston Society of Natural History to
the Museum of Comparative Zoology) contained two cotypes of
Lafresnaye's, Procellaria diabolica ^ and two other specimens of Black-
capped Petrel somewhat similar to the cotypes but smaller. These
specimens were all collected in Guadeloupe by L'Herminier in 1842.
On the label of one of the smaller pair (Lafr. coll. No. 8003) the data
reads Maupingue ou Maubingue, and on the other (Lafr. coll. No.
8004) Mauping ou Maupingue.

' The third cotype (Lafr. No. 8001) was exchanged in 1886 with Prof. Alfred Newton for a
specimen of the now extinct Aeslrelatajamaicensis of Jamaica.

noble: the resident birds of Guadeloupe. 371

Lafresnaye (Rev. zool., 1844, p. 168) in his original description of
Procellaria diabolica, referring to the larger specimens, says: —

"Une espece de Petrel, le Petrel Diable, du pere Labbat, Diablotin a
la Guadeloupe, Procellaria diabolica L'Herminier, qui y arrive vers la
fin de septembre, y niche en decembre dans les Falaises; une seconde
espece, en tout semblable de plumage a celle-ci, et n'en differant
que par une taille moindre, y arrive a une autre epoque, niche dans les
memes falaises, mais a un etage different en hauteur, 'ce qui les fait
distinguer a la Guadeloupe par les noms de Petrels des hauts et Petrels
des bas. Ces deux oiseaux seront pris pour la meme espece par tons
les ornithologistes qui les possederont sans renseignements sur leurs
moeurs. Cependent M. L'Herminier les regarde comme constituant
deux especes bien distinctes, differant essentiellement de moeurs et
d'epoque de passage. Mais n'anticipons pas sur les futurs documents
que nous promet notre savant collegue, et qui auront un bien autre
interet sous sa plume et racontes de visu."

The two pairs of Black-capped Petrels from the Lafresnaye collec-
tion are different from each other not only in size but in coloration
and in shape of the nostril tubes. The smaller ones have the grey of
the cap extending down the back of the neck and not terminating
abruptly on the nape as in the larger birds, and the nostril tubes of
the smaller birds are higher and end more abruptly than those of the
larger specimens. In this respect as also in size the smaller birds are
similar to Aestrelata jamaicensis. Each pair represents, I believe, a
distinct species of Aestrelata.

Which species, then, is Aestrelata haesitata? This is a difficult
question to decide because of the uncertainty of the original descrip-
tion. Kuhl (Beitrage zur zoologie, Frankfurt, A. M., 1820, p. 142)
described a petrel "in Museo Bullokiano, nunc in Temminkiano"
and calculated his measurements in terms of the "pollex."

If we assume that the pollex was the Frankfurt a. M. inch of that
time, as determined by the Bureau of Standards at Washington and
sent me by letter, it is then possible that Kuhl's specimen could be
referable to either of the Guadeloupe species under consideration or
better still to neither. Dr. Stejneger, however, recently told Mr.
Bangs that Kuhl was a student of Temminck and would very likely
have used the French system. Changing Kuhl's measurements
from French inches (Ridgway, Nomenclature of colors, 1886, pi. 17
note) into millimeters and comparing them with the measurements
of the Guadeloupe birds we find the figures more closely approximat-
ing the measurements of the small than of the large species. But

372

bulletin: museum of comparative zoology.

still there is so great a discrepancy in the measurements, those of wing
and tail being like the large bird, those of bill and tarsus like the small,
that it is impossible to determine to which species Kuhl's specimen
really should be referred. Since Lafresnaye described the large
species as Aestrelata diabolica I prefer to restrict the name Aestrelata
hacsitata to the small Black-capped Petrel of Guadeloupe. The fol-
lowing table shows the difference in size between the two species.

Measurements in Millimeters.

Bill to

Middle Toe

Angle of

(including

Wing

Tail

Mouth

Tarsus

Claw)

Culmen

Kuhl's specimen com-

306.3

162.1

42.9

38.1

56.1

—

puted by Pied du

Roil

Kuhl's specimen com-

268.71

142,26

37.54

33.58

49.39

—

puted by the Frank-

furt a. M. inch ^

Aestrelata diabolica

287

124

46.5

43

55

36

M. C. Z. 73222

Aestrelata diabolica

288

135

46

42

55

35.5

M. C. Z. 73221

Aestrelata haesitata

264

113.5

42

38

52

32.5

M. C. Z. 73219

Aestrelata haesitata

276

116.5

41.5

37.5

51

33

M.C.Z. 73220

Aestrelata jamaicensis

278

128.5

40

38

51.5

32

M. C. Z. 73218

L'Herminier's list of Guadeloupe birds {cj. Lawrence, Proc. U. S.
N. M., 1878, 1, p. 451) includes both species of Guadeloupe Black-
capped Petrels under the names Procellaria diabolica L'Herm. and
Proccllaria maupinc/ L'Herm. These species are marked with a cross
to indicate that L'Herminier also found them on Martinique. I have
no other information in regard to the Black-capped Petrel in Mar-
tinique.

Investigation will probably show that Aestrelata diabolica and not
Aestrelata haesitata as here restricted is the American Black-capped
Petrel mostly represented in collections. Mr. J. T. Nichols ha?

1 Ridgway, Nomenclature of colors, 1886, pi. 17 note.

2 Determined by Bureau of standards.

noble: the resident birds of Guadeloupe.

373

kindly furnished the following measurements of the specimens in the
collections of the American Museum of Natural History and of Jona-
than Dwight, Jr.

Measurements in Millimeters.

Nostril

Wing

Tail

Tarsus

Culmen

Tubes

Remarks

Amer. Mus. G212

293

133

35

32

Short

Sooty, not greyish

Type of P. meridion-

above, crown

alis LawT.

ending

scarcely marked off

Florida Coast

abruptly

by lateral white
from back

Amer. Mus. 46145

—

—

35

35

Low

Bill and foot only

Long Island

ending

July 1850

gradually

Amer. Mus. 11212

277

143

38

31.5

Low-

Back grey, rump

cf Central Park Zoo.

ending

darker, nape

Jan. 8, 1912.

gradually

broadly white

Captive bird

marking ofi' cap
from back

Coll. J. D. Jr.

290

146

35

32

Low

Nape whitish

Blacksburg, Mont-

marking off cap

gomery Co., Va.

from back. Cap

Aug. 31, 1893,

and back sooty lat-

E. A. Smyth Jr.

ter somewhat
greyish

Coll. J. D. Jr.

290

143

37

35

Low

Back and cap sooty

Cayuga Co., N. Y.

sharply marked off

Sept. 1893

by white nape

It appears from this table that all of the birds in these collections
except P. meridio7ialis Lawr. are referable to Aestrelata diabolica
Lafresnaye. The actual measurements of Lawrence's type are
somewhat larger than those considered typical Aestrelata haesitata as
represented by the two specimens in the Museum of Comparative
Zoology, but the characters of bill and coloration make it referable
to that species.

I have included in the table the measurements of Aestrelata jamai-
censis to bring out the similarity between that species and Aestrelata
haesitata. It would be rash to consider Aestrelata jamaicensis simply
a color phase of Aestrelata haesitata. Yet further study may reveal
that these two species are very closely related.

During the course of my investigation of the status of Aestrelata

374 bulletin: museum of comparative zoology.

hacsitata and Aestrelata diaboHca one more point came to light which
may be of interest. Ahhough Temminck (PI. col., 1826, no. 416)
wrote a brief description of Aestrelata haesitata he did not figure that
species. His plate represents a petrel with grey upper tail-coverts
similar to Aestrelata cervicalis Salvin.

14. Columba squamosa Bonnaterre.
Ramier.

Seven specimens, adult and half-grown birds, from Ste. Claude,
July 2nd, from Ste. Rose July 13th, and from Goyave, August 31st.

The Ramier is the principal game-bird of Guadeloupe. It is a bird
of the rain forest and is found only high up on the " roof of the jungle."
In the early morning and late afternoon scattered flocks dash by high
over head making for their favorite feeding grounds among the taller
fruit trees upon the mountain slopes. The native hunters learn to
know these routes of daily migration and kill great numbers of the
Ramier for market.

15. Zenaida zenaida aurita (Temminck & Knip).

Tourterelle.

One adult female from Goyave September 1st.

The Wood Dove is not rare in Guadeloupe but it is less abundant
than Gcotrijgon mystacca in company with which it is sometimes found.
But unlike the latter, it is widely spread over Guadeloupe and Grande
Terre. It prefers the water's edge and is rarely met with in the moun-
tains. In the open fields, especially those that are bordered with
Mangrove swamps, the Tourterelle is common. This environment is
very different from the hot, sandy hill-sides frequented by the closely
related Zenaida z. lucida Noble (Proc. N. E. zool. club, 1915, 5, p.
101-102) of St. Croix. The Tourterelle thrives well in captivity, and
is perhaps the commonest cage-bird seen in Guadeloupe.

noble: the resident birds of Guadeloupe. 375

16. Chaemepelia passerina trochila (Bonaparte).

Ortolan.

Twelve adult and half-grown specimens from Goyave, August 25th
to September 1st.

On the east coast of Guadeloupe, in the lowlands of the north coast,
and all over Grande Terre, the little Ground Dove is abundant. It
is the commonest bird in the cane-fields, and in spite of its small size
the natives snare great numbers for food.

Long after the young have been fully fledged, the parent birds stay
with them. These family groups feed together about the edges of
the cane-fields. While on the island I never saw an Ortolan, that was
not associated with its family flock.

17. Geotrygon martinica (Linne).
Perdrix Rouge. Perdrix Gris.

One adult female from Goyave, August 20th.

Few of the chasseurs of Guadeloupe know that the Perdrix Rouge
is the male and the Perdrix Gris the female of one and the same species,
but all agree that both are nearly extirpated from the island. The
habits of this species are similar to those of Gcoirygon mystacea but
unlike that species it seems to have been unable to adapt its habits to
the introduced Mongoose. Today it is probably the rarest bird on
Guadeloupe but fifteen or twenty years ago it was abundant and
was considered excellent game.

18. Geotrygon mystacea (Linne).

Perdrix Croissant.

Fifteen adults from Nez Casse, Ste. Rose, and Goyave, taken late
in June, in July, and August.

Since the introduction of the Mongoose all of the species of Per-
drix have suffered because they build their nests near the ground and

37G bulletin: museum of comparative zoology.

within easy reach of even a beast which is strictly terrestrial. But the
Perdrix Croissant is still locally abundant in many of the mountainous
parts of Guadeloupe. Like the other species, it is a rain forest
bird, frequenting the dense wet undergrowths. The dense woods,
however, form an easy approach for the Mongoose. But the Perdrix
Croissant seems to be slowly adapting itself to new conditions. To-
wards evening small flocks fly down from the mountains to feed with
the Tourterelles and Ortolans in the open clearings about the old
cane-fields. My guides informed me that this habit had been re-
cently acquired. It certainly helps to protect the birds from the
stealthy approach of a Mongoose.

The Perdrix Croissant is ranked throughout Guadeloupe, as one of
the best game-birds. The natives formerly caught them with hoops
fitted with' wire snares, and brought great numbers to market. During
my stay on the island I never saw a single Perdrix sold in a village
market.

19. CoccYZUS minor DOMINIC ae Shelley.
Coucou Manioc. Oiseau de Pluie.

Four adults from Ste. Rose, July 13th-19th, six adult and half-
grown specimens from Goyave, August 20th, September 1st.

I have compared a series of eight specimens from Dominica with
eight adults from Guadeloupe and have not been able to find any
appreciable difference between them. My Guadeloupe birds, in spite
of the fact that they are in the worn summer plumage, seem to be a
trifle darker than the Dominica birds and they also average slightly
larger. But I prefer to regard this rather a tendency toward differ-
entiation than a real racial distinction. The Guadeloupe birds mea-
sure:— wing 142.62; tail 163.69; exposed culmen 29.14; tarsus
29.74.

Locally distributed throughout the lower uplands of Guadeloupe,
the Cuckoo is a conspicuous bird because of its slow and clumsy move-
ments. It is confined to the low wooded hills, and not met with at
all in the Mangrove swamps where I had been led to expect it.

The only cry I heard was a resonant guttural chuckle. On dull
days this peculiar call is often heard in the hills. The natives believe
the call to be a sure sign of heavy rain and hence they call the bird
rOiseau de Pluie.

noble: the resident birds of Guadeloupe. 377

20. Streptoceryle torquata stictipennis (Lawrence).
Martin Pecheur. Pie.

One adult male from Goyave, x\ugust 27th.

The Martin Pecheur is found very locally distributed over the whole
of Guadeloupe. It frequents the mountain torrents and pools, flying
overland from one stream to another.

At Goyave I observed two Martin Pecheurs but was able to secure
but one. At least one pair bred in the spring of 1914, in the high
sand banks six miles up the Riviere de Goyave. The nesting hole
resembled that of Streptoceryle alcyon, but was of course much larger.

21. Speotyto guadeloupensis guadeloupensis Ridgway.

Coucou.

One adult specimen, which I was told came from Marie Galante,
presented by the Musee L'Herminier. It now bears the catalogue •
number M. C. Z. 66347. The type of the species is M. C. Z. 74167.

The Burrowing Owl was formerly found on the cliffs of Marie
Galante, but, since the introduction of the Mongoose some twenty
years ago, it has completely disappeared. I could find no evidence
of its ever having existed on Guadeloupe, Grande Terre, or Les
Saintes.

The Musee L'Herminier had five well-mounted specimens of this
species. Unfortunately they bore no data nor was there any record
of their presentation to the Musee. The conservateur, however,
assured me that they came from Marie Galante many years ago.

22. Nephoecetes niger ja.maicensis (Ridgway).

Gros Martinet Noir. Hirondelle de Montague.

Two adult males and one female from Goyave, September 1st.
It seems advisable to refer the Guadeloupe bird to this race, typical
of Jamaica. No specimens from Haiti or Santo Domingo have been

378

BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY.

available for study. Both Ridgway's specimens from Guadeloupe
(Bull. 50, U. S. N. M., 1911, pt. 5, p. 706) and my series average
slightly smaller than the Jamaican birds while the specimens from
Dominica average larger.

Ridgway {Loc. cit., p. 704) refers the Cuban bird to N. nigcr nigcr.
The single specimen before me from Cuba, M. C. Z. 61113, is equally
dark as the specimens from Jamaica and seems indistinguishable from
them. It is included in the following table of measurements.

• Measurements in Millimeters.

Exposed

M. C.Z.

Locality

Sex

Wing

Tail

Culmen

Tarsus

61113

Cuba

cf

156

67.5

6.5

11.5

20446

Jamaica

d^(?)

151.5

66

20447

((

c?

162

63

6

62034

u

d^

154

67

6

20448

((

d^

153.5

60

5.5

21445

u

9

151

60

6

11.5

13615

Dominica

d"

155

67

6

11.5

66334

Guadeloupe

cf

147.5

59

5.5

66333

li

cf

149

62.5

6

13615

u

9

149

62.5

5.5

The Black Swift is especially abundant on the edges of the " Grand
Bois." It was observed at nearl}^ every localit}^ visited except the
flat land of Grande Terre. The bird is called L'hirondelle de Mon-
tagne by the natives because it appears just before sunset flying in
great flocks from the mountains. I found it to be wholly an early
morning and late afternoon flier. None of my guides knew anything
of its breeding habits. But since the bird always comes from and
retreats to the deep rain forest it is not improbable that it may roost
in some of the big hollow trees on the mountain slopes.

23. Chaetura acuta (Gmelin).

Petit Martinet Noir. Hirondelle-Mouche.

Eleven specimens of both sexes from Goyave, August 29th and 30th.

The Lesser Antillean Swift was observed at only a few localities,

first on July 4th at Ste. Claude flying with a number of Black

noble: the resident birds of Guadeloupe. 379

Swifts, and again on July 22d several were seen in an evening flight
over Grand Etang, Deshaies. But at Goyave, August 29th the
species was flying by day. SmaU flocks passed back and forth across
a cow pasture throughout the heat of the day. These flocks remained
in the same locahty for several days and did not seem to be migratory.
Xo flocks were observed m the evening, but scattered individuals were
flving with the Black Swifts.

24. Melanerpes herminieri (Lesson).
Tapeur.

Seven adults from Ste. Rose, July 16th-19th and eight adults from
Goyave August 20th-September 1st.

Ridgway (Bull. 50, U. S. N. M., 1914, pt. 4, p. 12) has created
a monotypic genus for this species. The tendency to split up the
Antillean woodpeckers into separate genera was carried further still
l)y Miller (Bull. Amer. mus. nat. hist., 1915, 34, p. 518) in his descrip-
tion of the Porto Rican form. There are, to be sure, some small
differences between the Guadeloupe bird and any other species of
Melanerpes, but I believe it is entirely a matter of individual opinion
as to whether the Guadeloupe species should be separated from its
mainland relatives and placed in a separate genus. It is perhaps
worthy of note that in Antillean reptiles and amphibians we find most
of the genera to be of wide distribution.

The Tapeur is certainly not a rare bird on the island but seems to
be \ery local in distribution. It is confined to the hardwood belt
covering the hills. Only once did I observe a woodpecker near a
village. Then the bird was clinging to a half-decayed tree in front
of the Gendarmerie at Ste. Rose.

The first time I became familiar with the Tapeur was in a sunny
valley among the hills of Cluny. Every morning upon rising I would
hear the roll of a woodpecker far away on some distant slope. Then a
bird perhaps nearer at hand would answer until finally three or four
would be rolling at once. Often they would call, or rather bray,
and then the note reminded me of the warning cry of the Yellow-
bellied Sapsucker (Sphyrapicus varius varius). There were many
deserted nesting holes nearby, but none was lower than fifty feet from
the groimd.

380 bulletin: museum of comparative zoology.

In other parts of the island the species was less common, rare in
fact on the west coast. Mr. Delphine Duchamp of Goyave believes
the woodpecker had become more abundant about his plantation
within the last ten years.

25. Eulampis jugularis (Linne).
Gros-Colibri. Oiseau-Mouche a Gorge Rouge.

Four adults from the Soufriere June 26th- July 4th, and five adults
from Ste. Rose July 8th-10th.

The Gros-Colibri is the commonest hummer on the island. It
seems to prefer the higher altitudes. About the coffee plantations
of Ste. Claude where the banana flowers were a further attraction this
Red-throated Hummer is abundant. It likes to bask in the morning
sun and often selects a roadside twig for a perch. The small boys
set out straws smeared with gum on these perches and catch the bird
as it alights. Like all of the island hummers this one is easily excited
and comes readily to the "squeak." It then utters at short intervals
a clear "seep."

On July 18th near Cluny I found a nest with its coriipliment of two
eggs. The nest was larger than one of the Ruby-throat's {Archilochus
cohihris) but it was about the same in structure. The eggs were
chalky white.

26. Sericotes holosericeus holosericeus (Linne).
Colibri Bleu. Oiseau-Mouche Bleu.

Two adult ihales from Ste. Claude July 2nd and Goyave August 30th.

The Green-throated Hummer is the rarest of the three species of
Colibri found in Guadeloupe. I observed it in the uplands associated
with Eulampis jugularis, but never in the lowlands.

27. Orthorhyncus exilis exilis (Gmelin).

Fou-Fou. Oiseau-Mouche huppe.

One pair of adults from the Soufriere, June 26th and 29th, and one
adult male from Ste. Rose July 12th.

noble: the resident birds of Guadeloupe. 381

In the rain forest the tiny Fou-Fou is one of the few birds which
one is sure to meet. It is often very pugnacious and on several occa-
sions, when I have excited it by "squeaking" it has darted ahnost
into my face. Although the Fou-Fou is the smallest bird on Guade-
loupe it will drive away Sucriers, Gros-Becs, and even Grives from
its favorite honey tree. In spite of its tiny size, it makes a great
deal of noise during one of these attacks, — a crackling volley of psist I
psist!! psist!!! loud enough to frighten even a man. The species is
by far the noisiest of the Guadeloupe hummingbirds.

28. TyRANNUS DOiMINICENSIS VORAX (VieiUot).

Pipiri.

Five adults from Ste. Rose July 12th, 14th, and one from Goyave
September 2nd.

Ridgway (Bull. 50, U. S. N. M., 1907, pt. 4, p. 708) states that
Tyrannus d. dominicensis occurs in Guadeloupe. All of the six speci-
mens, which I collected,, are typical Tyrannus d. vorax. One, how-
ever, is smaller than the other Guadeloupian birds but larger than the
average for the Greater Antilles. The occurrence of Tyrannus d.
vorax in Guadeloupe is to be expected on zoogeographical grounds for
the same race is found north as well as south of the island (cf. Riley,
Smith, miscell. coll., Nov. 8, 1904, 47, p. 2).

The Pipiri is perhaps the most conspicuous if not abundant bird
of the lowlands. It resembles the Kingbird (Tyrannus tyrannus) in
that it selects a high perch overlooking some bit of pasture and from
these sallies forth after the passing flies. As it darts out the snap of
its bill may be heard for some distance. Of all the bird notes about
the cane-fields, the one that catches a stranger's ear first is the sharp
pip-piree, pip-pirr-ee of this bird. In the early morning the bird is
especially active and its call-note seems to arise from every corner of
the plantation. I imagined the bird received its name from its call
but Ballet (L'histoire de la flore, la faune etc. de la Guadeloupe, Basse
Terre, 1895, 1, 2, p. 21) says: — "Pipiri vient sans doute du mot
breton pipirette, expression dont on se sert en Bretagne pour designer
I'aube ou piperette du jour."

382 bulletin: museum of comparative zoology.

29. Myl\rchus oberi oberi Lawrence.
( Pipiri Gros-Tete.

One immature male from Ste. Rose, July 11th.

The single specimen secured was one of the two birds of the species
I saw on the island. The pair was observed in an area of deciduous
scrub about four miles south of Ste. Rose. My guide, one of the best
chasseurs of that village, said he had never seen the bird before on
the island. The species was doubtfully recorded from Guadeloupe,
and Ridgway (Bull. 50, U. S. N. M., 1907, pt. 4, p. 618) questions
Guadeloupe as a locality for it.

30. Bl.\cicus bruxxeicapillus La^Tcnce.
Gobe-INIouche Brun.

One adult male from the Soufriere June 29th and eight adults and
half-grown specimens from Ste. Rose July 12th-16th.

I found the Brown Flycatcher not rare in Guadeloupe, but rather
locally confined to the clearings in the deep woods. It seems to prefer
the solitude, of the forest, for it only comes into the open when hawking
flies.

31. Elainea flavogastr.\ martinica (Linne).
Petit Pintade. Gobe-]Mouche huppe.

Not a rare bird in the woody parts of Guadeloupe. Two specimens,
both sexes from Ste. Claude, July 3d ; two birds, one female and one
unsexed from Ste. Rose July 19th; and five specimens of both sexes
from the Soufriere June 29th.

Near the highest part of the island visited, I took my first specimen
of Petit Pintade. Then in the heavy cover about Ste. Claude, July 4th
and near Morne Rouge, August 22d I observed several scattered flocks
of a few^ individuals. In the lowlands where large berry producing
trees are absent this species was rare.

Clark (Proc. Bost. soc. nat. hist., 1905, 32, p. 208) has described
the song at some length. Only on rare occasions did I hear the bird
sing in Guadeloupe, and it was then a long clear whistle. In habits

noble: the resident birds of Guadeloupe. 383

this bird is very like a wood-warbler reminding one particularly of a
Black-poll {Dendroica striata).

32. HoLOQUiscALus guadeloupensis (Lawrence).

Holoquiscahis martinicensis Ridgway.
Merle. Bout de Petun.

Nine specimens, adults and half-grown males from Ste. Rose, July
12th-18th; four specimens from Basse Terre, July 3d.

I fail to find any characters by which to separate H. martinicensis
from H. guadeloupensis. Ridgway (Bull. U. S. N. M., 1902, 50,
pt. 2, p- 232) says of H. guadeloupensis "Similar to H. martinicensis
but wing averaging slightly longer." His measurements for the wing
of the male of H. martinicensis are: — 119.4-124 (120.7) and for the
female 102.4-108.5 (105.4). For the male of H. guadeloupensis, on
the other hand, his wing measurements are: — 119.9-124.5 (121.9),
for the female 104.4-109.2 (106.7).

My averages for the wing of H. guadeloupensis are also slightly
larger than those of H. martinicensis, but this difference is very small,
and there is, apparently, no other distinguishing character. I do
not believe a distinction can be made between the two forms, and it
seems best to refer the ^Martinique bird to H. guadeloupensis. On
geographical grounds alone there would seem to be a real difference
between these two birds for the species has never been taken on
Dominica, lying just between and in plain sight of Guadeloupe and
Martinique. But if isolation has tended to make a distinction be-
tween the two forms, this distinction is, to my mind, at present not
great enough to recognize two species.

Holoquiscalus martinicensis Ridgway.

Exposed

M.C.Z.

Sex

Locality

Wing

Tail

Tarsus

Culmen

10895 1

d^

Martinique, F. W. I.

120.5

96

36

27.5

276SS

cT-

u u

llS-i-

100+

35.5

27

28696

d^

a u

120

102

38-

28

28695

9

a u

lOS-

87

34

24

11272

9

u u

105

85

33

24

11273

9

u u

104+

80

37-

25

1 CoU. E. A. and O. Bangs.

384 bulletin: museum of comparative zoology.

Holoquiscalus guadeloupensis (Lawrence).

Expssed

n.cz.

Sex

Locality

Wing

Tail

Tarsus

Culmen

66571

cf

Guadeloupe, F. W. I.

123.5

104-

36

28

66570

cf

a a

119,

100

35

29

66571

d^

( u

121

102.5

35

27.5

66572

d"

I u

120+

102+

37.5

27

66566

9

I ' u

104.5

80

31.5

21 +

66567

9

I u

107.5

88

31

25.5

6656S

9

i u

106

80

32

21+

66578

9

( u

105.5

83

30

24

14853

9

< «

81

32-

24-

Although the Merle never ascends to any of the high pastures at
the edge of the rain forest, it is still fairly abundant over the rest of
the island. I did not meet with it at all during my stay at Ste. Claude,
Vieux Habitants, INIorne Rouge, or any of the other high regions I
visited. But about the low plantations of Ste. Marie and Goyave
it was very abundant. At Ste. Rose from July 12th-20th the Merle
was seen nearly every morning in the Mango trees surrounding a large
cow pasture. In fact wherever herds of cattle are grazing one may
feel certain of seeing or hearing some INIerles providing the altitude

IS not too great.

Ballet {Loc. cit., p. 23) in speaking of the Merle says: —
"Cet oiseau vit par bandes nombreuses, aime a se percher sur les
grands arbres, notamment sur les palmistes, se perche sur les boeufs
pour devorer les tiques et autres vernlines dont ils sont converts, est
tres famillier, suit le laboureur, et, pose sur les boeufs, ou la charrue,
se precipite sur les larves et les insectes mis a decouvert par cet instru-
ment aratoire. Sa chair n'est pas bonne. Aussi, grace a son peu de
gout, il a echappe a la destruction et rend a notre agriculture d'im-
menses services. Ses bandes multipliees couvrent la Grande - Terre
et une partie de la Guadeloupe."

The notes of the Merle are as varied as those of the Starling (Sturnus
vulgaris). When in flocks the Merle keeps up an incessant jabbering.
Sometimes, especially in the early morning, the bird gives its true call-
note, a double whistle of two syllables, the second rather prolonged.
But it is the conversational jabbering which is most characteristic
of the bird.

noble: the resident birds or Guadeloupe. 385

33. Pyrrhulagra noctis dominicana Ridgway.
Pere Noir. Gros-Bec.

Thirteen specimens, adults and half-grown, of both sexes from Ste.
Claude July 3, and from Ste. Rose July 15th, 16th, and 20th.

When we consider the tendency of Pyrrhulagra to break up into
island races in the Lesser Antilles, we might readily expect the Guade-
loupe bird to be distinct. This series, however, is much too small to
distinguish the Guadeloupe from the Dominica bird. My measure-
ments fall within the limits shown by Ridgway 's specimens (Bull.
50, U. S. N. M., 1901, pt. 2, p. 556).

The Pere Noir is the common sparrow from the high woods to the
mangroves. It is the only bird that is evenly distributed over the
whole of Guadeloupe. Several of these birds were seen June 27th-
29th at the "Club des Montagnards" near the crest of the Soufriere,
others July Ist^th about Ste. Claude, and then on July 20th it was
seen again in equal abundance in the mangrove swamps about Ste.
Rose. The natives apply the name Gros-Bec to both sexes while
they reserve the name Pere Noir for the male. The birds are often
found in flocks in the dooryards of the houses. Its sharp chirp and
clear song, pseepi pseep!! pseep!!! are characteristic sounds of the
country villages.

34. TiARis BicoLOR OMissA (Jardinc).
Olive. Mangeur d'herbes.

Two adults, male and female from Goyave, September 5th and 7th.

The grassquits of Grenada, the Grenadines, St. Vincent, and Bar-
bados instead of being referable on Grenada and Barbados to T. h.
marchii, and on the other islands to the more wide-spread T. h. omissa
as believed by Ridgway (Bull. 50, U. S. N. M., 1901, pt. 1, p. 538,
541) really represent a well-defined race which is apparently confined
to these southern islands of the Lesser Antilles. It may be distin-
guished at once from the two races mentioned by its different size and
coloration, and it may be known as

Tiaris hicolor expectata, subsp. no v.

Type: M. C. Z. No. 13109, (E. A. & O. Bangs Coll.) from St. George,
Grenada, June 19, 1904, Austin H. Clark. In measurements it is

386 bulletin: museum of comparative zoology.

smaller than T. h. marchii. The average of 14 males, and 9 females
from Grenada and Grenadine, 4 males and 2 females from St. Vincent,
and 6 males and 4 females from Barbados, — 39 specimens in all are : —
wing 51.3 (48-52.5); tail 38.2 (36-40); exposed culmen 9.1 (8.5-
9.5); depth of bill 7.4 (7-8); tarsus 17.3 (15.5-18). While on the
other hand a series of 39 specimens, 26 males and 13 females from
Jamaica, representing T. b. marchii, the average is : — wing 52.4
(50-54.5); tail 40.6 (38.5-42.5); exposed culmen 8.9 (8.5-9.5);
depth of bill 6.8 (6.5-7.5); tarsus 17.1 (16.5-17.5).

In coloration it differs widely from T. h. marchii in that the dark
area of the breast is not sharply divided, but merges gradually into
the white of the belly-region with often scattered spots in blotches of
darker color, extending posteriorly along the sides of the belly and
encroaching considerably into the white area.

It is similar to T. b. omissa but has a longer tail and different color-
ation. Clark (Proc. Bost. soc. nat. hist., 32, p. 286) considers both
the Grassquit from Grenada and St. Vincent as referable to T. b.
omissa. The series of skins before me, however, show that grass-
quits from Grenada, the Gernadines, Barbados, and St. Vincent differ
from those of the other islands of the Lesser Antilles by having more
white on the belly. The tails of the former birds are relatively longer
averaging 38.2 against 37.1 of those of the latter. These characters
are rather constant throughout and are sufficient, I believe, to dis-
tinguish separate geographical races.

I have examined a series of twenty-seven skins from other islands
of the Lesser Antilles but fail to find any characters upon which to
separate the bird occurring on the islands from St. Lucia to Porto Rico
from the mainland specimens. It seems anomalous to find T. b. omissa
in Tobago and Venezuela'and then skipping Grenada, the Grenadines,
and St. Vincent occurring again on the other islands of the Lesser
Antilles. But until a larger series of skins can be examined it is
perhaps premature to cite this as a case of convergent evolution.

Ridgway (Bull. 50, U. S. N. M., 1901, pt. 1, p. 539) records two speci-
mens of T. b. omissa taken in Cuba. Since there are no other Cuban
records I incjuired into their authenticity. Mr. C. B. Cory, of the Field
Museum, informed me by letter that the birds were given to him by
Gundlach and Cory supposed that they came from Cuba. Like so
many other of Gundlach's birds they were probably collected during
one of Gundlach's three trips to Porto Rico. The Grassquit is not so
abundant on Guadeloupe as in the northern Lesser Antilles. On
Guadeloupe it is confined to the lowlands where it prefers the hot

noble: the resident birds of Guadeloupe. 387

road-sides. The bird is rather shy and upon approach it disappears
quickly into the nearest thicket. Its monotonous call has something
of the buzz of a locust and something of the call of the Sucrier in it.

35. Saltator albicollis Vieillot.
Grive-Gros-Bec.

Twelve specimens, adults and half-grown from Ste. Rose July 12th
to 16th; and from Goyave August 20th to September 4th.

I never met with the Grive-Gros-Bec during my stay on the west
coast, but in the north and east of Guadeloupe I found that the bird
was not rare. It usually freciuents the small stands of hardwood on
the mountain slopes. In the rain forest or about the plantations it
was less often seen.

This species is the "Grive" most frequently shot. It is ranked
as a game-bird in spite of its laborious movements. Its low chuckle,
its stealthy but nevertheless clumsy approach by hopping from twig
to twig and finally its loud whistle are all very characteristic. In
flight and general habits the Grive-Gros-Bec reminds one most of a
very young and awkward Pine Grosbeak (Pinicolor enucleator leucura).

36. EuPHONiA FLAViFRONS FLAViFRONS (Sparrman).
Perrouche. Perrique de Matouba.

Four adults, three females and one male, from the slopes of the
Soufriere near Matouba taken June 28th.

The Mistletoe bird is confined to the steep and heavily wooded
slopes of the interior. It feeds in flocks on soft fruits and berries.
All four of the specimens taken had their crops stuffed with gelatinous
coated fruit-seeds. The plaintive whistle of the Perouche is often
heard in the deep, vine-covered gorges of the Soufriere but because of
the bird's small size and its retiring habits one rarely catches a glimpse
of this, the most beautiful of Guadeloupe birds. In fact the natives
believe that at certain seasons of the year when the wild fruit is ripe,
flocks of these tanagers fly over from Dominica. It seems more
likely, however, that they had previously overlooked the bird.

388 bulletin: museum of comparative zoology.

37. Progne dominicensis (Gmelin).

Hirondelle de Dominique.

Six specimens of both sexes from Goyave, August 30th and 31st
This species is not rare about the low plantations of the east coast of
Guadeloupe and of Grande Terre. The first bird I saw upon reaching
Guadeloupe, June 22d, was a martin flying about the stern of the 'S'essel.
A flock seemed not only in color but in flight like great tree-swallows
hovering overhead. On the west coast the bird was exceedingly
rare and it was not until I had taken up my abode at Goyave that I
really became acquainted with it. On warm moist evenings, so char-
acteristic of the Windward Islands, widely scattered flocks of Progne
dominicensis and Ncphoecctes yiiger jamaicensis would appear and
skim over the cane-fields. During the heat of the day neither of these
birds is visible. In the early morning especially along the water-
front I found this Martin rather abundant.

38. Vireosylva calidris barbadensis Ridgway.

Siffleur.

Five specimens from Goyave and the Soufriere taken the first week
in July and the last week in August.

When passing through the patches of hardwood trees that skirt
the deep forests one is sure to hear the whistle of this bird coming
from the top of some lofty forest giant, and even among the scrubby
growths of the lowlands one may often hear that same clear note.
In writing of this bird Ballet (Loc. cit., p. 22) says: —

" On lui a donne ce nom, parce qu'il imite parfaitement le sifilet de
la voix humaine. II se tient dans les bois. On croit toujours quand
on I'entend que c'est un homme qui en appelle un autre. II n'y a
point d'etranger qui n'y soit trompe."

This bird resembles the Red-eyed Vireo not only in its choice of a
home but also in its actions, in its song, and its nest building.

. noble: the resident birds of Guadeloupe. 389

39. Coereba dominicana (Taylor).
Sucrier.

Common in the cultivated regions. Ten specimens from the plan-
tations about Ste. Claude and Ste. Rose.

I have examined a series of five adults of this species from Guade-
loupe, four from Antigua and five from Dominica, and I find a con-
siderable variation in both color and size. In adults from the same
island the breast and belly varies from yellow-green to bright ochra-
ceous while the upper parts differ considerably in their intensities.
The specimens from Antigua hAve a constantly shorter tarsus, it
averaging 17.2 mm. against 18.2 for those of Guadeloupe and 18.4
for those of Dominica. The other measurements are all within the
limits of individual variation.

The white wing-spot does not seem to be a good specific character.
In two of the specimens from Antigua, one from Guadeloupe, and two
from Dominica this mark is just visible. Upon examining a series of
nine adults from St. Croix and another of nineteen adults from
Grenada and Grenadines I fail to find that the characters given by
Ridgway (Bull. 50, U. S. N. M., 1902, pt. 2, p. 400) in his key hold true.
Instead of C. saccharina having a smaller wing-spot than C newtoni,
it has in every case at least as large if not a larger one. Again, the
superciliary stripes of the latter are not broader than those of the
former, rarely, however, they come closer together on the head (2
specimens) and spread out so as to touch each other. C. neivtoni may,
however, generally be distinguished from C. saccharina by its lighter
color, especially on the breast and throat, and again by its shorter
tarsus averaging 18.3 mm. (fourteen specimens) against 19.1 for the
latter.

Everywhere in the lowlands and as far up the mountain as the
Grand Bois extends, the Honey Creeper is abundant. About the
banana plantations I found them in greatest numbers, often nesting
in the vicinity of the houses. When examining a flower this species is
as acrobatic and agile as a nuthatch, while its undulating flight is
much more graceful. The song is a characteristic little wheeze, —
Zee ! — Zee ! Because of its great curiosity the Honey Creeper is
often killed by the small boys. All the country gamins know how to
"squeak up" this little bird into range of their blow-pipes or they can,
at least, catch them by means of straws smeared with bird-lime.

390

bulletin: museum of comparative zoology.

40. Dent)roica plumbea Lawrence.
Fauvette Grise.

Ten specimens, adult and half-grown, taken on the slopes of the
Soufriere in the vicinity of, and above Ste. Claude, during the latter
part of June and the early part of July.

The avifaunas of Guadeloupe and Dominica are by no means identi-
cal, nevertheless many of the small birds are the same on both islands.
I have examined a large series of adults and young and find tha't this
species does not differ in the two localities.

Adults from Dominica.

Exposed

I.C.Z.

Sex

Wing

Tail

Culmen

Tarsus

60950

9

61.5

49.5

10

21.2

13569

9

61.8

51

10.5

21.2

13610

9 (?)

64.5

52

10.5

21.5

28767

d'

60

51.5

10.5

20.5

28766

cf

60

50.5

10.2

20.5

13570

cf

67

56

11

22

(albinistic)

Adults from Guadeloupe.

Exposed

Sex

Wing

Tail

Culmen

Tarsus

9

60.2

53.5

10.2

20.5

9

59.8

49

10.5

20

c?

63

51

10

20.2

cf

63.5

51.5

10.2

19.8

cf

62.5

51.5

10.8

19.8

cf

63.5

50.5

10.8

20.5

c?

67

55.5

10.5

21.5

I never met this bird an\T\-here but in the deep woods. It is a true
wood-warbler and as such does not descend into the lowlands unless
following some extension of the forest. In the dense dripping woods
that cover the sides of the Soufriere, it is often met with and is always
one of the first birds to answer to the " squeak." At Ste. Rose where
the heavy woods of Soffire are so near at hand, this little warbler
may be found just outside of the town.

noble: the resident birds of Guadeloupe. 391

Camping near the top of the Soufriere, I found the bird common
about the mountain streams. Near the base of the huge mass of
denuded lava which forms the summit of this volcano, an albino
female was taken. The head and neck are white, a series of white
blotches extend down the back and sides while the rest of the plumage
is the normal olive-grey.

( 41. Dendroica ruficapilla ruficapilla (Gmelin).

Oiseau Jaune.

A common species in the lowlands. Fifteen specimens from
immediate neighborhood of Goyave, Ste. Rose, and Ste. Claude taken
on various days throughout July and August.

Clark (Proc. Bost. soc. nat. hist., 1905, 32, p. 294) says "The sub-
species of this form, D. r. rvficapilla (Guadeloupe and Dominica),
D. r. rufivcrtcx (Cozumel Island), D. r. flavida (St. Andrew's) and
D. r. rufopileata (Curasao) appear all to fall wnthin the range of indi-
vidual variation, if we can judge from the great differences exhibited
by a series of sixteen specimens of the closely related D. capitalis
of Barbados. The only specimen from Cozumel Island which I
have been able to examine, as well as three specimens from Dominica
* * * are inseparable from Grenadine examples." An examination
of a large series of this species including the specimens taken by
Mr. Clark as well as those collected by myself shows that the dis-
tinguishing characters of the described races have no more value than
Clark gave them. It is clear that only one form should be recognized.

Among the mangroves, about the plantations and ascending the
sparsely wooded hills this bird is common. Ober says of it (Proc.
U. S. N. M., 1878, 1, p. 453) "with the two sparrows the bird is most
commonly met with in the gardens and coffee plantations. In the
latter, I find it chiefly in the pois douce trees, which, originally planted
as wind-breaks for the coffee plants' protection, seam the hills all
around in long rows." About Goyave and Ste. Rose I found it most
common in the small plantations bordering the mangroves. It
sometimes occurs in numbers upon the high uplands, but I have never
taken it higher than Ste. Claude some two thousand feet above the
sea. Its habits, nesting, and song are all very much like those of the
Yellow Warbler (Z). aestiva acstiva) but unlike this species it seems

392 bulletin: museum of comparative zoology.

to prefer the hot scrubby fields to the streams and swamps, although
some are always to be found about the mangroves.

42. Troglodytes guadeloupensis (Cory).
Rossignol.

One adult female taken July 13th, near Ste. Rose.

This species is now practically extinct owing to the Mongoose.
Twenty-five years ago the bird was widely distributed all over the
island, and like the House Wren would frequent the gardens about
the villages. For more than ten years it has been extremely rare and
local, found only in the high woods which have been more or less
cut over. Although I visited over a dozen distinct localities on
Guadeloupe it was only seen on the wooded hills back of Ste. Rose.

It was in the evening, shortly after the sun had set that I heard for
the first time the song of this wren. It was long and varied with more
of a warbler quality than that of a wren. But the profusion and
bubbling merriment of the song could be given only by a wren. As I
advanced through the brush, the bird darted off to a fallen log, ran
nimbly- along it, hopped into a low tree and began to flit upwards from
one branch to the next till it had reached the top. Then it flew off to
another tree to again start its spiral climb upwards. When finally
shot it proved to be a female, and although unsuspected until the
specimen was dissected I had probably been following a pair of wrens
and not a single individual.

43. Cinclocerthia ruficauda tremula (Lafresnaye).
Trembleur. Grive Trombleuse.

Eight specimens, adults and half-grown individuals, taken through-
out July and the latter part of August from the Soufriere, Ste. Rose,
and Goyave.

Confined entirely to the deep woods, this bird is one of the few
species one meets while struggling through the forest. When flushed
from the ground where it is habitually to be found, it flits up to a low
branch and begins to shake as if in the grasp of a tropical fever. At

noble: the resident birds of Guadeloupe. 393

the same time it jerks the tail nervously up and down just as a Spotted
Sandpiper does, accompanying these movements by a bobbing of its
head in every direction. My guides said that the small heaps of snail-
shells quite often found upon the forest-floor were made by this bird
which feeds almost entirely upon the snails. The species is so rare
and local that I was unable to verify this assertion.

The Mongoose, now found in every part of the island, has almost
exterminated it. The greater part of my time on the island was
spent in the deep woods where only a few were found. If one moves
very quietly to a suitable place for "squeaking" this bird may be
easily induced to come within gunshot, but I have never heard it
utter any answering call except and only rarely a low guttural sound.
The coloration and actions of this bird are in keeping with its environ-
ment. Its uniform dark plumage makes it invisible among the dark,
decaying leaves of the forest-floor, while its silent flitting to and fro
are in harmony with the great hush of a tropical jungle. The peculiar
trembling habit is probably some sort of a warning motion, but during
this action the bird is not very unlike a bunch of dried leaves shaking
in the wind.

44. CiCHLHERMiNiA HERMiNiERi (Lafrcsnayc).

Cichlherminia coryi Ridgway.
Grive a Pieds Jaunes.

Twenty-four specimens from the region about Ste. Rose and Goyave
taken during the latter part of July and August.

This series of specimens together with Lafresnaye's and Ridgway's
types make it clear that Cichlherminia coryi Ridgway (Smiths, misc.
coll., 1904, 47, p. 112) is the adult oi Cichlherminia herminieri (Lafr.).
These specimens show a gradual change from one to the other kind
of plumage, a change which is undoubtedly one of age. Lafresnaye's
original labels for the types of these two species show that all the
birds came from Guadeloupe. The specimens were sent to Lafresnaye
by Dr. F. L'Herminier in 1844, and are now together with their
original labels in the M. C. Z.

Ober (Proc. U. S. N. M., 1878, 1, p. 452) in his field-notes on this
bird writes : —

" A resident of the wooded hills and mountains; found in Dominica

394 bulletin: museum of comparative zoology.

in the same localities as the Perdrix, woods sufficiently free from under-
brush to afford places for scratching. The places where they have
disturbed the earth by scratching are frequently seen in the paths,
where the woods are thick, and in the open forest. They will come
quickly at the call if within hearing, but are shy, flying cautiously
from tree to tree never long at rest."

According to my experience they are extremely local requiring a
thick wet forest but not one so thick as the rain forest of the high
ranges. In Guadeloupe this species is practically unknown on the
west coast. To the north of the island it is only back in the hills of
Bellevue that the bird is found. In the south and east I found it
not rare about the foothills of Morne Rouge, near Goyave, and again
at Ste. Marie.

As Ober has said this species is a ground inhabiting bird. For this
reason it was one of the first victims of the Mongoose. Since as Ballet
(L'histoire de la flore, la faune etc. de la Guadeloupe — Basse Terre
1895) expresses it "la chair est la plus tendre et la plus delicate,"
the natives for many years have set out great hoops each with sixty
to a hundred snares attached. The bird while scratching would
become entangled and would eventually be found and taken to market
where it would be sold for five cents. Now everything is changed,
the Mongoose has nearly exterminated the species while those few
individuals that remain seem to have adapted themselves to the new
conditions, and if one can believe in hear-say are on the increase.
This change is probably a forsaking of the ground for the low dense
shrubbery. I did not once surprise the bird scratching among the
leaves. Often I have heard its bell-like note coming from high up in
the trees. It is a very characteristic call, mellow, resonant, and
repeated at frequent intervals.

45. Allenia apicalis (Hartlaub).
Grive Fine. Grive Cendree.

Thirteen specimens from five different localities on Guadeloupe,
taken during both July and August. This species is not rare wherever
the environment is suitable.

An examination of a large series of specimens from the islands of
the southern Lesser Antilles shows a tendency for this species to

noble: the resident birds of Guadeloupe. 395

become darker in color as it progresses southward. This diflFerence
does not seem to be seasonal or sexual. Between any two adjacent
islands there is no marked difference in color, but the birds from
Barbados and Grenada are distinctly darker than those from Guade-
loupe. Because of this intergrading, it seems hardly advisable to
draw a line between the northern and southern birds by separating a
race.

J. H. Riley (Smith, misc. coll., 1904, 47, p. 288) has noted this
slight difference and in speaking of the Barbuda and Antigua birds
says : —

"The above series, when compared with a series from the other
Lesser Antilles, averages more olive brown above, without the reddish
cast in the plumage seen in the other series before me. The measure-
ments are also slightly larger as the following will show : Four males
from Barbuda and Antigua average: wing, 129; tail, 104.6, culmen,
20. Seven males from Saba south to St. Vincent average: wing,
121.3; tail, 95.6; culmen, 19.4."

I find the Guadeloupe bird compares favorably in measurements
with the Barbuda-Antigua birds. Three males from Guadeloupe
average: — wing, 121.3; tail, 93.6; exposed culmen, 18.3; tarsus,

29.2. Three females from Guadeloupe average: — wing, 124.9;
tail, 98; exposed culmen, 18.8; tarsus, 31.

The measurements of the Dominica birds, on the other hand, fall
well within the range of variation in a series of twenty birds from the
islands to the south as far as Grenada. Three males from Dominica
average: — wing, 114.6; tail, 90.5; exposed culmen, 18.8; tarsus,

28.3. It seems evident then that if a southern race were to be sepa-
rated from a northern the Guadeloupe bird would be included in the
northern and the Dominica bird in the southern. Such a splitting
up of races on these two closely associated islands is not at all the rule
of subspecific differentiation in the avifauna of the Lesser Antilles.
On Guadeloupe this species is the commonest of the three Grives, but
I did not meet with it at all during my short staj^ on some of the more
northern islands. It prefers the borders of woodlands made up of
pure stands of deciduous trees or again the scrubby upland fields;
still it is not rare even in the deep woods wherever there is a clearing
formed by a tree which has crashed down.

The song of this bird consists of a few high notes uttered slowly and
deliberately, generally from the top of some small tree standing at
the edge of a clearing. The birds congregate only when feeding on
the small fruit-trebs, and it is rare that you find them together.

396 bulletin: museum of comparative zoology.

46. Margarops fuscatus densirostris (Vieillot).
Grosse Grive. Grise Corossol.

Eleven specimens from the deep woods near Ste. Rose, the Soufriere,
and Goyave taken in the latter part of July and the latter part of
August.

Few birds of Guadeloupe are more strictly confined to the deep
woods than this species. Very shy and retiring in habits it seeks the
tallest trees of the rain forest. On my homeward voyage from
Guadeloupe I was greatly surprised to find the closely related Mar-
garops f. fuscatus in the streets of Christiansted, St. Croix; for I asso-
ciated such a bird with anything but the noise and bustle of traffic.
Perhaps the lighter color of this form to the north has been brought
about by its open and sunny habitat.

The Grosse Grive is considered throughout Guadeloupe as one of the
best game-birds to be ranked even with the Ramier and Perdrix; and
since it lives on the roof of the tropical forest covering the higher parts
of the island, the hunters are put to considerable trouble to obtain it.
When disturbed the bird utters a sharp cluck, entirely different from
the alarm-note of any other bird of the region. The cluck is repeated
at intervals and is accompanied by a simultaneous lowering and
jerking upwards of the tail. The bird's song is loud and clear con-
sisting of a series of long whistles. In attracting this Grive, the
natives give a long call of low vibrant sounds, — shush! shush! — not
unlike the puffing of a distant locomotive. They explain this call
as the imitation of a mother Grive hovering over a young one which
has fallen from the nest or been overtaken by some other calamity.

SEP 28 hj)6

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LX. No. 11.

THE STANFORD EXPEDITION TO BRAZIL, 1911, JOHN
C. BRANNER, DIRECTOR. THE ANTS OF BRAZIL.

By William M. Mann.

With Seven Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

September, 1916.

No. 11. — The Stanford Expedition to Brazil, 1911. John C.
Branner, Director. The Ants of Brazil.

Contributions from the Entomological Laboratory op the Bussey
Institution, Harvard University, No. 114.

By William M. Mann.

As Entomologist to the Stanford Expedition, it was my privilege
to spend the summer and fall of 1911 making collections in various
parts of Brazil. Natal and Ceara on the East Coast were our head-
quarters for the first three months, and from these stations side trips
were made to nearby points. In the early part of July, when the
other members of the party returned to the States, Dr. Fred Baker
and I remained at Para, and subsequently went up the Amazon to
Manaos. We hoped to be able to get into the region of the upper
Rio Madeira. Our hopes were more than realized, for Mr. May,
senior member of the firm, May, Jeckyll & Randolph, then engaged
in constructing the Madeira-Mamore Railroad, took an active interest
in our work and provided us with transportation from Manaos to
Porto Velho on the little steamer owned by the company, and with
letters to several of the engineers and medical men. From the time
of our arrival at Porto Velho till we left, we were shown the greatest
hospitality by everyone in the foreign colony. Dr. James Laidlow
and Messrs. Nixon and Troop cared for us in their homes at Porto
Velho, while along the line of the railroad and in the construction
camps everyone gave us all possible assistance. Some of the men, as
Messrs. H. N. Burton and Fry, themselves interested in insects,
collected during spare hours and generously turned over to us such
specimens as we wanted. It is impossible adequately to express our
appreciation of the kindness with which we were treated while on the
Madeira-Mamore.

As a great deal of territory was covered, and several distinct faunal
regions visited, it seems advisable to give a list of the localities in which
collections were made. These are:

400 bulletin: museum of comparative zoology.

State of Rio Grande do Norte.

Natal. On the coast, in an arid, sandy and wind-swept district.

Ceard-Mirim. North of Natal. The immediate vicinity of the
town is well cultivated, with fields of cane and cotton, and considerable
woody land and abundant water.

Baixa Verde. The terminus of a little railroad running out of
Natal in a northwesterlv direction. The surrounding countrv is
hilly and very arid, with much scrub and cacti. At the time we col-
lected here everything was dry and collecting was to be had only
beneath stones, with which the ground was strewn.

State of Parahyba.

Indepcndencia. Dr. Heath and I spent a week at the little village
of Itamatahy, near Indepcndencia, as guests of Messrs. Nye and
Tessire, engineers on the local railroad. The country is hilly, with
abundant vegetation and water. Bamboo breaks afforded interesting
collecting.

State of Gear a.

Geard. On the coast in extremely arid surroundings. There is a
good stream near the city, along which was a fairly abundant fauna.
This and the following are type localities of many of the ants described
by Mayr and Forel.

Baturite Mountains. Mr. Williams, the Director of the railroad
between Ceara and Inixada gave the members of the Expedition an
excursion to the latter village. En route it was possible to collect at
several points in the mountains.

Maranguape Mountains. INIr. Lieb, Assistant Geologist of the
Expedition, and the writer made a side trip into these mountains and
collected for a day. In the humid canyons and on the hi^sides were
taken several species not found elsewhere.

State of Pard.

Para. Nearly all of the material labeled Para was taken in the
forest on the outskirts of the suburb Souza.

Santarem. Visited on our return trip down the river. We were
able to spend only a few hours here, chiefly among the scrub in the
sandy region back of the town.

MANN: THE ANTS OF BRAZIL. 401

State of Amazonas.

liacoaiiara. On the north bank of the Amazon, about forty miles
below the mouth of the Rio Madeira.

Porto Vclho. The starting point of the Madeira-Mamore Railroad,
near the border of Matto Grosso, about three miles from the Brazilian
village of Sao Antonio.

State of Matto Grosso.

Ahund. On the Brazilian side of the river, nearly opposite the
mouth of the Rio Abuna which forms the boundary of Bolivia and the
Brazilian State of Matto Grosso.

Madeira-Mamore R. R. Camps. Collections were made at a number
of construction and other camps along the line of the railroad. The
numbers of the camps are given as data. The locality can be most
exactly expressed by giving the distances in kilometers from Porto
Velho. These are:

Camp No.

28

170 kms

« u

39

284 "

11 u

41

306 "

U it

43

325 "

My study of the collection of Formicidte has been made possible
through the aid of Prof. W. M. Wheeler who has constantly followed
the course of my work and generously permitted me to use his exten-
sive collections and library at all times.

I wish to thank Dr. J. C. Branner, through whom I was enabled to
accompany the expedition, and Dr. Fred Baker, Dr. Harold Heath,
Prof. E. C. Starks and Messrs. Olaf Jenkins, Earl Leib and George
Branner, members of the Stanford Expedition, as well as Prof. Chas.
T. Brues of the Bussev Institution, all of whom have assisted me in
various ways.

PONERINAE.

1. Acanthostichits brevicornis Emery.

Tliree workers taken from beneath a deeply imbedded stone at
Independencia, agree closely with Emery's description of A. brevi-
cornis from Cayenne. The length varies from 3 to 6 mm. The head
is considerably longer than broad, with straight, parallel sides. The

402 bulletin: museum of comparative zoology.

antennal scapes are short and broad. There was only a small number
of workers together and all but three escaped.

2. Paraponera clavata (Fabricius).

Found commonly at Para, Manaos, and along the Rio Madeira
at Abuna, Porto Velho, and Camps 39 and 41. ^ 9.

Next to Dinoponcra grandis Guerin, this is the largest of the Brazil-
ian ants. It is much more widely distributed than D. grandis, occur-
ring from Central America to Paraguay. In habit it is diurnal. The
colonies are composed of a small number of individuals, which nest
in the ground, generally among the roots of trees or shrubs.

3. Platythyrea angusta Forel.

A series of workers from Porto Velho, Abuna; and Madeira-Mamore
Camp 39 agrees well with the description, but have the anterior femora
much swollen at the base. Dr. Forel recently showed me the type,
which also has enlarged femora. This species, originally described
from Trinidad, is more slender than the other South American forms,
and the petiole is twice as long as broad. These characters, and the
structure of the femora distinguish it from the others.

^O"^

4. Platythyrea incerta Emery.

A single worker from Madeira-Mamore Camp 41 agrees closely
with Emery's description. It differs from P. punctata in being much
larger (length 8 mm.) and in having the head shorter and the punc-
tuation considerably coarser.

5. Platythyrea meincrti Forel.

Plate 1, figs. 2, 3; Plate 7, fig. 53.

A colony of this species was found at Para in one part of a termite
nest, a favorite nesting place of the genus, which is probably largely
termitophagous in habit. This species is characterized by the strongly
bisinuate petiolar node and the large eyes, which are as long as their
distance from the anterior border of the head. A full-grown larva is
shown Plate 7, fig. 53.

MANN: THE ANTS OF BRAZIL. 403

The following table may help in separating the South American
species of Platythyrea.

A. Mandibles dentate.

Length 8 mm. (Surinam) sinnata Roger.

AA. Mandibles without teeth.

a. Eyes as long as their distance from anterior border of
head; petiole strongly bisinuate, the middle portion pro-
longed into a short beak. Length 7.5 mm. (Venezuela,

Brazil) meinerti Forel.

aa. Eyes smaller, petiole feebly sinuate.

b. Petiole from above more than twice as long as broad,
as long as first segment of gaster ; femora swollen. Length
6.7 to 7.7 mm. (Trinidad, Brazil) angusta Forel.

bb. Petiole from above less than twice as long as broad,
femora normal.

c. Punctures on epinotum very coarse, dilatation of frontal
lobes more marked. Length 8 mm. (Venezuela, Brazil).

incerta Emery.

cc. Punctures on epinotum finer, dilatation of frontal

lobes less marked. Length 6 to 7 mm. (Antilles, Central

and South America) 'punctata Smith.

6. Typhlomyrmex rogenhoferi Mayr.
One small colony, taken from a rotten log at Para, y .

7. Rhopalopone relida, sp. nov.
Plate 1, fig. 4, 5.

Worker. (Plate 1, figs. 4, 5). Length 2.5 mm.

Head, excluding mandibles, distinctly longer than broad, broadest
behind eyes, the width at occiput about equal to that at base of cly-
peus; sides slightly rounded, posterior border feebly concave. Man-
dibles long, rather slender, the blade minutely dentate; outer border
arcuate at middle. Clypeus subtriangular, broadly rounded at ante-
rior border, flattened in the middle. Frontal carinae very short, their
basal lobes rounded. Antennae robust; scapes arcuate, extending
barely to the occipital corners ; first funicular joint as long as the two
succeeding joints together; joints 3 to 9 broader than long; apical

404 bulletin: museum of comparative zoology.

joint twice as long as the penultimate. Eyes small, convex; situated
at middle of sides of head. Prothorax slightly transverse, rounded
dorsally and laterally. Promesonotal suture faintly impressed. Meso-
epinotal suture not discernable. Epinotum in profile slightly convex
above; its base rounding into the declivity and equal to it in length.
Petiole from above transverse, sides rounded; in profile deeper than
thick, anterior surface nearly straight, the posterior concave; ventral
surface with a large flat tooth anteriorly. Gaster about as long as
the thorax and epinotum together; first and second segments subequal,
the former with a large anteroventral tooth. Legs robust; posterior
coxa armed with a large curved spine.

Sublucid. Head regularly, longitudinally striate and rugose.
Thorax and epinotum shining, sparsely punctate, their pleurae
coarsely striate transversely. Epinotal declivity transversely striate.
Petiolar node with transverse rugae. Gaster with coarse longitudinal
striae. Legs finely punctate and shining.

Head, antennae, body, and legs with fine short pilosity.

Color brownish red ; legs, mandibles, and antennae brownish yellow.
Pilositv white.

Described from several workers taken at Madeira-Mamore Camp
39.

This distinct species is the first Rhopalopone to be recorded from
America, the four other known species inhabiting Borneo and New
Guinea.

8. Holcoponera striatula Mayr.

Very common in the vicinity of Natal, where it was nesting beneath
stones and logs, in rather large colonies. One colony was found at
Para. S 9.

9. Holcoponera moUcri Forel.

A single worker from Madeira-Mamore Camp 39 agrees with
Forel's description of this species from Blumenau. It is larger than
H. striatula, the mandibles are more coarsely striate and all the funicu-
lar joints are distinctly longer than broad. In its other characters
it is very similar to H. striatula.

MANN: THE ANTS OF BRAZIL. 405

10. Edatomma {Edatomma) quadridens (Fabricius).

Plate 7, fig. 54.

This species was very common at Independencia, Ceara-Mirim,
Para, Itacoatiara, and Manaos. In a nest excavated at the first
named locality, the brood chambers were about two and a half feet
beneath the surface, in very hard dry earth. Although E. quadridens
is very common in collections, the male appears to be undescribed.

Male. Length 10 mm.

Head, excluding mandibles, as broad as long, with rounded occipital
border. Cheeks two thirds as long as eye. Mandibles well developed,
shaped like those of the worker, but smaller. Eyes and ocelli large and
convex. Antennae slender, the scape thick and short, about half as
long as eye; first joint of the funiculus one third as long as the scape,
joints 3-12 cylindrical, four to five times as long as broad; apical joint
one and a third times the length of penultimate. Pro- and mesothorax
rounded above and at sides. Epinotum rounded ; the declivous sur-
face broad and flat, feebly marginate at base, with small lamellate
tubercles. Node rounded, transverse, its height about equal to its
length; anterior surface flat; with a tubercle anteroventrally.

Head and thorax opaque, rugulosely striate. Mandibles coarsely
striate. Antennae finely punctate. Node and first two segments of
gaster subopaque, the latter densely striolate longitudinally.

First four joints of antennae sparsely, the rest thickly pubescent,
with a few short, erect hairs. Head and thorax with pubescence and
a few erect hairs. Node devoid of pubescence, but pilose. Gaster
sparsely pubescent and abundantly pilose. Legs finely pilose.

Color black; genitalia brown. Wings (length 6.5 mm.) infuscated;
veins and stigma fuscous.

The larva is shown Plate 7, fig. 54.

IL Edatomma {Edatomma) ruidum (Roger).

Less abundant than the preceding species. Found at Ceara-
Mirim, Manaos, and Camps 39 and 41 Madeira-Mamore RR. ^ .

12. Edatomma {Edatomma) tuberculatum (Olivier).

Common at Para, Itacoatiara, Manaos, and along the Rio Madeira.

S 9.

406 bulletin: museum of comparative zoology.

Workers are often seen moving slowly about or remaining motion-
less on the stems and leaves of trees and shrubs. The species is
partly nocturnal in habit and often enters houses and hunts about for
other insects that come to the lights. A nest which I dug out was in
the ground among the roots of a plant, about twelve inches below^ the
surface.

13. Edatonima {Eciatomma) confine Mayr.
Plate 1, fig. 8.

A single worker (Plate 1, fig. 8), which agrees well with Mayr's
description, was taken at Porto Velho. Apparently the species has
not been recorded since Mayr described it from a Colombian specimen.

The structure of the pronotum is very characteristic. The middle
tubercle is prolonged and flattened above, and laterally compressed
in front of the pleural spines so that it has the form of a short, thick
carina. The lateral spines on the pronotum are short, flattened, and
triangular; the epinotal spines are prominent. The sculpture of the
head is coarse, that of the rest of the body delicate. The head,
thorax, and abdomen are sparsely beset with coarse, erect hairs.

14. Ectatomma (Gnaviptogenys) concinnum (F. Smith).

Plate 1, fig. 7.

Workers (Plate 1, fig. 7) were found at Porto Velho, Abuna, and
Madeira-Mamore Camp 39.

15. Ectatomma (Gnamptogenys) tortuolosum F. Smith.
Plate 1, fig. 6.

Worker. (Plate 1, fig. 6.) Length 7 mm.

Head, excluding mandibles, a little longer than broad, with slightly
convex sides, narrowly rounded posterior corners and concave border.
Clypeus nearly as long as broad, the surface depressed; anterior
border straight. Mandibles slender, the blade edentate. Antennae
robust; scape thickened apically, extending one fourth its length
beyond the occipital corners; funicular joints 1 to 3 elongate, sub-
equal; joints 3-6 globose, a little longer than broad. Eyes moder-
atel}' large, convex, situated at middle of sides of head. Thorax

MANN: THE ANTS OF BRAZIL. 407

above without sutures. Prothorax rounded above in front and at
sides. Epinotum in profile nearly straight at base, the declivity
sloping gradually. Petiole longer than broad, broadest behind, with
rounded posterior border, straight sides and straight, margined
anterior border, the corners of which project angulately; in profile
higher than long, rounded above. Gaster short and thick. Legs
rather slender.

Head shining, longitudinally striate, the striae coarse and regu-
lar, becoming oblique at sides, perpendicular on cheeks; the
intervening ridges rounded. Thorax shining, sculptured similarly
to head; the prothoracic striae longitudinal at middle, arcuate at
sides, forming a somewhat concentric pattern. Striae of epinotum
transverse at middle, obliquely longitudinal at sides. Petiole con-
centrically striate. Gaster subshining, the first two segments longi-
tudinally striate.

Head, body, and legs with abundant, rather fine erect hairs.

Color piceous, legs and antennae fuscous.

The specimen before me has the antennae mutilated.

Described from a worker found dead, at Madeira-Mamore Camp
41. The specimen runs to E. tortiiolosum in Emery's key to the
species, and answers to Smith's very superficial description, but
differs from the form considered as this species by Emery (Studi sulle
formiche della fauna Neotropica. Bull. Soc. ent. Ital., 1896, 28, p. 51)
in the structure of the petiole, which forms an acute angle above in
profile in E. tortuolosum. It is doubtful whether this or Emery's speci-
men (which came from Para) belongs to Smith's species.

16. Edaiomma (Gnamptogcnys) sulcatum (F. Smith).

Several workers and females from Porto Velho. The striae extend
the entire length of the thorax; those of the node and first gastric
segment are also longitudinal, but slightly arcuate on the sides. The
color is bright ferruginous, with the head black and the mandibles.
pale yellow.

17. Ectatomma {Gnamptogcnys) sulcatum var. nitens, var. nov.

Three workers from jManaos and Independencia.
In form and sculpture this variety is identical with the typical form,
but is entirely black in color, except the mandibles which are yellow.

408 bulletin: museum of comparative zoology.

18. Ectatonima (Gnamptogenys) annulatuvi Mayr.

Not uncommon at Porto Velho, and Camp 39, Madeira-Mamore
R. R., where several females and workers were taken. Two workers
from Para are also in the collection. A nest was found in a cavity of a
fence post, where the wood was rotten.

The striolation of E. amiulatum is very delicate, longitudinal on the
pro- and mesonotum, transverse on the epinotum and petiolar node.
The specimens vary in color from fuscous to ferruginous. The legs
of all are yellowish brown, with a fuscous blotch at apex of tibiae and
femora. v

19. Dinoponera grandis (Guerin).

Several specimens of the typical form of the species, as designated
by Emery (Ann. Soc. ent. Belg., 1901, 45, p. 47) were taken at Para,
in the suburb Souza. The subopaque gaster, abundance of hairs on
the body and the well-developed ventral tooth of the pronotum, are
characteristic of this form. The petiole in profile is quadrangular, as
high behind as in front, and the thorax and gaster bear fine recumbent
pubescence, which gives a brownish appearance to the body.

20. Dinoponera grandis subsp. lucida Emery.

Two workers from Porto Velho.

This subspecies differs from the typical D. grandis in having the
prothorax, node, and gaster shining; the petiole from above is more
slender and in profile has the upper surface convex instead of nearly
straight; the prothoracic spine is lacking; the pubescence of the thorax
is more dense.

21. Dinoponera grandis subsp. mutica Emery. .
Plate 7, fig. 55.

Abundant at Natal, Baixa-Verde, Ceara-Mirim, and Independencia.
The prothoracic spine is absent; the node shorter than in the typical
D. grandis, in profile pointed in front, rounded behind. The body is
more shining, the pilosity and pubescence less conspicuous.

There were several colonies of D. mutica in the vicinity of our house

MANN: THE ANTS OF BRAZIL. 409

at Natal, nesting among scrubby vegetation. The typical D. grandis,
in the forest, is seen foraging all through the day, but D. mutica, living
in more open localities, is crepuscular or nocturnal, though it forages
also on cloudy days. The formicaries were always in thickets, among
the roots of trees. The mounds thrown up are low, generally not over
six inches in height, and often up to three feet in diameter. Dr.
Heath and I dug out one nest. The tunnels extended along the vmder-
side of roots, which formed projecting roofs. Along these tunnels
were frequent broad and flat chamljcrs, which contained the brood.
In spite of the large size and powerful sting, the ants were not very
pugnacious, though those in a chamber would sally out when it was
cut into.

Dinoponera grandis and its varieties are known to the Brazilians as
the "Tocandero," and according to them its sting causes fever.

A larva (Plate 7, fig. 55) probably immature, in alcohol measures
13 mm. in length. The body is thick and the neck short. All the
segments are distinct, with fine, short hairs. The head is glabrous,
from above a little broader than long; the mandibles long and acumi-
nate. The thorax and abdomen are tuberculate, the tubercles verv
large and prominent, rounded above, each bearing a small sensory
papilla at the middle. Each segment has three of these large tubercles
laterally, and a smaller, less conspicuous one basally.

22. Dinoponera grandis var. or subsp.

Male. Length 21 mm.

Head, including the mandibles, as broad as long, very convex behind.
Eyes very large and long occupying the entire sides of head, the inner
border deeply emarginate; ocelli very large and convex. Clypeus
convex, the anterior border truncate. Mandibles small, pointed at
apex, with a small tooth at middle of inner border. Antennae a little
shorter than the body; first funicular joint twice as broad as long;
joints 2-11 very long, cylindrical, each slightly shorter and more
slender than the preceding. Thorax robust; scutellum short, tri-
angular, broadly rounded at apex. Epinotum evenly rounded, with-
out distinct base or declivity, unarmed. Petiole nearly twice as long
as broad, narrowed in front, with nearly straight sides; in profile
longer than high, flattened above, the anterior slope gradual, more
abruptly sloping behind, the anteroventral surface with a broad,
triangular projection. Gaster long, and slender, the length three

410 bulletin: museum of comparative zoology.

times the breadth. Genitalia prominent; the valves broad, rounded
at apex; cerci long and slender. Wings large, extending almost to
tip of gaster. Legs very long and slender.

Body and legs shining. Antennae opaque, coarsely, densely
punctured; sparsely pubescent, and having much very long, fine
erect hairs, which on the apical joints are shorter and confined to the
tips; pubescence of apical joint more dense than the rest. Thorax
with long silky pubescence, most abundant on the pleurae, and very
fine erect hairs sparsely distributed. Node without pubescence, but
with abundant erect hairs. Gaster with a thin mat of silky pubes-
cence, shorter and finer than that of the thorax; lateral and apical
portions with fine erect hairs.

Color rufous, the antennal scape and first five funicular joints
fuscous. Wings slightly infuscated, veins and stigma reddish brown.
Pile and pubescence yellowish white, except the long antennal hairs
which are black.

Described from three examples, which were taken at lights in
Independencia. This form which is probably the male of the variety
D. midica, the commoner form in this vicinitv, is verv much like a
male thynnid in general habitus. The antennae bear unusually
long hairs, which are abundant basally, but thin out and become
shorter toward the apex.

23. Neoponera {Neoponera) commutata (Roger).
Porto Velho and Camp 39, Madeira-Mamore R. R.

24. Neoponera (Neoponera) apicalis (Latreille).

Para, Porto Velho, Abuna, and Camps 39, 41, Madeira-Mamore
R. R.

25. Neoponera {Neoponera) obscuricornis Emery.

Abuna and Porto Velho. ^

In general appearance this species is very similar to the preceding.
It can be distinguished by the structure of the node. In N. apicalis
this is sharply margined at the sides, and longitudinally impressed
medially to the margin. Neoponera obscuricornis has the surface of
the node slightly convex, not impressed; the margin is very feeble
at the base and entirely obsolete on the apical half.

MANN: THE ANTS OF BRAZIL, 411

26. Neoponera (Neoponera) latreillei Forel.

Para, Porto Velho, and Camps 39, 41, Madeira-Mamore R. R. y
The large size and active movements, with the contrast in color

between the black body and yellow tipped antennae make N. latreillei

one of the most conspicuous ants in the forest.

A small colony, of perhaps thirty individuals, found at Para, was

nesting beneath a banana stalk in a dense thicket. The ants were

very timid and ran away cjuickly, but returned one by one to carry

off their brood.

27. Neoponera {Neoponera) unidentata (Mayr).

Para, Manaos, Porto Velho, and Camps 39, 41 Madeira-Mamore
R. R. y 9

28. Neoponera (Neoponera) hakeri, sp. nov.
Plate 1, fig. 9.

Worker. (Plate 1, fig. 9). Length 8 mm. Near N. unidentata
(Mayr) . Head, excluding mandibles, longer than broad, subquadrate ,
sides feebly convex, posterior angles obtuse; occiput concave; carina
on cheek strong; clypeus at middle prolonged into a sharp point, con-
cave at sides. Mandibles rather thick, triangular, with twelve small
teeth, the apical and two subapical larger than the others and subequal.
Eyes convex, a little in front of the middle of head. Antennal scapes
extending three eighths their length beyond the occipital corners;
apical joint of flagellum nearly twice as long as penultimate. Thorax,
seen in profile, moderately convex. Pronotum convex, strongly
margined at sides, the margin indistinctly serrate. Mesonotum nearly
circular, disc-like. Mesoepinotal impression shallow, but distinct.
Metanotum convex basally; declivity gradual, the surface nearly (flat,
on either side with a prominent marginal carina on which are three
distinct triangular teeth. Node highest in front; anterior surface
straight to apex; posterior surface, seen in profile, strongly convex,
base slightly concave, margined at sides, margin with three to five
distinct teeth. Gaster as long as thorax and head excluding mandibles.

Head coarsely, longitudinally striate, the striae irregular and
wavy; mandibles sublucid, smooth. Pronotum more finely trans-
versely striate. Striae of meso- and epinotum coarse, on mesonotum

412 bulletin: museuim of comparative zoology.

and disc of epinotum transverse, on sides longitudinally oblique.
Node shining, with coarse, transverse striae, which extend around it,
broken only by the strong margin. Gaster sublucid; first segment
minutely punctate.

Head and thorax with fine recumbent pubescence, and abundant,
scattered, erect pile. Node without pubescence, and with scarce
pilosity. Gaster with a dense mat of fine recumbent pubescence, and
abundant, scattered, erect pile.

Femora and tibiae with short erect hairs, smooth and shining.

Color black; mandibles, antennae, tibiae, and tarsi brown, femora
yellow. Pile and pubescence pale yellow.

Female. Length 11 mm.

Similar to worker. The eyes are small, ocelli minute. The thoracic
striae are proportionately much finer than in the worker, and trans-
verse. The declivity of the epinotum is abrupt, the base shallowly
margined; margin with fine teeth. Node as in worker.

Pile and pubescence as in worker. Color black; legs, mandibles,
and extreme tip of antennae brown.

Wings. Length 7.5 mm. Infuscated, veins and stigma fuscous.

Described from one female from Porto Velho and several workers
taken on the Rio Madeira at Porto Velho, and Camps 39 and 41 on
the Madeira-Mamore R. R. This species in form is similar to N .
unidentata (Mayr), which was common in the same region, but the
very distinctive coarse sculpture of the head, thorax, and especially
of the node, and the denticulate margins of the epinotal declivity and
the posterior surface of the node, as well as the different nodal struc-
ture seem to constitute differences more than subspecific.

29. Neoponera {Neoponera) villosa (Fabricius).

Para, Manaos, Itacoatiara, Porto Velho, Abuna, and Camps 39, 41
Madeira-Mamore R. R., Brazil; and Abuna, Bolivia.

This is one of the commonest ponerine ants of Brazil, and one of
the most widely distributed species, ranging from Texas to Paraguay.

30. Neoponera (Neoponera) carinulata (Roger).

Plate 1, fig. 10.

Worker. (Plate 1, fig. 10). Length 8 mm.

Head, excluding mandibles, scarcely longer than broad, very slightly
narrowed in front; occipital border nearly straight; sides feebly

MANN: THE ANTS OF BRAZIL. 413

convex; clypeus prolonged at middle, the apical portion narrowly
rounded; frontal carinae above antennal lobes slightly diverging out-
wardly. Mandibles triangular; about | as long as head, the blade
with twelve distinct teeth. Carinae on cheeks very distinct. Anten-
nae slender; the scapes extending one third their length past occipi-
tal corners; funicular joints all longer than broad, increasing in size
toward the tip. Prothorax transverse, longitudinally carinate at
middle, flattened above; sides strongly margined, in profile slightly
convex. Mesonotum circular, with a median, longitudinal carina,
on either side of which the surface is depressed; in profile flat. Epi-
notum in profile evenly convex, with no appreciable angle between
base and declivity; anterior portion rounded above and at sides;
declivity margined, with an indistinct tubercle at middle of margin;
surface flat. Petiole as broad as long, broadest behind, sides evenly
convex; posterior margin straight, in profile three fourths as thick
as high, with anterior surface horizontal from base to two thirds the
distance to apex, then inclined to apex; posterior surface with a disc
at middle, rounded on the sides. Gaster cylindrical, as long as thorax
and petiole together.

Somewhat shining. Mandibles finely striate. Head foveolately
punctate; thorax and node finely punctate, the epinotal pleurae strio-
late.

Head, thorax, and node with sparse, fine, appressed pubescence
and scattered, erect hairs. Posterior surface of node smooth and
shining. Gaster rather densely pubescent and with scattered, long
pile. Legs sparsely pubescent and pilose.

Color black; antennae, mandibles, and legs brown. Pile and
pubescence silvery.

Described from two workers taken at Abuna.

31. Neoponera (Neoponera) crenata Roger subsp. moesta Mayr.
One colony was found at Porto Velho.

32. Neoponera (Neoponera) stipitum Forel.

A single worker from Camp 41, Madeira-Mamore R. R. agrees
very well with a cotype from Colombia in the Wheeler collection.

414 bulletin: museum of comparative zoology.

33. Neoponera (Neoponera) cavinodis, sp. nov.
Plate 2, fig. 14.

Worker. (Plate 2, fig. 14). Length 10 mm.

Near N. lutcoJa (Roger). Head, excluding mandibles, longer than
broad; sides convex, base slightly concave, posterior angles rounded;
clypeus extended at middle into a triangular projection, the sides
arcuately impressed. Mandibles elongate, acute at apex, with twelve
teeth, the basal of which are small and those apically larger; the apical
tooth is long and pointed, attached at base to this is a smaller sharp
tooth, about half as long, and basal to this another, slightly longer
and more acute than the subapical. Eyes convex, situated a little
in front of middle of head. Scape of antenna extending three eighths
its length past occiput; joints 1-10 of flagellum subequal in length,
gradually' thicker towards the apex; apical joint 1| times length of
penultimate. Pronotum slightly convex, sides feebly margined,
forming an angle with the nearly flat pleurae; disc at middle with a
shallow, but distinct longitudinal impression. Mesoepinotum in
profile very slightly convex, rounded at sides posteriorly, and in back,
the mesoepinotal suture scarcely distinguishable: declivity of epino-
tum rather flat, with feeble margin at sides; above forming a rounded
angle with the basal portion ; seen from behind spear-shaped, the apex
with a short longitudinal impression. Petiole higher than thorax,
anterior surface evenly rounded to apex, the middle of base very
slightly convex; apex projecting backward over the posterior surface;
posterior surface receding anteriorly, concave above, slightly convex
below. Gaster as long as head and thorax.

Mandibles subshining, with scattered, coarse punctures. Head sub-
opaque, very finely transversely striate and with minute punctures.
Thorax and anterior surface of petiolar node sublucid, finely punctate
and delicately reticulate. Gaster finely punctate. Legs shining. Head
and thorax with fine pubescence and long scattered pile. Clypeus with
several long hairs. Anterior surface of node with appressed pubes-
cence, the posterior surface glabrous. Gaster thickly pubescent and
with long scattered hairs. Femora and tibiae with suberect hairs.

Color brown, head and gaster darker than the thorax. Legs yellow.

Pile and pubescence pale yellow.

Described from a worker taken at Porto Velho.

In general appearance N. cavinodis resembles N. crenata but is much
larger and the node is entirely different.

MANN: THE ANTS OF BRAZIL. 415

34. Pachycondyla crassinoda (Latreille).
Taken at Para, Porto Velho, and Madeira-Mamore Camp 39.

35. Pachycondyla harpax (Fabricius).

Common at Para Manaos, Porto Velho, Abuna, and Camp 41,
INIadeira-Mamore, R. R.

36. Euponera (Trachymesopus) stigma (Fabricius).

This species, which is common everywhere in the American tropics
is represented in the collection by worker^ and females from Para,
Itacoatiara, Manaos, Porto Velho, Abuna, and Camps 39 and 41,
Madeira-Mamore R. R. Rotten logs and beneath bark are its
favorite nesting places.

37. Belonopelta jeckylli, sp. nov.
Plate 2, fig. 12, 13.

Worker. (Plate 2, fig. 12, 13). Length 4.5 mm.

Head very thick dorsoventrally ; excluding mandibles, a little
longer than broad, broadest anteriorly, with straight sides, narrowly
rounded occipital corners and feebly concave occipital border. Cly-
peus convex; at anterior border armed with a stout spine at middle.
Mandibles five eighths as long as head, slender, bidentate at apex;
the inner border with a large tooth at middle, and a minute one mid-
way between this and the subapical tooth. Antennae robust; the
scapes slightly arcuate, thickened distally, almost attaining occipital
corners. First funicular joint longer than broad, joints 2-10 slightly
transverse, those at the apical end subglobose; apical joint as long
as the three preceding together, evenly narrowed from base to apex.
Eyes minute, situated at anterior third of sides of head. Prothorax
with a thick neck. Pronotum from above, excluding the neck, a
little broader than long, rounded in front and at sides, the. posterior
border concave. Promesonotal suture very strongly impressed.
Mesonotum about half the width of pronotum, rounded in front,
sides nearly straight, in profile slightly convex. Mesoepinotal suture
feebly impressed. Epinotum twice as long as the mesonotum, and a

416 bulletin: museum of comparative zoology.

little broader, broadest behind, sides nearly straight; in profile
gradually elevated from base to angle of declivity; declivity half as
long as the basal part, with the surface flattened. Petiole from
above very little broader than long, anterior margin straight, at
corners of base with prominent lamellate projections, that appear as
spines from above; sides rounded, posterior border feebly rounded;
in profile deeper than thick, with nearly straight anterior and posterior
surfaces, slightly convex above. Gaster cylindrical, strongly nar-
rowed toward apex; the first segment noticeably broader than the
second and equal to it in length; a distinct ventral spine present at
base. Sting comparatively large. Legs very slender.

Head, thorax, and petiole subopaque, closely, foveolately punctate
throughout; pleurae of pro- and mesothorax and anterior coxa
obliquely rugulosely striolate; collar transversely rugulose-striolate.
Mandibles shining, finely punctate. Antennae and legs subshining,
finely punctate. Gaster shining, minutely punctate.

Body, legs, antennae, and mandibles with semierect pilosity.

Head, mandibles, thorax, and petiole fuscous to piceous; borders
of the frontal lobes, a round spot on the pronotum, the gaster and
legs red. The borders of the gastric segments are infuscated.

Described from several specimens taken at Camp 39, Madeira-
Mamore R. R. The colony was discovered quite accidentally by
scratching away some of the leaves and debris with which the ground
in the forest is always covered. As far as I could ascertain the ants
were traveling in a definite direction. Some of those before me are
red in color, evidently immature. Nothing seems to be known re-
garding the habits of the species of this interesting genus, specimens
of which are rare in collections. The four known species have each
been found only once.

38. Ponera trigona Mayr.
One small colony was found at Manaos.

39. Ponera oyaciceps Mayr.
Several colonies were taken at Ceara-Mirim.

40. Ponera distinguenda Emery.
Numerous workers were taken at Camp 39, Madeira-Mamore R. R.

MANN: THE ANTS OF BRAZIL. 417

41. Anochetus (Anochetus) mayri Emery.

A single worker of this widely distributed species was taken at
Baixa Verde.

42. Anochetus {Anochetus) hispinosus (F. Smith).
Plate 1, fig. 11.

JVorker. (Plate 1, fig. 11). Length 8 mm.

Head, excluding mandibles a little longer than broad; broadest at
eyes; occipital border deeply excised; sides above eyes concave,
below sinuate, with the lower surface flattened. Eyes large, and con-
vex. Clypeus between frontal lobes triangular, the surface flat,
anterior portion bilobed; border concave at middle. Frontal carinae
short. Mandibles about two thirds as long as head, narrow at base,
broad and flat anteriorly, the blade edentate to near apex, where
there is a short, thick tooth; apex with two long teeth, between which
is a smaller, acute tooth. Antennae slender, the scape curved,
thickened at middle, extending one eighth its length beyond the
occipital corners; funicular joints long and slender, 1 to 10 subequal,
apical joint nearly twice the length of subapical. Prothorax much
narrowed in front, rounded above and at sides, the posterior border
slightly concave. Promesothoracic suture deeply impressed; meso-
thorax transverse, rounded above and at sides. Mesoepinotal im-
pression deep and broad; epinotum much narrower than prothorax,
straight at sides and above to the declivity, which is very short, and
concave; angle with a pair of short, acute, erect spines. Node
triangular, the apex with a pair of strong spines, between these
broadly concave. Gaster short and thick, the first segment with
a small ventral tooth. Sting long and powerful. Legs long and
slender.

Head in front subshining, with dense striolae, which extend upward
and outward from the frontal area; the remainder of head and
the mandibles smooth and very shining. Antennae sublucid, finely
pubescent. Thorax shining, rugosely punctate; the mesothoracic
pleurae smooth. Node shining, transversely rugose in front, smooth
behind. Legs and gaster shining, finely punctate.

Head in front with sparse pubescence, and a ^ew erect hairs. Pro-
thorax, gaster, and coxae with very sparse erect gray hairs. Femora
and tibiae thinly pubescent.

Color black, the mandibles, posterior corners of head, and the nodal

418 bulletin: museum of comparative zoology.

spines brownish; legs and tip of gaster ferruginous. Pubescence
white.

Described from a single worker taken at Porto Velho. The speci-
men on which Smith based his description was from Ega (Bates), so
it is probable that the species extends over much of the Amazonian
region.

43. Anochetus (Stenomyrmex) emargmahis (Fabricius) subsp.

rucjosiis Emery.

Male. Length 7 mm.

Head, excluding eyes, considerably longer than broad, rounded
behind. Eyes very large, each three fourths as broad as the distance
between them, convex, as long as sides of head. Ocelli very large and
convex. Mandibles, short, feeble, straight, and without teeth.
Clypeus truncate at anterior border; surface broadly foveolately
depressed at sides. Antennae very long and slender; the first joint
twice as long as the second; joints 3-11 cyHndrical, subequal, very
long. Thorax narrower than head. Promesonotum convex above.
Epinotum in profile evenly rounded. Node longer than broad, in
profile triangular, much longer than thick. Gaster long and slender,
without a distinct constriction between the first and second segments.
Legs long and slender.

Body shining throughout, smooth, with extremely fine semierect
pilosity, which is thickest on the antennae. Head and mandibles
with a few coarser hairs.

Color light testaceous; antennae, excepting first joint, fuscous;
eves black. Pilositv white.

Described from several specimens taken at Manaos, from colonies
which were nesting beneath the bases of living palm-tree leaves.
Other colonies were found at Porto Velho beneath logs. The male is
very active and takes flight readily.

44. Odoniomachus affinis Guerin subsp. 7nayi Mann.

The single colony of this form was found living in parabiotic rela-
tions with Dolichodenis debilis Emery var. nifescens Mann, an account
of which has already been published.

45. Odontomachus hacmatoda (Linne).
Para, Manaos, Porto Velho, and Camp 41, Madeira-Mamore R, R.

MANN: THE ANTS OF BRAZIL. 419

46. Odontomachus haematoda subsp. imbescens Roger.

Very common at Ceara, in the Maranguape Mts., Independencia,
Natal, and Ceara-Mirim. This form is distinguished from the typical
0. haematoda by the more abundant pile and pubescence, and by the
coarser sculpture of the node, which is only feebly striate in the latter
form.

47. Odontomachus haematoda subsp. laticeps Roger.

Porto Velho and Camp 41, Madeira-Mamore R. R.
This was much rarer than the above mentioned forms.

DORYLINAE.

48. Eciton {Eciton) hamatiim (Fabricius).

This species, which ranges from Mexico through Central America
and over all of tropical South America, was very abundant in the region
of the upper Rio Madeira. The armies, found in the woods almost
every day, contained enormous numbers of individuals, mostly minors
and mediae. One of the big-headed soldiers was encountered at
intervals of from ten to twenty feet in the procession. This form is a
very conspicuous object on account of its large, light-colored head and
the long mandibles which necessitate the body being held high.

The march of the army is exceedingly rapid, and at times very
definite in direction. Often it divides and sends some branches up
into the tallest trees, while others cross and recross the trails. If an
object such as a grub, lizard, or small snake be thrown near the column,
it is almost instantly covered with the workers, which bite and sting
severely. Unlike some of the other species, E. hamatuvi marches in
the daytime and the column travels beneath or over the leaves, over
logs, and along the trails. The trunks of fallen trees are a favorite
runway. Other species of ants seem to be the usual prey of E. hama-
tum, for larvae and pupae of these made up the greater part of the
booty carried by the workers. I observed on several occasions
columns descending trees bringing numbers of larvae, pupae, and even
adults of Dolichoderus lugens Emery, an ant which secretes from the
anal glands a large drop of yellow liquid, apparently for defense.

In spite of its large size and the number of individuals in a column,
E. hamatum is a timid species in comparison with some of the others,
such as E. Taga7is. When the column was disturbed by my picking

420 bulletin: museum of comparative zoology.

up some of the individuals, those nearest would turn about and run
back, zig-zagging from one ant to another, apparently missing none.
An instantaneous antennal communication took place, the warned ant
turned also, and almost instantly the whole army was retracing its
steps as rapidly as it had come. In a few moments some few would
return and then more and in a short time the army would resume its
march. At other times a new path was followed. This antennal
communication and quick change of direction was observed also in
E. pilosum, E. crassicorne, and E. burchelli.

49. Eciton {Eciton) lucanoides Emery.

Columns were found at Madeira-Mamore Camps 39 and 41.

This species was originally described from Peru. The mandibles
are armed with a strong spine on the inner side, a little in front of the
middle. I am unable to distinguish the minors and mediae of E.
lucanoides from those of E. hamatum by any single character.

50. Eciton (Eciton) burchelli (Westwood).

This species was moderately abundant on the Rio Madeira, where it
was encountered several times at Abuna and at Camp 41, Madeira-
Mamore R. R.

It is rather more aggressive than the other species. When dis-
turbed it attacks the intruder very fiercely and persistently. The
pain produced by the sting is severe, but of short duration.

51. Eciton {Eciton) rapax F. Smith.

Two small files of this beautiful species were found at Porto Velho
and Camp 41, Madeira-Mamore R. R. These were running along
beneath the loose covering of leaves on the ground.

The ants were very timid and the file dispersed as soon as it was
disturbed.

52. Eciton {Eciton) rogeri Dalla Torre.
Several colonies were taken at Camp 39, Madeira-Mamore R. R.

53. Eciton {Eciton) vagans (Olivier).

Found at Independencia and Baixa Verde, where files were fre-
quently seen toward evening.

MANN: THE ANTS OF BRAZIL. 421

54. Ecitoii (Labidus) coecum (Latreille).

Ceara, Baturite Mts., Natal, Independencia, and Para, Brazil and
at Abiina, Bolivia. This is the most widely distributed and one of the
commonest species. It is subterranean in habit, and often found
beneath deeply embedded stones and logs. The specimens from
Abuna were taken from beneath the putrid carcass of a sheep, and
most of the individuals were dead or nearly so, possibly having been
overcome by the gases of decomposition.

55. Eciton (Labidus) praedator F. Smith.

Common along the upper Rio Madeira, where it was taken at Porto
Velho and Camps 28 and 39 Madeira-Mamore R. R. One army was
seen emerging from the commissary building at the first named sta-
tion, carrying with it an incredible number of insects, mostly cock-
roaches. Houses along the railroad were frequently raided at night
by E. prcedator, which is well known to the Brazilians and called by
them "cazadoro" (hunter). I had the opportunity of observing one
hut while the ants were in possession. The ground was covered with
the ants, which swarmed also in the cracks and on the few pieces of
furniture, while the owner of the place, a Barbados negress, not
accustomed to such intrusions, stood for safety in a puddle of soapy
water with which she had attempted to drive the ants away, and
begged me to tell her what to do to get rid of them.

56. Eciton (Labidus) praedator subsp. emiliae, subsp. nov.

Worker. Differing from the typical form in having the head and
body largely opaque; the head more densely punctulate; the thorax
and petiole with very few hairs. The pilosity of the gaster is much
shorter and the pubescence is more dense; the color is reddish brown,
becoming darker on the head.

A large series of this new subspecies was sent to me by Dr. Emilia
Snethlage, who collected it at Colonia de Veado, near Obidos. Possi-
bly this is the var. ferruginea Norton, which was described from
Mexico, but has not been recognized since.

422 bulletin: museum of comparative zoology.

57. Eciton (Labidus) crassicorne F. Smith.

Common at Ceara-Mirim and at Carnahubinha (near Natal) where
armies were frequently seen toward evening, either on the ground or
in the nests of ground-inhabiting ants.

58. Eciton (Labidus) esenbecki (\Yestwood).
A single male of this species was taken at light at Porto Velho.

59. Eciton (Labidus) sidcatum (Mayr).
Several males were taken at light at Independencia.

60. Eciton {Acamatus) nitcns Mayr.

A rather large colony of this distinct species, was found emerging
at dusk from beneath our house at Independencia.

61. Eciton {Acamatus) pilosum F. Smith.

Taken at Independencia and Para. Several armies were encoun-
tered. I collected a number of specimens from a large column which
was crossing a railroad track at Independencia, whereupon the column
broke up, but fgrmed again and continued its march, this time beneath
the rail. A column found in the woods at Para was travelling under-
ground, except where it crossed a path. Here it emerged and con-
structed a trail with embankments on each side.

62. Ecito7i {Acamatus) Icgionis F. Smith subsp. crenulatum,

subsp. nov.

Plate 1, fig. 1.

Worker. (Plate 1, fig. 1). Length 3.5 to 5 mm.

Head, excluding mandibles, longer than broad, with nearly straight
sides and rounded border; posterior corners angulate. Frontal
carinae nearly straight. Outer border of antennal pits strongly
carinate. Anterior margin of clypeus flat, projecting at middle.
Mandibles stout; the blade edentate. Antennae robust; scape
extending a third its length beyond the occipital corners, constricted

MANN: THE ANTS OF BRAZIL. 423

near apex, then incrassate; funicular joints a little longer than broad.
Eyes distinct, convex. Pronotum twice as long as broad, flat above,
the lateral borders elevated into a rounded ridge, which is somewhat
thickened at middle. Mesonotum separated from pronotum by a
transverse ridge, elevated, concave at middle, strongly margined at
sides; mesial to the margin is a longitudinal furrow and on the upper
surface two parallel ridges. Base and declivity of epinotum subequal
in length, the former flat above with margined sides; surface of the
latter evenly convex. Petiole from above longer than broad; flat
above, with straight, roundly margined sides. Postpetiole a little
longer than broad, broadest behind, rounded above and at sides,
anteroventral surface with a distinct tooth.

Head, antennae, thorax, and epinotum sublucid; coarsely, granu-
losely punctate, throughout, except for a rounded area mesial to the
eye, which is finely punctate. Antennae, mandibles, petiole, post-
petiole, and legs more shallowly punctate. Gaster smooth and shining.

Head, body, antennae, and legs with long erect hairs; funiculus
pubescent.

Color black, antennae, and legs dark reddish brown. Pilosity
yellow.

Described from several workers taken from a file that was running
beneath the loose bark of a felled tree at Madeira-Mamore Camp No.
39. This is a very distinct form because of its peculiar sculpture and
the strong carinae on the thorax. The latter, especially those at the
middle of the mesothorax, are interrupted so that in profile they ap-
pear as tubercles.

Myrmicinae.

63. Pseudomyrma arboris-sancti Emery.
Plate 3, fig. 21.

This species is widely distributed in northern South America, and
many observations have been made on its relations to the trees of the
genus Triplaris with which it appears to be always associated. The
tree is well known to the Brazilians and Bolivians by the name of
"palo santo," and the ant is called the "taschi." I was told that no
living tree was ever without the ants and that the ants never nested
elsewhere than in this plant. At Madeira-Mamore Camp 43
Triplaris was fairly common in the woods and I examined some

424 bulletin: museum of comparative zoology.

dozens, always finding the ant. The colonies contain an enormous
number of individuals living throughout the whole plant, all parts of
which are hollow. The workers are very aggressive and their sting
is quite severe. Whether or not the ants derive any advantage from
the tree other than a convenient place to nest I do not know, but
there is no doubt that they protect the plant from almost any possible
enemy. A Swiss rubber explorer. Otto Schmidt, who has spent many
years in the forest and is a keen observer, told me that dead Triplaris
plants never contain ant colonies. This suggests that the living
plants do offer some attraction to the ant, other than shelter.

64. Pseudomyrma ocidata F. Smith.

Many workers and females were found at Natal, nesting in hollow
twigs. This is one of the smaller, more delicate species, the workers
measuring 4 mm. in length. The head is two and a half times as
long as broad, the sides parallel with large, flat eyes. The antennal
scapes are short, extending only to anterior third of eyes. The petiole
is flattened above, twice as long as broad, in profile nearly twice as
long as thick. The color is dark fuscous throughout.

65. Pseudomyrma caroli Forel.

Many colonies were found at Itacoatiara, nesting in twigs on small
trees near the river.

66. Pseudomyrma nigriceps F. Smith.

Several workers were found at Abuna and Madeira-Mam ore Camp
41.

This species resembles P. rufa in structure and size. The color is
different and striking, the head being black, and the rest of the body
and appendages testaceous. It was not common.

67. Pseudomyrma fiavidula F. Smith.

Common at Independencia and Ceara-Mirim, nesting in grass culms,
the characteristic nesting site of the species.

MANN: THE ANTS OF BRAZIL. 425

68. Pseudornyrma rufa F. Smith.
Plate 7, fig. 56.

Worker. Length 6 mm.

Head, excluding mandibles, a little longer than broad, slightly
narrowed behind, with convex sides and straight posterior border.
Clypeus strongly keeled at middle; anterior border straight. Man-
dibles well developed, the blades finely dentate. Antennae thick;
first funicular joint two and a half times as long as broad; apical
joint as long as the two preceding. Ocelli distinct, very close together.
Pronotum transverse, depressed, strongly margined at sides, with
narrowly rounded humeri. Mesonotum longer than broad, the
surface with a disc-shaped impression at the posterior half. Meso-
epinotal impression shallow. Epinotum more elevated than pro-
and mesonotum, margined at sides basally, the surface distinctly
concave; declivity as long as base, in profile straight. Petiole in
profile elevated behind, above evenly rounded from base to apex,
nearly as high as long; from above, more than twice as long as broad,
narrowed above, with roundly margined sides and a narrow longitudi-
nal impression at middle; an tero ventral surface with a minute tooth;
posterior surface concave. Postpetiole from above a little broader
than long. Femora slightly thickened.

Subopaque, granulosely punctate, with very sparse gray pile;
pubescence not abundant, poUinose, white. Mandibles finely punc-
tate and pilose.

Color ferruginous, the meso- and epinotum infuscated.

Female (dealated). Length 7.5 mm. (Plate 7, fig. 56).

Head, excluding mandibles, one and a third times as long as broad,
the anterior and posterior borders of equal width; sides slightly con-
vex, posterior border straight. Clypeus keeled at middle, the ante-
rior border projecting, with a stout, triangular tooth at middle.
Mandibles stout, the blade with a stout, triangular, basal tooth, then
three minute teeth and long, pointed apical and subapical teeth.
Thorax slender. Pronotum narrower than mesothorax, twice as broad
as long, with subparallel, margined sides and narrowly rounded humeri.
Petiole seen in profile much as in worker; from above it differs in
having the surface flat, and not depressed except at the apex, where
there is a deep excavation dividing it into two triangular portions.
Sculpture and pilosity much as in the worker; the pubescence of
gaster very fine and silky.

426 bulletin: museum of comparative zoology.

Color as in worker, but the head is more infuscated.
Described from a female and several workers found in a twig lying
on the ground at Para.

69. Pseudomyrma gracilis (Fabricius).

Many colonies of what seems to be the typical form of this species
were taken from twigs at Manaos. The workers (7.5 mm. in length)
are colored black, excepting a narrow border at the anterior of head,
the clypeus, mandibles, the tips of the front, and intermediate femora
and the tarsi, which are ferruginous.

70. Pseudomyrma excavata Ma^T.

Three workers from Manaos agree with specimens from Costa Rica
received from Dr. Forel.

71. Pseudomyrma laevigata F. Smith.

One small colony found in a twig at Manaos. The specimens are
yellow in color, without trace of maculation.

72. Pseudomyrma mutilloides Emery.

A worker each from Itacoatiara and Camp 39 Madeira-Mamore
Railroad.

73. Pseudomyrma subtilissima Emery subsp. tenuissima Emery,

Taken at Natal and Maranhao. The colony from the latter locality
was taken from beneath a loose piece of bark, quite an unusual situ-
ation for Pseudomyrma. The type of P. tenuissima is from Matto
Grosso. Emery records a specimen also from Cayenne.

74. Pseudomyrma maculata F. Smith.

One colony from Natal is very doubtfully referred to this species.
The worker is 4 mm. in length.

Head large, broadest in front, sides convex, posterior margin trun-
cate, much broader than pronotum; antennae thickened at apex.

MANN: THE ANTS OF BRAZIL. 427

first funicular joint as long as second and third together. Metano-
tum evenly convex. First petiolar node elongate, slightly petiolate
in front, about 1| times as long as second node, which is globose.

Color yellowish brown, abdomen fuscous; epinotum with a faint
median longitudinal fuscous stripe.

75. Pseudomyrma clcgans F. Smith.

Found frequently in the vicinity of Manaos. This is the only known
ground-inhabiting species of the genus. The nests found were in
bare places along the roadside. The entrance is circular, about two
millimeters in diameter, and directly on the surface, without a mound.

76. Pseudomyrma elongata Mayr.

Many colonies found at Ceara-Mirim and Independencia. One
large tree at Ceara-Mirim contained hundreds of colonies nesting in
hollow twigs and in galls, scarcely a dead twig or a gall being without
them. The twigs inhabited are those hollowed naturally or bored by
other insects. Pseudomyrma elongata is decidedly beneficial to the
tree in keeping away insect enemies. The tree mentioned was
especially well protected, having several Azteca colonies and some
populous wasp nests in addition to the Pseudomyrma.

77. Pseudomyrma sericea Mayr var. altinoda, var. nov.

Worker. Length 4 mm.

Head one and one third times as long as broad, slightly broader in
front than behind, with convex sides and concave posterior border.
Clypeus with distinct notch at middle. Eyes large. Antennal scape
reaching one third its distance from place of insertion to posterior
border of head. Basal funicular joint not as long as two succeeding
joints taken together. Pronotum with well-defined margin. Pro-
mesonotal and mesoepinotal sutures equally pronounced. Mesono-
tum small, seen in profile lower than pro- and epinotum, transverse,
about one half the length of the pronotum. Epinotum longer than
pro- and mesonotum together, evenly convex above; its declivity
shorter than length of basal surface; angle between the two sur-
faces much rounded. Petiole two thirds as long as thorax, its nodes
higher than epinotum, equal in length; anterior node convex in front.

428 bulletin: museum of comparative zoology.

declivous behind; posterior node nearly globose, broader than base
of first segment of gaster. Gaster about as long as thorax. Legs
robust.

Body opaque, ever\'iv\iiere with a fine mat of pale pubescence, and
short, erect, pale pile.

Color fuscous, antennae and legs lighter. Differs from typical
P. sericea in smaller size and fuscous color. Pseudomyrma sericea is
black throughout, and the gaster is more robust.

Described from two workers from Porto Velho and one from Camp
39 Madeira-Mamore R. R.

78. Pseudomyrma gracilis Fabricius subsp. carapiina, subsp. nov.

Worker. Length 10 mm.

Head, excluding mandibles, slightly longer than broad, contracted
behind, with convex sides and slightly concave occipital border.
Mandibles large and thick, with finely dentate blades. Clypeus
small, slightly convex, anterior border truncate at middle. Antennal
scape extending past opposite the middle of eye; first funicular joint
three times as long as broad and longer than the second, apical joint
as long as the two preceding together. Pronotum flat, with straight,
distinctly margined sides and narrrowly rounded, projecting humeri.
Mesothorax transverse. Mesoepinotal suture as long as the. pro-
and mesonotum together, nearly as broad behind as in front; its
basal surface margined at sides, shorter than the declivity. Peduncle
of petiole nearly as long as the node; node elongate-globose. Post-
petiole longer than broad, pyriform, nearly twice as broad as petiolar
node. Gaster long and slender. Legs with slightly thickened
femora.

Subopaque; petiole, postpetiole, and gaster shining; finely, densely
punctate tlu-oughout; mandibles subopaque, finely punctate.

Pubescence more abundant than in typical P. gracilis, white.
Pilosity short and abundant on head and thorax, longer and more
sparse on petiole, postpetiole, and gaster; black in color. Head,
thorax, epinotum, and legs black, the tarsi brown; petiole, post-
petiole, and gaster bright ferrugineous.

Described from two workers taken on a shrub at Abuna. This
form is very distinct from the other varieties of P. gracilis in its much
larger size, and the bicolored body. In the latter respect it resembles
P. agilis Emery from Central America, but is very much larger.

MANN: THE ANTS OF BRAZIL. 429

79. Pheidolc {Elasmopheidoh) aherrans Mayr.

One colony was found at Independencia. This species is distrib-
uted along the east coast,' in the more arid localities from north of
Cape San Roque to Argentina. Among the specimens before me
some of the soldiers have the vertex of head, the thorax and epinotum
piceous. Others from the same colony are reddish brown tliroughout,
agreeing in this respect with a series from Buenos Ayres (Silvestri
coll.) in the Wheeler collection.

80. Pheidole (Pheidole) ginlelmi-muUeri Forel subsp. mamore,

subsp. nov.

Soldier. Length 4.75 mm.

Head, excluding mandibles, a little longer than broad, slightly
narrowed in front, with nearly straight sides, broadly rounded occipi-
tal corners and deeply, narrowly concave border. Clypeus convex,
the carina short and thick; anterior border nearly straight. Mandi-
bles short, very thick and blunt, the blade without teeth. Frontal
carinae broad at base; antennal grooves longer than scapes. Scapes
extending a little more than half the distance from eyes to occipital
corners. Club shorter than remainder of funiculus; funicular joints
2-7 as long as broad. Eyes small and convex, located at sides of head
at anterior third. Pronotum transverse, the sides extended into
blunt cones. Promesonotal impression feeble. Mesonotum flat an-
teriorly; with a narrow transverse depression before the basal margin,
which is somewhat elevated; declivous behind. Epinotum broadly
depressed at base, the spines very stout, triangular and strongly
diverging. Petiolar node in profile wedge-shaped; from above twice
as broad as long. Postpetiole three times as broad as the petiole,
conical at sides. Gaster elliptical. Legs short, femora and tibiae
thickened.

Head subopaque, coarsely striate, striae extending over occiput;
clypeus with several coarse punctures anteriorly. Mandibles sub-
opaque; coarsely, sparsely punctate. Thorax subopaque with
transverse, interrupted rugae. Epinotum subhicid, with a few fine
transverse striae. Petiole and postpetiole closely, finely punctate.
Gaster finely punctate, sublucid. Legs shining. Head, antennae,
body, and legs with abundant long, rather stiff pile.

430 bulletin: museum of comparative zoology.

Color black, except the legs and apex of antennae which are cas-
taneous.

Worker. Length 2.55 mm.

Head, excluding mandibles, as broad as long; sides and occipital
margin evenly rounded. Mandibles long, slender, with three teeth
apically. Clypeus slightly convex; anterior border truncate. An-
tennal scapes extending half their length beyond the occiput. Pro-
thorax rounded, elevated. Promesonotal impression indistinct.
Mesonotum in profile convex above, declivous behind. Epinotum
fiat and narrow above, the sides and posterior border slightly mar-
gined; the spines barely perceptible. Node of petiole in profile
rounded above. Postpetiole a little broader than long, with rounded
sides. Legs long, the femora and tibiae somewhat swollen. Mi-
nutely punctate and shining. Base of epinotum transversely carinate.
Pilosity as in the soldier.

Color fuscous; petiole, postpetiole, and legs testaceous fuscous.

Female. Length 5.5 mm.

Head, excluding mandibles, broader than long, sides slightly con-
vex; border of occiput only slightly concave. Mandibles similar to
those of soldier. Antennae short, scapes extending three fourths the
distance to occiput. Eyes large, located in front of middle of head.
Other characters as in the soldier. Thorax flattened above, declivous
in the epinotal region, the epinotal spines short and stout. Post-
petiole twice as broad as the petiole, the sides drawn out into distinct
cones. Gaster one and one half times length of head. Sculpture of
head similar to that of the soldier. Pro- and mesothorax longitu-
dinally striate. Epinotum sparsely and coarselj^ punctured. Gaster
with fine punctures, shining.

Wings. Length a little over 5 mm. Slightly infuscated. Veins
light fuscous.

Described from workers, soldiers, and a single female from a colony
taken at Madeira-Mamore Camp 36. This is a very distinct
form, the soldier differing from that of typical P. guilelmi-muUeri in
its smaller size, darker color, stronger cephalic sculpture, and in being
much less shining. The worker is more shining and much darker
than the soldier. It is much smaller than the closely related P.
hohenlohei Forel from Brazil. The very minute epinotal spines of the
worker are scarcely more than angles of the margin at the bases.

MANN: THE ANTS OF BRAZIL. 431

81. Pheidole (Pheidole) wheeleri, sp. nov.
Plate 3, fig. 23.

Soldier. (Plate 3, fig. 23). Length 3.5 mm.

Head, excluding mandibles, a fourth longer than broad, narrowed
in front, with slightly convex sides, broadly rounded occipital corners
and narrowly excavated border. Mandibles thick, blunt at apex,
with one blunt subapical tooth. Clypeus flat, without a keel; the
anterior border at middle rather strongly bisinuate. Frontal area
large, triangular. Frontal carinae moderately elevated at base,
nearly straight, diverging, extending as far as apex of antennal scapes.
Antennal scapes slightly bent at base, thickened at apex; reaching to
half the distance between eye and occipital corners. Funiculus short
and thick, the club as long as the remainder; fu"st joint twice as long
as broad, joints 2-7 broader than long, joints 8-9 longer than broad;
apical joint of club as long as the two preceding together. Eye small,
flat, located at sides of anterior third of head. Pronotum twice as
broad as long, in front narrow and flat; sides from above drawn out
into strong conical projections, in front of which the straight side
margins converge to the flattened portion. Promesonotal suture
discernible, but not impressed. Mesonotum flat above to the pos-
terior fourth, where it is declivous. Mesoepinotal suture broad and
deep. Base of epinotum longer than the declivity, narrowly excavated
longitudinally; the sides rounded; base strongly depressed; spines
stout, erect, half as long as the base. Petiolar node wedge-shaped in
profile; from above, more than twice as broad as long. Postpetiole
transverse, one and a half times as broad as the petiole, the sides
bluntly conical. Gaster short and oval. Legs short, the femora very
much swollen; tibiae thickened at anterior half.

Body, except epinotum, head, mandibles, antennae, and legs shining.
Mandibles with very distinct, regular, scattered punctures; clypeus
smooth and shining. Cheeks and sides of the anterior portion of
front longitudinally striate, the striae short, extending to about
opposite the eyes; the spaces between shining. Front with rather
strong, regular, widely separated punctures. Pro- and mesonotum
with sparse punctures; the conical lateral pronotal projections densely
punctate at apex. Epinotum sublucid, shallowly granulosely punc-
tate. Petiole, postpetiole, and gaster sparsely punctate, the posterior
portion and sides of the postpetiole transversely rugulose.

432 bulletin: museum of comparative zoology.

Head, mandibles, antennae, body, and legs with rather abundant
long silky pile.

Color dark reddish brown, gaster, epinotum, and a spot on the
vertex much darker than the rest.

Worker. Length 2 mm.

Head, excluding mandibles, a little longer than broad, as broad in
front as behind, with moderately convex sides and nearly straight
occipital border. Mandibles long and slender, with long apical and
subapical teeth. Clypeus convex, flattened anteriorly, the border
concave at middle. Antennal scapes scarcely surpassing the occipital
corners ; all the funicular joints longer than broad. Eyes little convex ;
situated at middle of sides of head. Structure of thorax and abdomen
similar to that of soldier. Sculpture and pilosity similar to that of the
soldier, but the mesonotum is more closely punctate.

Color dark reddish brown throughout.

Described from soldiers and workers taken at Madeira-Mamore
Camp 39. The strongly prolonged sides of the pronotum, which,
with the mesonotum, forms a diagonal quadrangle from above, is very
characteristic. Both soldier and worker are very brightly shining,
with the exception of the epinotum. The punctation of the head and
especially that of the mandibles is coarse.

82. Plieidolc {Pheidolc) carapuna, sp. nov.
Plate 3, fig. 22.

Soldier. (Plate 3, fig. 22). Length 3.24 mm.

Head, excluding mandibles, one and a fourth times as long as broad,
as broad in front as behind, with slightly convex sides and deeply,
narrowly excavated border. Mandibles long and thick, the blade
with a large subapical tooth, a large tooth near base and feeble denticles
between. Clypeus depressed at sides and anteriorly, with a strong,
narrow carina at middle and one on each side, the border rounded.
Frontal carinae extending about half as far as the tips of antennal
scapes. Eyes small, flat, located at anterior third of sides of head.
Antennal scape slightly bent at base, thickened at apex, extending
half the distance to occipital corners. Funiculus short, the first joint
three times as long as thick; joints 2-8 as broad as long; club as long
as the rest of the funiculus with the first two joints subequal, apical
joint longer than the two preceding, connate. Pronotum and mesono-
tmn in profile rounded, the sides of the former extended at the sides

MANN: THE ANTS OF BRAZIL.

433

in the form of a blunt cone. Promesonotal suture basally discerni-
ble. Epinotum at base strongly, broadly impressed for its entire
length, the spines short, triangular and erect. Petiole thick, the node
as deep as thick, from above one and a half times as long as broad;
postpetiole from above transverse, narrowly rounded at sides. Gaster
oval, short. Legs moderately long.

Sublucid. Mandibles shining, with sparse fine punctures at base.
Front of head and cheeks with strong parallel striae, which become
confused and disappear toward the vertex, where they are replaced
by dense granulose punctation; this granulation extends forward,
outward from the frontal carinae to the cheeks. Clypeus strongly,
but more sparsely striate and shining. On the front and cheeks the
spaces between the striae are smooth and shining; the occiput is very
shining. Thorax and epinotum granulosely punctate. Node of
petiole and postpetiole and base of first gastric segment with sparse,
more shallow punctures ; rest of gaster smooth and shining.

Head, thorax, and abdomen with long hairs, which are most abun-
dant on the gaster. Sides of head in front sparsely pilose; antennae
and legs with semierect hairs.

Worker. Length L5 mm.

Head, excluding mandibles, a little longer than broad, with slightly
convex sides, broadly rounded occipital corners, and very slightly,
narrowly concave border. Mandibles long, slender, the blades mi-
nutely denticulate. Clypeus convex, the anterior border rounded.
Frontal carinae short, parallel. Antennal scapes surpassing the occi-
pital corners by about one sixth of their length. Eyes large and con-
vex, located at middle of sides of head. Thorax in profile rather low,
evenly rounded. Promesonotal suture not discernible. Mesoepinotal
suture deeply impressed. Epinotum with equal base and declivity,
the former depressed; spines very short, triangular. Petiole and post-
petiole similar to those of worker.

Head, thorax, and epinotum subopaque; coarsely, densely granu-
lose; cheeks with a few coarse striae. Petiolar node punctate; post-
petiole and gaster smooth and shining. Head and body with sparse,
fine hairs.

Color testaceous.

Female. Length 4 mm.

Head, excluding mandibles a little longer than broad, slightly
narrowed in front, with feebly convex sides, shallowly excavated
border. Mandibles and clypeus as in soldier. Antennal scapes
extending three fourths the distance to occipital corners. Eyes large

434 bulletin: museum of comparative zoology.

and convex, situated at anterior third of head. Ocelli large, arranged
in a triangle. Epinotal spines shorter than in the soldier. The rest
similar to the soldier, with the usual sexual differences.

Sculpture of head similar to that of soldier, except that the coarse
striae extend onto the vertex and occiput. Thorax smooth and
shining. Epinotum transversely striate. Nodes and a space at the
base of the first gastric segment granulously punctate. Pilosity as
in the soldier.

Color dark fuscous, cheeks and clypeus lighter. Antennae and legs
testaceous. Wings strongly infuscated; veins and stigma fuscous.

Described from several workers and a single female collected at
Madeira-Mamore Camp 39. The color of the soldier is striking.
The head is rich reddish brown, darker on the vertex and front and
lighter on the cheeks and clypeus. The mandibles are a rich red
wine-color. The thorax and epinotum are dark fuscous, the gaster
is similarly colored, but with the first segment lighter basally. This
species approaches P. susannae Forel from which it differs in the
much longer and differently sculptured head, the feeble promesonotal
impression, and in color.

83. Pheidole (Pheidole) triconstrida Forel var. laidloivi, var. no v.

Soldier. Length 3.5 mm.

Head, excluding the mandibles, a little longer than broad, slightly
broader in front than behind; posterior corners rounded; occipital
margin rather deeply impressed; occiput at middle with a deep
impression which extends to clypeus, becoming shallow anteriorly.
Frontal carinae more than half the length of the scape, sharply defined.
Eyes at sides of the anterior third of head. Mandibles convex, biden-
tate, the teeth thick and subequal. Clypeus trapezoidal, flattened,
truncate in front and behind, with a narrow median impression from
middle of front extending two thirds the distance to base. Frontal
carinae diverging behind. Antennal scapes reaching three foiu"ths
the distance to occiput. Antennal club a little shorter than the rest
of the funiculus, funicular joints 2-7 broader than long. Prothorax
rounded above, sides a little in front of middle drawn out angulately.
Promesonotal impression distinct. Mesonotum seen from above
trapezoidal, narrowed in front, deeply, transversely constricted at
middle, the posterior portion elevated into a ridge. Mesoepinotal
constriction deep and broad. Base of epinotum as long as declivity.

MANN: THE ANTS OF BRAZIL. 435

deeply concave; on either side vnih a high ridge which terminates in
an acute, erect spine, about two thirds as long as the base. Petiole
twice the length of postpetiole, the node broader than long, with
conical sides; seen from the side concave in front, higher than post-
petiole. Postpetiole broader than long, the sides conical.

Gaster smaller than the head. Legs slender. Head and gaster
shining, thorax and petiole subopaque. Mandibles very finely punc-
tate. Head densely, minutely punctate. Thorax, petiole, and post-
petiole with more dense punctures. Legs finely punctate, shining.

Thorax without hairs, head, petiole, and gaster with sparse,
scattered erect hairs.

Color pale brownish yellow, mandibles dark ferruginous.

Worker. Length 2 mm.

Head, excluding the mandibles, a little longer than broad, with
convex sides and evenly rounded posterior margin. Eyes at sides of
head in front of middle. Mandibles with two very acute teeth.
Antennal scapes extending one third their length beyond the occipital
margin. Thorax similar to that of the soldier, but the prothorax
more rounded at sides. Epinotum, petiole, and postpetiole similar
to those of soldier. Gaster as large as head.

Head, legs, and gaster shining, the rest of the body opaque.

Pilosity and color like those of the soldier. The hairs are even
sparser. Mandibles and border of clypeus darker.

Described from several soldiers and workers from Madeira-Mamore
R. R. Camp 37.

Differing from the typical form in the smaller size, and pale color.
The antennal scapes of the soldier are proportionally longer, reaching
three fourths the distance to corners of occiput.

Named after Dr. James Laidlow, of the medical corps of the Madeira-
Mamore Railroad.

84. Pheidole (Pheidole) biconMrida Ma\T.

Manaos, Porto Velho, Abuna and Camps 39, 41, 43 Madeira-
Mamore R. R.

On the Rio Madeira this species was very common, nesting in or
beneath rotten logs. The colonies are very populous.

85. Pheidole {Pheidole) hiconstricta subsp. bicolor Emery.
Madeira-Mamore R. R. Camp 39.

436 bulletin: museum of comparative zoology,

86. Pheidole (Pheidolc) biconstricta subsp. burtoni, subsp. nov.

Plate 3, fig. 24.

Soldier. Length 6 mm.

Head, excluding mandibles, a little longer than broad, much nar-
rowed in front, with convex sides, narrowly rounded occipital corners
and deeply and narrowly excised posterior border. Mandibles thick,
the blade with five short, very blunt teeth and a large subapical tooth.
Clypeus rather flat, feebly carinate; anterior border strongly excised
at middle. Frontal area large, triangular, impressed. Frontal cari-
nae a third as long as antennal scapes. Antennal scapes broadly
rounded at base, somewhat thickened apically, extending less than
half the distance from the eyes to posterior corners of head. Club
as long as the remainder of funiculus ; funicular joints 2-7 as long as
broad. Eyes small, convex, located at sides of anterior third of head.
Pronotum similar to that of P. biconstricta, but the corners are some-
what more angulate. Mesonotum slightly convex above, with feeble
transverse impression posteriorly. Base of epinotum much less im-
pressed than in P. biconstricta. Petiolar node and postpetiole
broader than long; the latter conical at sides.

Subopaque, except the occipital corners, mandibles, and clypeus,
which are shining. Mandibles strongly striate at base and regularly
punctate throughout. Head coarsely shagreened, coarsely longi-
tudinally rugose between the frontal carinae. Vertex and occiput
wnth foveolate punctures. Thorax and abdomen shagreened, the
gaster more shallowly than the rest. Legs finely punctate. Each
of the foveolate punctures on the head bears a short, stiff, recumbent
hair. Body, antennae, and legs throughout with long, stiff pile.

Head, posterior portion of pronotum, and epinotum piceous; mandi-
bles and clypeus red, the rest castaneous, the gaster distinctly lighter
than the other parts.

Described from a single soldier taken at Porto Velho. The sculpture
of the head is very different from the other forms of P. biconstricta
and the transverse mesothoracic impression is broader and more
shallow.

87. Pheidole {Pheidole) fallax Mayr subsp. emiliae Forel.
Several soldiers and workers were taken at Alanaos.

MANN: THE ANTS OF BRAZIL. 437

88. Pheidole {Pheidole) fallax siibsp. jelksii Mayr.
Found at Abuna, Bolivia, and at Porto Velho, Brazil.

89. Pheidole {Pheidole) fallax subsp. jchlcii var. antillensis Forel.

Several workers and soldiers from Para and Abuna, Bolivia agree
closely with West Indian specimens before me.

90. Pheidole {Pheidole) impressa Mayr.

Two soldiers of this curiously colored species were taken at Baixa
Verde. Since the types were described from Baturite in Ceara, the
species has not been recorded. My soldiers differ a little in color
from those described by Mayr, but otherwise agree closely with his
description. The head is ferruginous instead of ochre-yellow and
the gaster black. It may be that the Baixa Verde specimens repre-
sent an undescribed color variety, but I think it more probable that
those before Mayr were more immature and the color not fully devel-
oped.

91. Pheidole {Pheidole) loallacei, sp. nov.

Soldier. Length 6 mm.

Head, excluding mandibles, longer than broad, narrowed behind;
sides slightly convex; occipital corners narrowly rounded, the border
with a shallow, narrow incision. Mandibles large and stout, the
apical tooth long; blade with a short, stout subapical tooth and
several very small teeth. Clypeus depressed; anterior border dis-
tinctly concave at middle. Frontal carinae strong and thin at base,
extending only slightly past the antennal pits. Frontal area strongly
depressed. Antennal scapes slender, extending almost to occipital
corners, moderately bent at base; funiculus long and very slender,
the first joint somewhat swollen, joints 2-8 cylindrical, three times as
long as broad ; club very slender, a little more than half as long as the
rest of funiculus, the joints abovit four times as long as broad, apical
joint a little longer than the penultimate. Eyes small, convex, located
at anterior third of head. Pronotum in profile evenly rounded; the
sides slightly drawn out and rounded a little posterior to the middle.
Mesonotum in profile angulate at middle, the anterior part flat, the
posterior declivous. Base of epinotum much longer than the declivity.

438 bulletin: museum of comparative zoology.

flat anteriorly, narrowly impressed posteriorly; spines triangular,
the width at base greater than the height. Petiolar node twice as
broad as long, in profile triangular. Postpetiole one and a half times
as broad as long, narrowly rounded at sides. Gaster short, oval.
Legs very long and slender.

Head shining; cheeks and front with widely separated rugae, which
extend only slightly posterior to the eyes; front and vertex with sparse,
distinct punctures. Mandibles sublucid, very coarsely and sparsely
punctured at base. Thorax, epinotum, petiole, and postpetiole with
sparse interrupted transverse rugae; the spaces between the rugae
shining. Gaster finely punctate, shining. Funiculus sparsely pubes-
cent, head, body, antennae, and legs with long hairs.

Color ferruginous; anterior border of head and the mandibles
brownish red. Pile golden yellow.

Worker. Length 5 mm.

Form very long and slender. Head, excluding mandibles, much
longer than broad, strongly narrowed behind to form a distinct neck,
the posterior border of which is broadly expanded and elevated.
Mandibles long, acuminate at tip, with a long subapical tooth and
several shorter teeth. Clypeus depressed in front, the border strongly
concave at middle. Frontal carinae subparallel, extending to opposite
the anterior border of eyes. Eyes large and convex, situated at sides
of front, anterior to the middle of head. Antennae very long and
slender, the scapes about twice as long as the head including the neck;
funicular joints similar to those of worker. Pronotum longer than
broad, strongly narrowed in front; broadest behind middle; the
pleurae extended behind into small elongate tubercles. Promesonotal
suture strongly impressed. Mesonotum longer than broad, little
convex. Mesoepinotal impression broad and deep. Base of epino-
tum distinctly longer than the declivity, rounded above; spines very
short, triangular, erect, located closer together than their distance
from the sides. Petiolar node one and a half times as long as broad,
triangular in profile. Postpetiole longer than broad, with rounded
sides. Gaster oval. Legs very long and slender.

Sublucid. Cheeks with distinct, widely separated rugae. Front
and vertex sparsely punctate. Mandibles sublucid, finely punctate.
Pronotum in front with very sparse, transverse rugae, the rest of
thorax and the abdomen finely punctate. Pile similar to that of
soldier, but more sparse.

Color light ferruginous, anterior border of head, the mandibles,
and the antennal club darker.

MANN: THE ANTS OF BRAZIL. 439

Described from several workers and soldiers, Madeira-Mamore
Camps 39 and 46. This very distinct and striking species is related
to P. hergi Ma}T and P. oxyops Forel, but is quite distinct from
either. From P. hergi the soldier differs in the much longer antennal
scapes, the more slender club, the differently shaped epinotum, which
in P. hergi is broadly depressed, and in the much smaller size of the
epinotal spines. Pheidole wallacei is a larger and more slender species.

The structure of the worker's head, long drawn out behind, and the
very slender thorax and epinotum, and the small size of the spines
distinguish it from the worker of P. hergi. The worker of the latter
species has the meso- and epinotum granulously punctate and
subopaque instead of smooth and shining as in P. wallacei. The
worker has extremely long legs. Those which I observed were on
the ground and entering a hole beneath a tree, where the nest was
probably located. #

92. Pheidole (Pheidole) opaca Mayr.

Numerous workers, soldiers and females from Para, Abuna, and
IVf adeira-Mamore Camp 39, where it occurred commonly in populous
colonies beneath logs.

93. Pheidole {Pheidole) fimhriata Roger.

A single soldier of this characteristic species was found at Ceara-
Mirim, and a female at Madeira-Mamore Camp 39.

94. Pheidole (Pheidole) flavens Roger.
Natal, Ceara-Mirim, Independencia, and Para.

95. Pheidole (Pheidole) flavens subsp. exigtia Mayr.
Taken at Ceara-Mirim and Independencia.

96. Pheidole (Pheidole) colohopsis, sp. nov.
Plate 3, figs. 25, 26.

Female (dealated). Length 4 mm.

Head, excluding mandibles, a little longer than broad; broadest

440 bulletin: museum of comparative zoology.

in front, with nearly straight sides, concave occipital border and
narrowly rounded corners; anterior corners right-angled. Vertex
slightly rounded above, a little longer than the front. Front sepa-
rated from vertex by a distinct angle, and forming with the mandibles
a disc; very strongly impressed with a deeper, longitudinal groove at
middle. Frontal area small, triangular, more deeply impressed than
the front. Clypeus deeply impressed at middle; the anterior margin
carinate at middle, broadly crenulate. Frontal carinae strong,
extending sinuately to ends of antennal scapes. Mandibles very
short and thick, the blade slightly concave, edentate. xVntennae
short, the scape extending a little more than half the distance to
occipital corners, strongly geniculate at the angle which separates
front from vertex, thickened distally. First funicular joint as long
as the two succeeding joints, joints 2-8 as long as broad; joints 9-11
forming a club as long as remainder of funiculus. Eyes large, oval,
flat, with distinct ommatidia; situated at sides in front of middle.
Antennal fossae deep; extending to apex of scape. Ocelli distinct,
the median much the largest, each situated in a foveole, the median of
which is extended forward as a groove. Thorax rather flat above, sides
rounded; mesothorax a little longer than broad; scutellum rounded
behind. Epinotum with subequal base and declivity; the base strongly
concave at middle, with elevated sides, which terminate in short thick,
acuminate spines; surface of declivity concave, but less so than the
base. Petiole elongate; the node transverse, in profile narrowed
above, constricted behind. Postpetiole twice as broad as petiole, the
sides produced into angles, rounded in front, constricted behind; in
profile evenly convex above. Gaster oval, broad at apex.

Shining, with abundant silky pilosity, which is partly recumbent on
gaster and erect on, the other parts of body. Cheeks, front, and
clypeus devoid of hairs. Vertex longitudinally, crenulately striolate,
front and clypeus with striolae arranged in a concentric pattern.
Mandibles with five transverse, diagonal carinae. Thorax, petiole,
and postpetiole minutely punctate; postpetiole above with five short
longitudinal carinae.

Color testaceous; mandibular blades and posterior margins of gas-
tric segments darker. Pilosity yellow.

Described from a single female taken at Madeira-Mamore Camp
No. 41. It was beneath a log, in company with a colony of Myrmico-
crypta foreli. This is the first known South American Pheidole
belonging to the group of species which have the front of the head
depressed and disc-likcj somewhat similar to that of the major
worker in Camponotus subgenus Colobopsis.

MANN: THE ANTS OF BRAZIL. 441

97. Crematogaster (Cremaiogaster) stolli Forel.

This very distinct species was abundant at all the localities visited
along the Rio Madeira. The covered galleries that it constructs
superficially resemble termite runways, but are built of fibrous
material. They extend along tree trunks, often in an irregular spiral
direction around the tree and branching to the limbs. The nest
varies from hemispherical to broad and flat, up to five or six inches
in width, though generally smaller. The brood-chambers are in deep
cavities. The acrid secretion from the anal glands of C. siolli is
much more pungent than that of any other Crematogaster which I
have observed.

98. Crematogaster (Crematogaster) stolli subsp. autruni, subsp. nov.

Worker. Length 4.5 mm.

Head, excluding mandibles, a little broader than long, as broad in
front as behind, with rounded sides, broadly rounded occipital cor-
ners; the occipital border nearly straight, but narrowly concave at mid-
dle. Clypeus about as broad as long, with slightly rounded surface
and broadly rounded anterior border. Mandibles short and stout.
Frontal carinae feeble, but distinct, extending to opposite the eyes,
which are located at middle of sides of head. Eyes small and flat.
Antennae short and stout, the scapes arcuate, much thickened dis-
tally, extending about two thirds the distance to occipital corners.
First funicular joint as long as the two succeeding joints taken to-
gether; joints 2-7 as long as broad. Pro- and mesonotum together
as broad as long, evenly rounded above and at sides. Epinotum with
subequal base and declivity, the base rounded in profile, the surface
of base concave. Spines moderately long and stout. Petiole in
profile about twice as long as broad; from above, flattened and two
thirds as broad as long. Postpetiole elongate, rounded. Legs stout.
Gaster of medium length, triangular.

Body sublucid. Head in front with sparse, fine punctation; cheeks,
sides of head and a space parallel to the frontal carinae densely
striolate. Mandibles rugosely striolate, with a few fine recumbent
white hairs. Antennae punctate, the funiculus with considerable
fine, recumbent pile. Promesonotum finely rugulosely punctate,
with a few very fine hairs. Epinotum and mesothoracic pleurae

442 bulletin: museum of comparative zoology.

coarsely striolate. Petiole and postpetiole rugosely punctate. Gaster
subopaqiie, densely punctate, with a few erect black hairs.

Color black, except the head, which is somewhat reddish.

Described from six workers taken near Manaos. This form, which
is named after Don Antonio Autrun, our genial host at Kete Purangi,
differs from C. stoJli in the shorter antennal scapes and in sculpture
and color, besides being smaller. The epinotal spines are shorter
and thicker and the head and thorax lack the long, erect hairs of C.
stoUi. I am in doubt whether C. autruni should not be considered as
specifically distinct rather than as a subspecies.

99. Crematogaster (Crematogaster) heathi, sp. nov.
Plate 3, fig. 27.

Worker. Length 2.5 mm.

Head about as broad as long, with slightly convex sides and broadly
rounded occipital corners; occiput faintly excavated. Eyes small,
little convex, located at sides of head slightly posterior to middle.
Mandibles short and stout, the blade strongly concave. Antennal
scape not reaching the occipital corners; funicular joints, except
those of the club, subglobose, as broad as long. Clypeus quadrate,
as broad as long, the anterior border broadly rounded. Prothorax
subglobose; pronptum as broad as long, with narrowly rounded sides,
so that the disc is transversely oval, broadly rounded above. Epino-
tum with a very short base, which is narrowly rounded, giving in
profile the appearance of a tubercle; the declivity is three times as
long as the base, with a broad, flat surface, on either side of which is a
strong spine curving outward and upward. Petiole flattened above,
not quite so broad as long, with straight posterior margin; postpetiole
subglobular. Gaster short and broad, triangular.

Head shining, minutely punctate throughout, the front with sparse,
coarse punctures. Thorax, petiole, postpetiole, and legs sublucid.
Gaster subopaque, densely punctate. Head and thorax, petiole and
postpetiole in front sparsely pilose, the pile fine and recumbent.
Antennse, legs, and abdomen with erect pilosity.

Color red, except the gaster, which is black, and the legs, which are
somewhat infuscated. Pile and pubescence yellowish.

Described from a large number of workers taken from twigs at
Independencia. This species is related to C. brevispinosa from which
it differs in the structure of the epinotum and in the absence of teeth

MANN: THE ANTS OF BRAZIL. 443

on the mandibular blades. In living specimens the black color of the
gaster is in vivid contrast to the red of the other parts.

100. Crematogaster (Cremaiogaster) brevispinosa Mayr.

Several colonies of this well-marked species were taken at Itacoatiara
and Abuna, Bolivia.

101. Crematogaster (Crematogaster) brevispinosa subsp. rochai Forel.

This is common at Natal and Baixa Verde, where numerous colo-
nies were found. The workers agree closely with cotypes from Ceara.

102. Crematogaster (Crematogaster) vidima F. Smith.

Many colonies were found in hollow twigs at Natal, Baixa Verde,
Independencia, Para, and Madeira-Mamore Camp 39.

103. Crematogaster (Crematogaster) sulcata Mayr.
Abuna, Rio Madeira.

104. Crematogaster (Crematogaster) limata F. Smith.

Common at Manaos and Porto Velho. A colony at the latter
locality was nesting in a leguminaceous pod.

105. Crematogaster (Crematogaster) brasiliensis Mayr.
Manaos and Porto Velho.

106. Crematogaster (CYematogaster) longispinosa Emery.
Many workers were taken at Abuna.

107. Crematogaster (Crematogaster) acuta (Fabricius).

Many workers are in the collection from Madeira-Mamore Camp.
39.

444 bulletin: museum of comparative zoology.

108. Monomorium (Mitara) siihterraneuvi, sp. nov.
Plate 4, figs. 29, 30.

Worker. (Plate 4, figs. 29, 30). Length 3.5 mm. to 5 mm.

Head, excluding mandibles, as long as broad, nearly as broad in
front as behind, with convex sides and feebly concave occipital border.
Eyes small, located at sides of head, a little posterior to the middle.
Clypeus very convex, unarmed, the anterior border somewhat concave.
Frontal carinae short, straight. Antennae 11-jointed; scapes extend-
ing a little past the occipital corners of head; first funicular joint
longer than broad and nearly twice the length of the second, which is a
Httle longer than broad; joints 3-7 as broad as long; joints 8-10 form-
ing an elongate club, almost as long as the rest of funiculus, the apical
joint of which is longer than the two preceding together. Mandibles
rather thick, the blade with five strong teeth. Pronotum transverse,
convex above and at sides, broadest a little behind the middle. Pro-
mesonotal suture barely perceptible; mesoepinotal impression strong.
Epinotum in profile evenly rounded; the sides very feebly margined.
Petiole slender, the node in profile deeper than thick, rounded above,
convex at the anterior, flat at the posterior surfaces. Postpetiole
transverse, subglobose. Gaster as long as thorax and epinotum,
broadly oval. Legs long and slender.

Sublucid. Mandibles, sides of clypeus, and frontal carinae coarsely
striate longitudinally; cheeks with strong concentric striae, which
terminate at the carinae, the vertex with sparse irregular rugae, the
rest of head and the thorax densely punctulate. Epinotum trans-
versely, and the posterior surfaces of the petiolar and postpetiolar
nodes longitudinally carinulate. Gaster shining, sparsely punctate.
Apical half of funiculus pubescent; the rest without pubescence but
abundantly pilose, the pile stiff and erect.

Color testaceous, with a fuscous blotch on the front and vertex.

Described from a number of workers from Madeira-Mamore
Camp 39. These were in the ground, beneath the base of a recently
uprooted palm, about three feet below the surface.

In a recent paper (Ann. Soc. ent. Belg. 1913, 57, p. 261) Emery has
divided those species of IVIonomorium which have eleven-jointed
antennae into three subgenera. Of these, M. suhtcrraneum must
be placed in Mitara, which hitherto has contained only African, Asi-
atic, and Australian species. The species in this subgenus have the
clypeus entirely without denticles or carinae, the eyes developed, and

MANN: THE ANTS OF BRAZIL. 445

the epinotum unarmed. Monomoriwn subterraneum is very different
from any other of the American Monomoriums on account of these
characters, and also because it is an unusually large species, with
considerable variation in size between the largest and smallest workers.
The small eyes, the abundant stiff hairs with which the body is covered,
and the color, as well as the location in which the colony was found,
indicate that this species is hypogaeic in habit.

109. Megaloviyrviex bituberculatus Forel.

Many workers which agree closely with specimens from the Rio
Purus, received from Forel, were taken at Manaos, Porto Velho, and
Madeira-Mamore Camps 39 and 41. This form, which is confined
to the upper Amazonian region, attends Membracidae and the workers
were generally found in company with these on shrubs in the deep
forest. The nest is subterranean, the entrance nearly always at the
base of a tree. The living insect is slow in its movements.

110. Megalomyrmex umUacei, sp. nov.
Plate 3, fig. 28.

Worker. (Plate 3,. figs. 28). Length 5 mm.

Head, excluding mandibles, one and a third times as long as broad,
broadest at clypeus, strongly narrowed above eyes, with slightly
convex sides and straight occipital border, the latter distinctly,
though narrowly, margined. Clypeus long and narrow, not separated
from frontal area, the sides extending to outer bases of mandibles;
anterior border projecting and narrowly rounded at middle. Frontal
cai'inae distinct, elevated at insertion of antennae, parallel, extending
to opposite anterior border of eye. Mandibles long and acuminate,
the blade with four pointed teeth. Eyes large and very convex,
situated at sides of front of head, a little anterior to the middle.
Antennae long and slender; extending about one third their length
beyond the occipital corners. First funicular joint nearly twice as
long as the second, joints 2-8 subequal, longer than broad, cylindrical,
joints 9 and 10 subequal, each twice as long as the eighth and a little
shorter than the apical. Thorax long and slender. Prothorax
slightly rounded above, sides rounded. Promesonotal suture faint.

446 bulletin: museum of comparative zoology.

Mesonotum pyriform, flattened behind, with a feeble transverse im-
pression at middle and a longitudinal impression in front, the sides
slightly margined posteriorly. Epinotum in profile rather flat above;
its base about twice as long as declivity, into which it passes at a broad
rounded angle; surface of base broadly concave, that of declivity
flat; base of declivity with broad, ear-shaped tubercles. Petiole one
and a half times length of postpetiole, without antero ventral tooth;
the node in profile as deep as broad, concave in front, rounded behind,
with a constriction near apex. Postpetiole broader than petiolar
node, rounded above, the anterior surface nearly straight; antero-
ventral border angulate in front, but without distinct tubercle or
tooth. Legs long and slender.

Gaster shining, sparsely, regularly punctate; rest of body and the
legs subopaque. Head and thorax rugulosely punctulate, a median,
longitudinal surface on the front less so and more shining than the
rest; front of head between frontal carinae and eye concentrically
striolate. Mandibles shining, coarsely striate. Epinotum and petio-
lar node transversely striolate; postpetiole rugulose like the front
of head. Body and legs everywhere with abundant erect hairs.
Mandibles, funiculus, and antennal scapes with semierect hairs, last
four funicular joints pubescent.

Color light ferruginous, pile and pubescence yellow.

Female (dealated). Length 6.5 mm.

Closely resembling the worker. The white ocelli are small, but
distinct. The smooth surface on the front is shorter but more dis-
tinctly shining than in the woi'ker.

Described from a single female and several workers found at Porto
Velho. This species is near M. iheringi Forel, from which it differs
in its larger size, the head more constricted behind, in the much
larger size of the eye, and in the subopacity of the tegument, which in
M. iheringi is shining.

in. Tranopelta gilva Mayr. var. albida, var. nov.

Several workers of a small variety of Tranopelta gilva Mayr were
taken at Camp 39, Madeira-Mamore R. R. These differ from
var. brunnea Forel, the only form of which the worker is known, in
the smaller size (length L5 mm.) and in the uniformly pale color
which in the series before me is almost white, with a faint tinge of
yellow.

. MANN: THE ANTS OF BRAZIL. 447

112. Solenopsis picea Emery var. subadpressa Forel.

This variety, somewhat larger than the typical Costa Rican form,
was described by Forel from specimens collected by Sr. Diaz da Rocha
at Ceara. Specimens which agree well with cotypes in the Wheeler
collection were taken at Porto Velho, Abuna, and Madeira-Mamore
R. R. Co. Camp 39.

113. Solenopsis glohularia (F. Smith).

One colony with numerous workers and females was found at Natal.
The large size and globular shape of the postpetiole distinguish this
from related species.

114. Solenopsis subtilis Emery.
One small colony was taken at Manaos.

115. Sole7iopsis geminata (Fabricius).

Colonies of the typical form of this widely distributed species were
found at Natal, Independencia, Baixa Verde, Itacoatiara, Manaos,
and Porto Velho.

116. Solenopsis geminata (Fabricius) subsp. medusa, subsp. nov.

Plate 4, fig. 31.

In a number of colonies from Ceara-Mirim and the Maranguape
mountains the largest headed of the soldiers differ from those of
typical S. geminata in having the sides of head at the anterior corners
broadly expanded, as shown in Plate 4, fig. 31.

I can detect no character in the smaller soldiers and workers that
will distinguish them from the same phases of S. geminata, but the
difference between the largest soldiers is striking and constant and
seems to be of subspecific value. It is probably a local race, limited
to the east coastal region of Brazil.

117. Solenopsis saevissima (F. Smith).

Very abundant at Para, Itacoatiara, and Manaos in Brazil and at
Abuna, Boh via There can be no doubt that this species is the one

448 bulletin: museum of comparative zoology.

described by Smith from specimens sent to him by Bates and of which
the latter has given an interesting account in his NaturaUst on the
River Amazons. It is the common fire ant ("formigo do fogo" of
the Brazihans) of the Amazonian region, where it occurs generally
in cultivated districts and is a bad pest. The colonies are large and
numerous and the individuals highly aggressive, and, because of their
numbers and painful sting, very formidable.

Smith's name has long been included with the synonyms of S.
geminata, and S. saevissima was redescribed as S. pylades by Forel,
but from field observations, compared with the notes of Bates and
Smith and compared with specimens of S. pylades determined b}'
Forel I am certain that the name given by the former author applies
to this species.

118. Leptothorax (Goniothorax) echinatinodis Forel subsp. spininodis

Ma>T.

Plate 4, fig. 36.

Many workers and males from colonies which were nesting in
twigs at Independencia, Ceara-Mirim, and Manaos agree closely
with Mayr's description, except that in the entire series before me the
tips of the femora are distinctly infuscated.

The types of this species were taken from an egg-case of a species
of Mantis from a doubtful locality.

119. Tetramorium {Tetrogmus) simillimum F. Smith.
Several workers of this tropicopolitan species were taken at Manaos.

120. Wasmannia auropunctata (Roger).

This is one of the most widely spread Neotropical ants. It was
taken in Brazil at Natal, Independencia, Itacoatiara, Manaos, Porto
Velho, Madeira-Mamore Camps 39 and 41, and at Abuna, Bolivia.

121. Cryptoccrvs iCephalotes) atratus (Linne).

Common throughout northern Brazil. Taken by the Expedition
at Para, Manaos, Itacoatiara, and on the upper Rio Madeira. The
species nests generally in hollowed branches of high trees, though one

MANN: THE ANTS OF BRAZIL. 449

nest was in the hollowed trunk of a small tree. It is omnivorous in
habit, frequenting garbage and eating even carrion. Some dead
macaws which I placed in the woods as bait for carrion-feeding insects
were continually covered by C. atratus, to the exclusion of other
insects. It is diurnal, and a striking form as it walks slowly about on
tree trunks and logs. The hard spiny armor is sufficient to protect
it from any ordinary enemy.

122. Cryptocerus (Cephalotcs) oculatus (Spinola).
One colony from Para.

123. Cryptocerus (Zacryptocerus) clypeatus (Fabricius).

This species was very common at all places along the Rio Madeira,
and at Itacoatiara and Santarem. A large colony was found nesting
in a hollow parasitic vine.

124. Cryptocerus {Cryptocerus) umbraculatus (Fabricius).

A colony found at Abuna, Rio Madeira was nesting in a hollow
branch near the top of a recently felled tree.

125. Cryptocerus (Cryptocerus) inaequalis, sp. nov.

Worker. Length 5.25 mm.

Head broader than long, broadest behind the eyes; narrowed in
front. Posterior margin straight, the angles projecting as lamellae,
and broadly concave at apex. Sides and anterior corners of head
evenly rounded. Anterior margin of clypeus deeply concave. Pro-
and mesothorax together as broad as long; prothorax angulate at the
anterior corners and with a pair of flat triangular teeth at posterior
half of margin. Sides of mesothorax with a short slender spine at
middle. Epinotum more than twice as broad as long, with two broad
teeth at sides. Petiole and postpetiole subequal in width. Petiole
with long, slender, backward curving spines. Spines of the postpetiole
short and broad, projecting forward. Gaster broadly cordiform,
very convex above; anterior third of sides broadly margined; excised
at middle of anterior border.

Subopaque. Sparsely, foveolately punctate above, except the gas-

450 bulletin: museum of comparative zoology.

ter, which is finely and densely punctate. Legs foveolately punctate.
Each puncture bearing a glistening scale-like hair.

Color black, with the marginate portions of the head and gaster
yellow, and the tibiae ferruginous.

Described from four workers from Abuna and Madeira-Mamore
Camp 41.

The very long, slender, and curved spines of the petiole, compared
with those of the postpetiole, and the lamellate occipital corners, with
the concave apices are characteristic of this species. It is related to
C. spinosus but differs in the form of the spines, and the absence of
pile on the dorsum.

126. Cryptocenis {Cryptocerus) spinosus Mayr.
One colony found at Porto Velho, Rio Madeira.

127. Cryptocerus (Cryptocerus) pusillus Klug.

Widely distributed throughout Brazil. Numerous specimens were
taken at Ceara, running along the wires of a fence, in the posts of which
they were nesting. At Manaos a colony was taken from a hollow
twig.

128. Cryptocerus (Cryptocerus) minutus (Fabricius).
Nesting in twigs at Natal, Manaos, Itacoatiara, and Porto Velho.

129. Cryptocerus (Cryptocerus) deprcssus Klug.
This was the commonest species of the genus at Manaos.

130. Cryptocerus (Cryptocerus) maculatus F. Smith.

A few specimens were taken at Natal, and a colony was found in a
hollow twig on a caju tree at Abuna, Bolivia.

131. Cryptocerus (Cryptocerus) multispinus Emery.
A single worker from Abuna.

132. Cryptocerus (Cryptocerus) cordatus F. Smith.
A small colonv from Porto Velho.

MANN: THE ANTS OF BRAZIL. 451

133. Cryptocerus {Cryptocerus) complanatus Guerin.
Plate 4, fig. 35.

A colony of C. complmialns, originally described from Cayenne, was
taken from a twig at Itacoatiara. The species belongs to the group
containing C. cordaius, C. spinosus, and C. inaequalis.

Worker. Length 5 mm.

The head is as broad as long, margin of occiput only slightly con-
cave, the corners broadly angulate, without spines; front narrower
than occiput; sides in front of eyes convex. Thorax longer than
broad. Sides of prothorax narrowly margined, nearly straight;
anterior corners angulate. Mesothorax with short spine at sides.
Epinotum three times as broad as long, with a blunt spine at middle
and a longer one at apical corners. Spines of petiole and postpetiole
curved backward, those of the petiole the largest. Abdomen cordate,
without dorsal impression at base.

Body subopaque, coarsely punctate, each puncture with a short
silvery hair.

Color black throughout.

Soldier. (Plate 4, fig. 35). Length 6 mm.

The body is less hairy and more shining than in the worker. The
transverse carina between the pro- and mesothorax is prominent at
sides and more feeble on disc. The angles at anterior corners of
prothorax are acute. Mesothorax with short, blunt spines at sides.
Epinotum without spine at middle; the corners with a short, pointed
spine. Spines of petiole and postpetiole similar to those of worker,
' but stouter.

Color black.

The worker agrees in outline with the Peruvian specimen figured
by Emery (Bull. Soc. ent. Ital. 22, 1890, 22, p. 75, pi. 9, fig. 6). The
outline of the soldier is shown in Plate 4, fig. 35.

13-4. Cryptocerus (Cryptocerus) pilosus Emery var. fiebrigi Forel.

Several workers and soldiers from Natal agree closely with the
description and with cotypes of this variety received from Dr. Forel.
This very distinct species has not been hitherto recorded from Brazil,
but undoubtedly is widely distributed along the coast.

452 bulletin: museum of comparative zoology.

135. Daceton armigerum (Latreille).

There were several colonies of this species in the grounds of the
Zoological Garden at Para, where workers were often seen on the
fences and trees. At Itacoatiara I found a very large colony nesting
in a hollow standing tree.

136. Acanthognathus occllatus Mayr.
Plate 5, fig. 39.

Female. (Plate 5, fig. 35). Length 3.5 mm.

Head cordate; excluding mandibles, twice as long as broad, with
convex sides and deeply excised occiput. Eyes large, convex, situ-
ated at middle of sides of head. Ocelli small, convex, white in color.
Clypeus longer than broad, its anterior border truncate. Antennae
slender and short; the scape not attaining the occipital corners, bent
and slightly thickened distally; first funicular joint two and a half
times as long as broad; joints 3-9 scarcely longer than broad; last
two much longer than broad, the apical nearly twice the length of
penultimate. Mandibles nearly as long as head, slender, straight;
apex with two long subequal teeth and a shorter tooth. On the inner
side near the base is a curved process, which extends upward a little
past the base. This is bidentate at apex. Pronotum very narrow;
in profile flat. Mesonotum evenly rounded above and at sides.
Scutellum rounded. Epinotum with base and declivity subequal in
length, the angle bearing a pair of stout, acute spines. Petiole elon-
gate, slender, the node globular. Postpetiole subglobose, about as
large as the petiolar node. Gaster short and thick.

Head in front somewhat shining; genae and occiput smooth and
shining. Mandibles finely punctate. Antennae and thorax sub-
opaque, the latter coarsely, rugosely punctate. Scutellum with five
feeble longitudinal carinae. Petiole, postpetiole, gaster, and legs
finely punctate, shining.

Front of head and occiput, thorax, petiole, postpetiole, gaster,
legs, and antennal scapes pilose.

Color ferruginous; the front of head, thorax, and gaster darkest,
legs lightest.

Described from one female which was found with a solitary worker
beneath a board on the ground at Para.

MANN: THE ANTS OF BRAZIL. 453

137. Strumigenys smithi Forel.
A single worker from Para is in the collection.

138. Strumigenys schdzi Emery.

A single female and several workers were taken at Para, the type-
locality.

139. Atta cephalotes (Linne).

Common throughout the forest regions, in enormous colonies.
Numerous specimens were taken at Porto Velho, Abuna, and Madeira-
Mamore Camp 41.

140. Atta sexdens (Linne).

Many specimens were collected at Natal, Ceara, Independencia,
Baixa Verde, Ceara-Mirim, Manaos, and Porto Velho.

This ant (the "sauba" of the Brazilians) ranges throughout the
tropical portions of South America, and is by far the most important
economically of all ants. It strips many cultivated plants of their
leaves, is especially attracted to citrus species. It takes also dried
vegetable matter, in particular farina, the staple food of Brazil. At
Independencia, back of our house was a large pile of kitchen refuse,
and this was visited nightly by hordes of workers, which collected
particles of farina, bread, and other vegetable material.

All sizes of workers forage for leaves, generally at night, but also
in late afternoons or on cloudy days. The smaller workers of this
and the preceding species often ride upon a portion of a leaf which is
being carried by a large one and this has given rise to the native belief
that the larger workers are blind and are guided by the smaller ones.

141. Acromyrmex {Moellerius) landolti Forel.

Common at Natal and Baixa Verde.

This species was described from a specimen from Colombia, and
Forel has since recorded it from Bahia, and Emery from Venezuela.
At Natal and Baixa Verde nests were very common. At the entrance
is built a turret of grass, from one to six inches in height. Through

454 bulletin: museum of comparative zoology.

this run circular tunnels, which vary in number, from one to eight.
From the entrance the tunnel runs perpendicularly to the first fungus-
garden chamber. This is about three inches in diameter and from
six inches to two feet below the surface. The colonies are rather
small. Although on the watch for inquilinous insects I failed to find
any spiders in the turrets, as reported by Emery (Ann. Soc. ent.
France, 1888, ser. p. 66.) in nests at Caracas.

142. Acromyrviex (Acromyrmex) coronata (Fabricius).
Several colonies were found at Para.

143. Acromyrmex (Acromyrmex) aspersa (F. Smith).
Common at Natal, Baixa Verde.

144. Acromyrmex (Acromyrmex) octospinosa (Reich),
Numerous workers were found at Itacoatiara.

145. Acromyrmex (Acromyrmex) nigrosetosa Forel.
One colony was found at Souza, near Para.

146. Acromyrmex (Acromyrmex) nigra (F. Smith).
Several workers are in the collection from Itacoatiara.

147. Acromyrmex (Acromyrmex) emiliae Forel.

A few workers were found at Madeira-Mamore Camp 39.
This large, opaque form is very distinct from the related species
in having no tubercles on the epinotum.

148. Trachymyrmex diversus, sp. nov.

Worker. Length 4 mm.

Head, excluding mandibles, a little longer than broad, broader
behind than in front, with convex sides, and pointed posterior corners;
occipital margin slightly concave. Eyes con^•ex, located in front of

MANN: THE ANTS OF BRAZIL. 455

the middle of the sides of head. Mandibles with two large apical
teeth and several fine ones basally. Anterior border of clypeus
broadly rounded. Lobes of frontal carinae large, flattened, sub-
triangular, the carinae extending back as strong, converging ridges
nearly to the occipital corners of the head. Antennal scapes dis-
tinctly thickened apically, extending about one third their length
past the occipital corners; funicular joints all distinctly longer than
broad. Occipital angles with one large, double spine, and anterior
to these several small conical tubercles. Preorbital carinae extending
backward toward, but not reaching the occipital corners of the head.
Vertex with distinct ridges, which converge posteriorly. Lateral
spine of pronotum long and stout, the apex excised to form two blunt
tips, the anterior of which is higher than the posterior. Median
spines very short, conical. Inferior spine two thirds as long as lateral
spines, stout, the apex broadly rounded. Mesonotum with two
pairs of lateral, subequal, stout spines, about one third the length of
the prothoracic spines, strongly denticulate. ISIesoepinotal con-
striction rather strongly impressed. Epinotum with subequal base
and declivity, meeting at obtuse angle; with a pair of short, blunt
spines a little in front of the middle of base; apical spines short and
conical. Petiole from above about as broad as long, very abruptly
narrowed anteriorly into a short petiole; node with a pair of stout
spines. Postpetiole as broad as long, twice as broad as petiole;
the posterior border distinctly margined, with a flattened impression
anterior to margin and the lateral borders with three strong tubercles,
the two anterior short and triangular, the one posterior longer and
blunt.

Gaster suboblong, broadest behind the middle; first segment with
prominent lateral ridge. Tubercles small, blunt, few in number.
Legs long, rather slender, without tubercles.

Mandibles faintly shining, with a few hairs; remainder of body
opaque and finely granular.

Hairs ferruginous, sparse, very short, and depressed, except on the
ventral surface of gaster, where they are longer and suberect.

Color ferruginous, the antennae and carinae of front and vertex
darker.

Described from workers taken on the trunk of a palm at Abuna,
Rio Madeira.

Most closely related to T. ocikeri Forel from Sao Paulo, but difYering
in the smaller size of the anterior pronotal spines, which in T. ocikeri
are much longer than the posterior pair.

456 bulletin: museum of comparative zoology.

149. Apterostigma branneri, sp. nov.

Plate 5, fig. 37.

Worker. Length 5 mm.

Near A. caherti Wheeler. Head, excluding mandibles one and
three fourth times as long as broad, with nearly straight sides, as
broad in front as behind. Posterior portion convex, behind con-
tracted into a rather short neck, the sides of which are straight, the
posterior angles somewhat prolonged and the border between them
truncate. Eyes convex, located a little behind the middle of sides of
head. Clypeus twice as long as broad, Avith rounded anterior border.
Mandibles with seven subequal teeth. Frontal carinae in front
strongly lobed, behind these becoming very feeble, the posterior por-
tion shorter than the lobes. Antennal scape extending nearly one
half its length past the occipital corners, first funicular joint nearly
as long as the three succeeding joints together; joints 2-9 broader
than long. Thorax elongate. Pronotum with a moderately strong
anterior border, which is slightly reflexed. Mesonotum with a pair
of subparallel longitudinal ridges. Epinotum with similar ridges ex-
tending along base, and to base of declivity. In profile the declivity
of the epinotum is nearly straight and about as long as the slightly
convex base; the angle between the two surfaces rounded. Petiole
from above less than twice as long as broad, sides convex, evenly
narrowed anteriorly to the point of insertion. In profile the node is
rounded, more than half as high as the length of the petiole. Post-
petiole slightly broader than long, broadest a little behind the middle,
the sides evenly rounded; posterior border very slightly convex.
Gaster elliptical, without a longitudinal ridge on each side. Legs
long, rather stout.

Mandibles feebly shining, finely longitudinally striate and sparsely
punctate. Funiculus of antennae slightly shining, the rest opaque,
very densely punctate.

Hairs abundant, long, suberect, dark at base, lighter towards apex.
Pubescence very short.

Color dark brown, funiculus lighter.

Female. Length 5.25 mm.

Closely resembling the worker. The eyes are larger. Mesothorax
without longitudinal ridges. Scutellum with two broad, blunt teeth.
Base and declivity of epinotum equal, meeting to form an obtuse
angle; base with faint parallel ridges. Wings opaque, dark fuscous,

MANN: THE ANTS OF BRAZIL. 457

with an elongate black spot between branches of cubital vein. Rest
as in the worker.

Male. Length 5 mm.

Head, excluding mandibles about one and one third times as long
as broad, rounded at sides, behind more suddenly constricted than in
the worker and female. Eyes and ocelli large, very convex. First
joint of funiculus one third as long as second; second joint one and
one fourth times as long as third; joints 3-11 subequal, two and one
half times as long as broad. Prothorax in front constricted; seen in
profile depressed; broadly margined. Mesothorax rounded, without
longitudinal ridges. Scutellum with broad, flat teeth, as in female.
Epinotum, petiole, postpetiole, and gaster similar to these parts in the
female.

Pilosity abundant, black, suberect on the head and body, more
depressed on the legs.

Wings. Length 4 mm.; fuscous, clouded with darker spots.

Color as in worker and female.

Described from specimens taken from two colonies at Abuna and
Madeira-Mamore R. R. Camp 39. Apierostigma hranncri approaches
most closely A. calrerti Wheeler from Costa Rica, but A. hranneri
has the head distinctly longer, the pronotum less declivous in front,
and without the deep impression posterior to the margin, and it lacks
the distinct ridge on the propleurae; the angle between the base and
declivity of the epinotum is much more obtuse. Seen in profile, the
slope of the epinotal declivity is more gradual. The size of the worker
of A. calverti is smaller (3.5-4 mm.).

150. Myrmicocrypta foreli, sp. nov.
Plate 4, fig. 32-34.

Worker. Length 3.25 mm.

Head, excluding mandibles, about one fourth longer than broad, a
little longer behind than in front, with slightly convex sides and
concave occipital border, the occipital corners drawn out into thick
blunt spines. Front broad and flat. Clypeus with broadly rounded
anterior border. Frontal carinae somewhat elevated at base of
antennal scape, anteriorly extending outward, almost attaining outer
corners of the clypeus; posteriorly very slightly diverging towards
occipital border, becoming much weaker at a little less than half the
distance from base of scape to occiput. Antennal scapes curved,

458 bulletin: museum of comparative zoology,

thickened at tips, extending one fourth their length beyond the occipital
corners. Joints 3-8 of funiculus a little broader than long, gradually
increasing in length towards apex; the basal joint of club one half as
long as the terminal. Eyes small, located at the middle of sides of
head. Mandibles long and rather slender with about seven short
teeth on the blade, the penultimate and apical larger than the others.
Pronotum concave at middle of front; on the side anteriorly with a
sharp, minutely tuberculated crest; the median surface posteriorly
with a pair of short, parallel tuberculated ridges, half as far apart as
those lateral and anterior, extending half the length of the pronotum
with a depression between them and the lateral margin ; inferior spine
short and acute. Promesonotal impression strong. Mesonotum
lower than the pronotum but higher than the epinotum ; at sides with
a bidentate crest. Epinotum with subequal base and declivity, the
latter very abrupt. Base margined with tuberculate ridges, which
terminate in spines, which are stout, acute and about half as long as
the declivity. Petiole elongate; the node globular, shorter than the
distance to point of insertion. Postpetiole twice as broad as petiole,
broader than long, sides evenly convex, posterior border truncate.
Gaster elongate, broadest in front of apex of first segment. Legs
long and stout.

Body and legs subopaque, except the gaster, the first segment of
which is shining. Head with very small tubercles arranged in more
or less circular patterns on the front. Mandibles longitudinally
striate, sublucid. Body everywhere finely granular, with minute
tubercles, which are coarser on the petiole, postpetiole, femora, and
tibiae, and more scattered on the gaster.

Body, antennae, and legs with thick, short, curved, recumbent,
glistening hairs.

Color light ferruginous, the blades of the mandibles darker.

Female (dealated). Length 4 mm.

Head similar to that of worker. Pronotum transverse, flattened
above, the sides thinly margined, the margins projecting anteriorly
as blunt lateral teeth. Mesonotum slightly rounded above, with a
longitudinal pair of denticulated crests at sides and middle and between
these a simple carina. The posterior border of scutellum extending
into a pair of blunt spines. Epinotum with two long, strong spines
at angle of base and declivity, which are ec[ual. The rest similar to
worker.

The color is somewhat darker, being light reddish brown.

Described from one female and several workers collected at Madeira-

MANN: THE ANTS OF BRAZIL. 459

Mamore Camp 41. The colony was under a log, about three
inches beneath the surface of the earth. A short perpendicular
passage led to a single chamber in which was the small fungus garden.

151. Cyphomyrmex rimosus (Spinola).

Natal, Baixa Verde, Para, Manaos, Abuna, Porto Velho, and Abuna,
Bolivia.

Occurs in small colonies, beneath stones or in rotten wood. The
death feigning instinct is strongly developed and the insect rolls up
and remains inert for some time when touched.

DOLICHODERINAE.

152. Dolichoderus (Dolichoderus) decollatus (F. Smith).

This species occurs commonly throughout the greater part of tropi-
cal South America. A large series of workers, collected at Itacoatiara,
Porto Velho, and Madeira-Mamore Camp 39, shows a great deal of
variation in color. Some have the head and thorax brownish red, in
others these parts are entirely black. These forms are connected, by
gradations.

Most of the workers that I observed were on the trunks of high
trees. They are slow in motion, and have a habit of remaining motion-
less for many minutes at a time. When alarmed they drop to the
ground.

153. Dolichoderus (Dolichoderus) imhecillus, sp. nov.

Plate 2, fig. 18.

Worker. (Plate 2, fig. 18). Length 10 mm.

Near D. atellaboidcs. Head, excluding mandibles and the neck,,
longer than broad, with convex sides. Occiput prolonged into a
neck which in profile is nearly twice as long as thick and moder-
ately reflexed at the posterior border. Mandibles slender, the blade
with ten minute teeth and two larger coarser ones apically. Pro-
notum as broad as long, rounded above and at sides. Mesothorax
similar to that of D. atellaboides, long and slender; in front with a
small, rounded, elevated portion, on each side of which is a strong
impression, extending backward and converging and terminating in

460 bulletin: museum of comparative zoology.

tubercles at the posterior third. Epinotum elevated, in profile a little
longer than deep, triangular; apex armed with a pair of strong, slightly
curved spines; base and declivity subequal in length, the surface of
the former convex, of the latter flat. Petiolar node in profile one and
a fourth times thicker than long; from above twice as broad as long;
the posterior surface sloping.

Sculpture similar to that of D. afellaboides, the head, pronotum, and
epinotum sublucid; petiolar node very coarsely and densely rugose;
mesothorax shining, less rugose; mesopleurae transversely striate.

Gaster finely punctate and shining. Mandibles punctate. The
body, antennal scapes, and legs with rather abundant, erect, stiff
hairs.

Color dark reddish brown to black, gaster black, legs lighter. Pile
of body and legs grey to white, of antennal scapes, black.

Described from eleven workers taken at Manaos. They were feed-
ing on the exudation of a small shrub along a trail in the forest.

This species is closely related to D. atellaboides, but differs in its
smaller size and more slender form, the longer neck and petiolar node
and the shining gaster. Dolichoderus rosenbergi Forel from Ecuador,
also has the gaster shining, but the first antennal joint in four times as
long as broad (in D. imbicellus it is less) and the size is larger (13 mm.).

154. Dolichoderus {Dolichoderus) atellaboides (Fabricius).

Several workers were collected at Abuna, and Madeira-Mamore
Camp 39. It is much less abundant than D. decollatus.

155. Dolichoderus (Dolichoderus) imitator Emery.

This is the smallest and most delicate species of the group, and the
least common. Five workers were taken at Madeira-Mamore Camp
39.

The following key serves to separate the Brazilian species of the
subgenus Dolichoderus.

Occiput not prolonged into a distinct neck 1.

Occiput prolonged into a distinct neck 2.

1. Large, coarse species, epinotal spines long, acuminate; node

flattened above, unarmed decollatus Smith.

Small, delicate species, finely sculptured, epinotal spines very

MANN: THE ANTS OF BRAZIL. 461

short; node in profile triangular, acuminate at apex, which is
armed with two small teeth imitator Emery.

Color black, petiolar node in profile flat above, twice as long as
thick, epinotal spines rounded, slender rugosus Smith.

Color, in part, red. Petiolar node in profile less than twice as
long as thick, not flat above; epinotal spines somewhat flat-
tened and slightly rugose toward base 3.

Node from above not twice as long as broad; gaster densely
punctured and opaque atellahoides Fabr.

Node from above more than twice as long as broad; gaster shin-
ing 4.

First antennal joint 4 times as long as thick; length 13 mm.
(Ecuador) roscnhcrgi Forel.

First antennal joint less than 4 times longer than thick; length
10 mm ivibecilhis Mann.'

156. Dolichoderus (Monads) hispinosus (Olivier).

Numerous colonies were found at Para, Itacoatiara, Porto Velh(^
and Madeira-Mamore Camp 39.

This is the commonest, and the most widely distributed species in
the genus. It builds carton nests, sometimes of large size, in the
branches of trees. The larger of these nests are solidly constructed,
similar to certain termiiaria. I am not certain that some which 1
examined were not termite nests that had been preempted by the ants.
AYhen the formicary is disturbed the workers defend it very pugna-
ciously. They bite hard enough to be disagreeable and the colonies
are very populous. I have, on several occasions actually been driven
from the near vicinity of a nest by this species.

157. Dolichoderus (Monads) spinicollis (Latreille).

^Yo^kers of this singular species were encountered only once, at
Madeira-Mamore Camp 41. They were found shortly before twi-
light, moving in a file along a slanting tree-trunk and going up
into a tangled mass of vines. Each was carrying in its mandibles
a portion of some fluffy, waxy substance. I could not locate the nest,
and did not see any more of the ants, though I returned to the same
locality several times for further search.

This species is characterized by the very long, acute spines which

4(]l2 bulletin; museum of comparative zoology.

project laterally and upward from the sides of the pronotum. The
meso- and epinotum each has a stout, triangular spine at the poste-
rior corners, and the node is armed at apex with a slender acuminate
spine as long as the node. The whole body is covered with fine
yellow pubescence.

158. Dolichoderus (Monads) laminatus Mayr subsp. lutcivcntris

Emery.

A female from Madeira-Mamore Camp 41 agrees with Emery's
description. It measures 5.5 mm. in length. The head, thorax,
and petiole are black, and the gaster testaceous, each segment of the
latter with a fuscous border. The legs are testaceous.

159. Dolichodertis (Monads) varians, sp. nov.

Worker. Length 5 mm.

Near D. lamellosus Mayr. Head oval, excluding mandibles, a little
longer than broad, the sides very convex; occipital angles narrowly
rounded, the border narrowly concave. Clypeus flat above, with
rounded anterior border. Frontal area very large, triangular. Fron-
tal carinae thin, slightly elevated, extending to opposite posterior
third of eye. Eyes small, convex, located in front of sides of head, a
little behind the middle. Antennae slender, the scape thickened
toward apex, extending a third its length beyond the occipital corners;
all the funicular joints longer than broad. Pronotum transverse,
broadest in front, the anterior angles drawn out to form acute tri-
angular spines, which are flattened above; anterior border convex at
middle, concave at sides; sides sharply but not broadly margined.
Mesonotum one and a half times as long as broad, narrowed behind;
the surface slightly concave; sides with a narrow elevated margin.
Base of epinotum from above triangular, one and a half times as long
as broad, sides roundly margined; apical portion transversely de-
pressed and posterior to this elevated into a very thin lamella; de-
clivity a little shorter than the base.

Petiolar node twice as broad as long; in profile twice as high as
thick; anterior and posterior surfaces slightly convex, the apex
rounded in front, behind elevated into a very thin broad lamella.
Gaster short and thick.

Subopaque. Head, thoracic dorsum, and anterior surface of

MANN: THE ANTS OF BRAZIL. 463

petiolar node shallowly, densely rugulose. Posterior surface of node
and the gaster finely, densely punctate, the latter regularly granulose.

Body thickly covered with short, fine, erect pile. Antennae mi-
nutely pubescent.

Color black; antennae and legs ferruginous.

Described from a single worker taken at Porto Velho. In color
and general form, as well as in sculpture D. varians resembles D.
lameUosus, from which it differs in the shape of the mesonotum. In
the latter species this is transversely oval, while in D. varians it is
distinctly longer than broad, with a faintly crenulate border. Dolicho-
derus varians is also much more thickly pilose. The other closely
related species, D. laminatus, has much narrower pronotal spines and
finer sculpture throughout.

160. Dolichodcrus (Monads) tristis, sp. nov.
Plate 2, fig. 17.

t

Worker. Length 4.5 mm.

Head, excluding mandibles, a little longer than broad, narrowed
in front, with slightly rounded sides and nearly straight posterior
border; occipital corners broadly rounded. Clypeus rounded, the
anterior border concave at middle. Frontal carinae nearly straight,
subparallel, extending to opposite middle of eyes. Eyes small, feebly
convex, located in front of sides of head a little behind the middle.
Mandibles short and thick. Antennae short, the scape curved,
thickened and flattened toward apex, extending less than one third
its length beyond the occipital corners; funicular joints subequal in
length, the first constricted at base, joints 3-10 cylindrical, about one
and a half times as long as broad; joint 10 as long as broad; apical
joint twice the length of penultimate. Pronotum transverse, nar-
rower than head; the short neck transversely depressed; sides convex,
acutely margined, the spines short and stout. Mesonotum trans-
verse, sides margined. Epinotum campanulate, unarmed, flat above,
the margin projecting laterally and behind; surface of basal portion
rounded. Petiole from above transverse; in profile thicker than long,
flat above, straight in front, concave behind; the posterior apical
border w4th a short, thick, triangular spine which projects upward
and backward; postero ventral surface extended into a triangular
process. Gaster short and thick, considerably broader than the
thorax.

464 bulletin: museum of comparative zoology.

Subopaque. Head regularly, rugosely punctate, each puncture
bearing a short, recumbent glistening white hair. Cheeks and
clypeus rugulose. Mandibles shining, coarsely punctured. Anten-
nae densely punctulate. Thoracic dorsum finely punctate; pleurae
shining; prothoracic pleurae densely punctate, meso- and epinotal
pleurae rugose. Node shining, rugose. First gastric segment densely,
irregularly striolate, the striolae longitudinal in front and transverse
at the posterior border; remaining segments very densely striolate
longitudinally. Legs somewhat shining, punctate.

Thorax and gaster finely pubescent and wath scattered erect hairs.
Legs sparsely pilose.

Color black; eyes, inner border of mandibles and tarsal joints red-
dish. Pilosity brownish.

Described from several workers taken at Abuna.

The curious structure of the petiole distinguishes this from the
other species of the subgenus.

16L Dolichoderus {Monads) debilis Emery.

Several workers from Madeira-Mamore Camp 37 agree closely
with cotypes received from Professor Emery.

162. Dolichoderus (Moriacis) debilis Emery var. rufescens Mann.

The single colony of this very distinct variety was found at Madeira-
Mamore Camp 39, living parabiotically with Odoniomachus affinis
subsp. mayi. It differs from Z). debilis in color, the greater length of
the petiolar spines, and the coarse cephalic sculpture.

163. Dolichoderus (Hypoclinea) abruptus (F. Smith).

Many specimens were taken at Porto Velho, Abuna, and Madeira-
Mamore Camp 41. The workers were found most frequently on
shrubs, where they attended Membracidae.

164. Dolichoderus (Hypoclinea) lugens Emery,

This species, originally described from Bolivia, swarmed in certain
parts of the forest near Porto Velho, and at Abuna, Brazil and Bolivia.
It forages on the ground more than do the other species of the genus,

MANN: THE ANTS OF BRAZIL. 465

but like the others, nests in trees. The colonies must be very large,
judging from the numbers of workers seen together. The workers
are able to exude from the anal glands a large drop of a mustard-yellow
secretion. This is not, so far as I could ascertain pungent, and it had
no effect when applied to my skin.

165. Dolichodcrus (Hypoclinea) bidens (Linne).
Taken at Para, Porto Velho, and Abuna.

166. Dolichodcrus (Hypoclinea) bidens (Linne) var. inferior, var. nov.

Taken at Itacoatiara and Abuna. This variety is somewhat
smaller than typical D. bidens, and is light ferruginous in color. The
punctation on the vertex is much finer.

167. Dolichoderus (Hypoclinea) analis Emery.

Several workers were taken at Para and Abuna. It is less common
than the preceding species.

^ . .

168. Dolichoderus (Hypoclinea) germaini Forel var. garbei Forel.

Many workers from Ceara and Independencia agree with Forel's
description of this variety from Bahia. The head has the sides little
convex and much longer than broad; the pronotum is as broad as long.
In all of the series the legs are light fuscous and in some of the speci-
mens the thorax is considerably lighter than the rest of the body.

169. Dolichoderus (Hypoclinea) ghiliani Emery.
Plate 3, fig. 19.

]]^orker. Length 4.5 mm.

Head one and a third times as long as broad, with convex sides,
rounded occipital corners and straight border. Clypeus rounded at
middle, the anterior border broadly rounded. Mandibles long and
thick, the blades finely dentate. Frontal carinae distinct, subparallel,
not extending to opposite posterior border of eye. Eyes oval, situ-
ated on sides of front, a little anterior to middle of head. Antennae

466 bulletin: museum of comparative zoology.

long and slender, the scape extending one third its length past the
occiput, funicular joints 1-7 subequal, cylindrical, nearly three times
as long as broad, joints 8-11 a little thicker than the others, apical
joint as long as the two succeeding joints together. Pronotum flat,
behind, sloping in front, the anterior part considerably narrowed,
with margined sides; border between anterior and posterior parts
angulate. Mesothorax in front flattened, disc-like; behind sloping
into the broad mesoepinotal impression; the mesothoracic spiracles
are large and form distinct tubercles. Base of epinotum longer
than the declivity, the surface two and a half times as long as broad,
flat and narrowest in front, broadly, shallowly impressed behind,
the impressed part acutely margined; posterior border margined, the
two margins joining in a prominent, rounded angle; face of declivity
rounded. Petiolar node thick, in profile very convex in front, nearly
straight behind, the two surfaces separated by a faint margin; the
anterior surface rounded, the posterior nearly flat. Gaster elongate
and narrower than is usual in the genus. Legs long and slender.

Head, thorax, epinotum, and gaster subshining; seriolately punctu-
late, devoid of pubescence but bearing sparse, long, erect hairs.
Antennae with moi'e abundant and finer, semierect, recumbent pile.
Legs finely pilose.

Gaster shining, with sparse, long, erect hairs.

Color light ferruginous, except the gaster, which is piceous.

Described from a single worker taken at Itacoatiara. It is related
to D. lutosus, but is a more slender species, the epinotal margin is
distinctly concave seen from behind, and the head is much longer,
with very convex sides, and narrowed behind, and not concave at the
border.

170. Dolichoderus {Hypoclinea) chmyipioni Forel var. ornatus,

var. nov.

Worker. Length 6 mm.

Near germaini. Head elongate oval, with convex sides and feebly
concave occipital border, the angles evenly rounded. Clypeus flat-
tened at middle, anterior border straight. Mandibles as in D. (jcr-
maini. Eyes nearly circular, located at sides of front anterior to
middle of head. Antennae slender, the scape bisinuate, extending
one third its length past the occipital border; funicular joints all dis-
tinctly longer than broad, the first three times, the others about twice.
Pronotum flattened, a little longer than broad, the sides rounded in

MANN: THE ANTS OF BRAZIL. 467

contour and feebly margined. Mesonotum oval, about one and one
half times as long as broad, in profile gently rounded from base to the
mesoepinotal impression, which is broad and deep. Epinotum more
than twice as long as broad, with the base a little longer than the
declivity; in profile the base is rounded, the declivity slightly convex;
the surface of the base is convex, except at the apical end, where there
is a narrow shallow transverse impression, posterior to which is a
slightly elevated margin; the basal surface is flat. Petiolar node thin,
in profile convex in front, concave behind, the apex thinly margined;
the margin not extending down the sides.

Head shining, minutely, closely punctate, thinly covered with
short, suberect pile; antennae thinly pilose; mandibles punctate and
pilose.

Thorax, epinotum, and petiole somewhat shining, punctate, simi-
lar to but more coarsely than the head, and more thinly pilose;
the posterior surface of the node with fine transverse striolae and
glabrous," except for several .long hairs at the apex. Thorax shining,
minutely punctate, thinly pilose.

Color black, except the legs, petiolar node, base of first gastric
segment and interrupted bands at the apices of the first and second
segments which are yellow. Apex of femora, tibia? and tarsi slightly
infuscated. Pile gray.

Female. Length 8 mm.

Head elongate oval, distinctly longer than broad, in general similar
to that of the worker. Ocelli large and distinct, arranged in a tri-
angle. Mesonotum flat behind, the sides with a low, rounded margin.
Scutellum flat. Epinotum similar to that of the worker, but the base
is much shorter in proportion to the declivity, slightly longer than
broad. Petiole as in worker. Gaster longer and more slender.

Head and pronotum with foveate punctation, the punctures shallow
and widely separated.

The color is the same as in the worker, except for the mandibles and
a transverse stripe across the clypeus, which are rufous. The tibiae,
tarsi, and tips of the femora are darker than in the worker. Wings
very slightly infuscated, veins pale brown. Pilosity as in worker.

Described from large series of workers from Para and Abuna.
This is a singularly colored variety, quite different from any of the
described forms. In the Para specimens the surface of the epinotal
declivity is yellow, and in those from Abuna it is black. Otherwise
the two are identical.

468 bulletin: museum of comparative zoology.

171. Dolichodcrus {Hypoclinca) lutosus (F. Smith).
Plate 3, fig. 20.

Several colonies, nesting beneath stones were found at Natal and
Baixa Verde. Those from the latter locality are small in size (length
3.5 mm.) and are evidently from an incipient colony. A figure of the
worker is given in Plate 3, fig. 20.

Female. Length 6 mm.

Ocelli very small, arranged in an equilateral triangle. Base of
epinotum shorter than declivity.

Color similar to that of worker but more pronounced. First three
gastric segments with a median longitudinal fuscous line; apex of
first segment with a narrow transverse fuscous band ; apical two thirds
of the second and border of the third segments fuscous. Wings
hyaline ; veins and stigma brown.

Male. Length 3.75 mm.

Head, excluding the mandibles, as long as broad, convex behind
and at sides, with large, moderately convex eyes, which occupy
about half the sides. Frontal area distinct, triangular; carinae fine.
Clypeus convex at middle, broadly concave at middle of anterior
border. Mandibles well developed, stouts, as long as the distance
from their base to the eye, blade with several fine teeth anteriorly.
Vertex elevated into a broad tubercle on wdiich the large ocelli are
located; in front of this, beneath the median ocellus the front has a
deep, triangular impression. Antennae short, extending to base of
gaster; first funicular joint a third the length of the second. Thorax
robust, mesonotum evenly rounded above, with feeble Mayrian fur-
rows. Epinotum short, base rounded, declivity flat, the two surfaces
separated by an obtuse angle. Petiole twice as high as thick, anterior
surface rounded, posteriorly nearly flat, the apex narrowly, thickly
margined. Gaster similar to that of worker, but flattened above.
Genitalia small. Hypopygium small, broadly triangular. Cerci
very small. Legs slender. ^Yings long and narrow.

Body shining, very minutely punctate. Mandibles punctate, with
several long hairs. Body without long hairs. Thorax and abdomen
with sparse, regular, very minute scale-like white hairs. Funiculus
densely pubescent.

Color dark fuscous, gastric segments at base lighter.

MANN: THE ANTS OF BRAZIL. 469

172. Aztcca schumanni Emery subsp. duhia, subsp. nov.

Plate 2, fig. 15.

Worker major. (Plate 2, fig. 15). Length 2 mm.

Head subquadrate, one and a third times longer than broad, slightly
narrowed in front, with feebly convex sides, broadly rounded occipital
corners and narrowly excavated border. Clypeus convex, the anterior
border broadly bisinuate. Antennal scapes extending about two
thirds the distance to occipital corners; strongly arcuate and thick-
ened toward apex; funicular joints 5-10 as broad as long. Eyes at
sides of head, well in front of the middle. Pro- and mesothorax as in
A. schumanni. Epinotum in profile rounded; the posterior portion
of base flattened. Node in profile twice as high as thick, evenly
rounded above. Antennal scapes and tibiae without erect pile.

Head, thorax, and abdomen with fine, long, appressed pubescence
and sparse, short, erect hairs.

Color fuscous, thorax lighter, gaster darker.

Worker minor. Length 2 mm.

Head longer than broad, the width at occiput equal to that at
clypeus; sides slightly convex, posterior border very feebly concave.
Clypeus convex, the anterior border truncate at middle, slightly
produced and rounded at corners. Eyes small, located in front of
sides anterior to middle of head. Antennal scapes extending about
two thirds the distance to occipital corners, funicular joints 6-10 as
broad as long. Thorax and epinotum similar to that of the worker
major, but the latter is somewhat flatter. Node and gaster as in the
worker major.

Subshining, finely punctate. Body above with sparse, short pile
and pubescence. Antennae and legs pubescent, without erect pile.

Color fuscous.

Male. Length 3.5 mm.

Head about as long as broad; sides and posterior border convex.
Mandibles slender, acuminate. Anterior border of clypeus rounded.
Eyes rather large and convex. Ocelli prominent, the lateral ones
situated at opposite ends of a transverse elevated tubercle, which is
rounded in front and behind. Antennae short and stout; first and
second joints subglobose, transverse, the second the longest; third
joint one and a half times longer than broad, very much thickened;
joints 4-6 distinctly longer than broad, joints 7-12 proportionately
shorter; apical joint one and a half times as long as penultimate.

470 bulletin: museum of comparative zoology.

Thorax robust, the anterior surface of the pronotum decUvous in
profile. Petiole and gaster similar to those of the worker minor.

Body and legs without erect pile, sparsely pubescent. Antennae,
excepting the first two joints, very densely covered with rather long,
erect pubescence.

Color black, legs and antennae dark fuscous. Wings hyaline.
Veins and stigma fuscous.

Described from four major workers, three minors, and a male from
Itacoatiara. This form differs from the typical A. schumanni in not
having erect hairs on the antennal scape, the head is less narrowed
in front, and the clypeus is not depressed at the middle. Azteca
schumanni var. taediosa Forel is more robust and has the head less
excavated behind and the antennal scapes longer.

173. Azteca miillcri Emery subsp. terminalis, subsp. nov.

Plate 2, fig. 16.

Worker major. Length 4 mm.

Head, excluding mandibles, as broad as long, narrowed in front,
with strongly convex sides, especially opposite the eyes, narrowly
rounded occipital corners and deeply excavated border. Clypeus
convex, the anterior border straight, except at corners, where it is
slightly produced and rounded. Mandibles thick, with six strong
teeth. Antennal scapes barely extending to occipital corners. Meso-
notum in profile evenly rounded. Epinotum with subequal base and
declivity, the former broadly flattened. Node rather low, rounded
above. First segment of gaster depressed in middle at base.

Shining, densely punctate throughout. Pubescence abundant, long
and recumbent. Pile of the scapes sparse and short, that of the body
longer.

Color very dark fuscous; terminal half of antennae yellow, the color
becoming more intense at apex.

Described from a series taken at Madeira-Mamore R. R. Camp 39.
The broadly flattened epinotum and the peculiar coloration of the
antennae distinguish this subspecies. It is evidently close to var.
A. nigella Emery from southern Brazil, but is larger and differently
colored.

174. Azteca aurita Emery subsp. silvae Forel.

A single colony of this distinct subspecies was found at Para, the
type locality.

MANN: THE ANTS OF BRAZIL. 471

175. Aztcca velox Forel var.

Colonies were found at Manaos, Abuna, Porto Velho, and Madeira-
Mamore R. R. Camps 39 and 41. There is considerable variation in
the amount of infuscation on the vertex.

176. Azteca angusiiceps Emery.

A single dealated female from Itacoatiara agrees closely with
Emery's figure and description of this species.

177. Azteca alfaroi Emery var.

Several workers of a variety of this species were taken at Abuna.
These are very close to var. A. aequilata Forel, but the sides of the
head are more convex.

178. Azteca trigona Emery.

Numerous workers were taken at Manaos and on the Rio Madeira
at Camps 35, 39, 41, 43, Porto Velho, and Abuna.

179. Azteca trigona Emery subsp. mathildae Forel.

This was the most abundant Azteca on the Rio Madeira. Speci-
mens were taken at Abuna, Porto Velho, and Madeira-Mamore
R. R. Camp 43. At Itacoatiara many colonies were nesting in high,
buttressed trees near the river bank.

180. Azteca trigona Emery subsp. mathildae Forel var. spuria Forel.

A number of colonies, in small carton nests, six or eight inches in
length, were found at Ceara-Mirim. Other colonies of a variety
identical with this were found at Abuna on the Rio Madeira. The
latter attend Coccidae of several species, over which they build sheds.
The coccids which I observed were on small bushes near the trees on
which the formicaries were built. The ants energetically defended
the sheds and the Coccidae.

181. Azteca barbifex Forel.

Very abundant at Abuna and Madeira-Mamore Camps 28, 39 and
41. The type specimens are from the Rio Purus and the species is
probably widely distributed throughout the upper Amazonian region.

472 bulletin: museum of comparative zoology.

182. Azteca chartifcx Forel var.
Several workers of a variety of this species were taken at Abiina.

183. Azteca fasciata Emery var. similis, var. nov.

IVIinor and major workers from colonies found at Madeira-Mamore
Camps 39 and 41 differ from Emery's description and figure of A.
fasciata from Santarem in color and in having the antennal scape
noticeably longer. In the smaller worker of A. fasciata it extendi
barely past the occipital corners while in A. similis it exceeds these
corners by a full third of its length. In A. similis the larger workers
have the head, pronotum, anterior femora, and antennal scapes red,
and the rest of the body dark fuscous. The smaller workers have the
vertex, and part of the pronotum infuscated. In other characters
it agrees closely with typical A. fasciata.

184. Azteca lanuginosa Emery subsp. pruinosa, subsp. nov.

Worker. Length 3.5 mm.

Head, excluding mandibles, as broad as long, appreciably narrowed
in front, with strongly convex sides, narrowly rounded occipital corners
and shallowly excavated border. Clypeus convex, the anterior border
bisinuate, projecting and rounded at middle. Mandibles with a thick,
blunt subapical tooth and five small teeth on the blade. Antennal
scapes in the largest workers barely attaining the occipital corners,
in the smaller ones slightly surpassing them. Mesonotum very con-
vex and elevated. Mesoepinotal suture strongly impressed. Node
in profile deeper than thick, evenly rounded above.

Mandibles subopaque, densely striolate longitudinally. Head,
thorax, and abdomen subopaque, finely, densely punctate.

Head and body evenly covered with fine pruinose pubescence. No
erect hairs present, except a very few on the gaster.

Color dark fuscous; clypeus, antennal scapes and legs lighter.

Described from a number of w^orkers taken at Abuna. This form is
less shining than A. lanuginosa, the pubescence is more abundant and
is closely appressed and not lanuginose in character. Otherwise the
variety agrees with the typical form.

/

MANN: THE ANTS OF BRAZIL. 473

185. Dorymyrmex pyramicus (Roger).

Many workers were found at Natal, and Itacoatiara. In the yard
of our house at Natal this and the succeeding variety were very
abundant, in small crater nests.

186. Dorymyrmex pyramicus (Roger) subsp. flavus McCook.

Very common at Natal. This and the preceding have a wide
distribution, ranging from Illinois in the United States to Argentina.
Strangely, in spite of the adaptive nature of this ant, it has not spread
out of the Americas.

187. Tapinoma melanocephahim (Fabricius).

Many specimens of this common tropicopolitan species were found
at Para and Porto Velho.

Camponotinae.

188. Brachymyrmex coactus Mayr.

One colony was found at Independencia, nesting in a twig.

189. Brachymyrmex admoius Mayr.
One colony was taken at Ceara-Mirim.

190. Brachymyrmex pictus Mayr.
Taken at Para and Manaos.

191. Myrmelachista (Decamera) bambusarum Forel.

Workers taken at Itacoatiara agree with Forel's description of this
species from Sao Paulo. The type specimens were nesting in bamboo.

192, Gigantiops destructor (Fabricius).

Found commonly at Para, Abuna, Porto Velho, and Madeira-
Mamore Camps 39 and 41. In life this was one of the most

474 bulletin: museum of comparative zoology.

attractive ants encountered. It lives always in the forest, where it
forages either among the branches of trees or on the ground. The
movements of the foraging worker are rapid, comparable to those of
some of our species of Cicindela, and the bicolored antennae are kept
constantly in motion.

The female is very similar to and scarcely larger than the biggest
worker and the eyes and ocelli are equally well developed in both.

193. Prenolepis {Nylcmderia) longicornis (Latreille).
Santarem, Natal, and Maranhao.

194. Prenolepis {Nylanderia) vividula (Nylander).
Several colonies were found at Para.

195. Prenolepis (Nylanderia) fidva Mayr.
Itacoatiara.

196. Prenolepis {Nylanderia) steinheili Forel.
Independencia.

197. Camponotus (Mynnoturba) maculatus (Fabricius) subsp. fusco-

cinctus Emery.

Several workers and females were found beneath bark at Natal.

198. Camponotus (Myrmoturba) macidatus subsp. fryi, subsp. nov.

Plate 6, fig. 52.

Worker major. Length 10 mm.

Close to subsp. C. spengleri Forel. Head a littl;e longer than broad,
truncate behind, with broadly rounded posterior corners and slightly
convex sides. Clypeus strongly carinate; the anterior border notched
at middle. Cheeks in front broadly rounded. Thorax and petiolar
node in profile thicker than in the other forms of C. macidatus, the
anterior face of the latter very convex. The front and vertex are

MANN: THE ANTS OF BRAZIL, 475

sparsely punctate, the punctures comparatively coarse. Pubescence
is lacking and pile sparse.

Color brown, the head darker; the basal parts of gastric segments
are transversely banded with light ferruginous.

Described from a single specimen taken at Madeira-Mamore
Camp 39. This subspecies is distinguished from the others by the
thicker thorax and petiolar node and the very sparse pile.

199. Camponotus {Myrmoturha) maculatus subsp. ahunanus,

subsp. nov.

Plate 6, fig. 44.

Worker major. Length 7 mm.

Head, excluding mandibles, longer than broad, narrowed in front,
with convex sides; posterior border excised, straight at middle, the
angles prominent, and clypeus longer than broad, very broadly
carinate at middle, the anterior border bilobed. Mandibles stout,
with five rounded teeth. Antennae slender, the scapes barely reach-
ing to occipital corners of the head. Thorax slender; pronotum
distinctly longer than broad. Epinotum from above four times as
long as broad; in profile slightly rounding from base to declivity, the
two surfaces joining in a broadly rounded angle. Petiolar node
wedge-shaped in profile, the anterior surface rounded, the posterior
nearly flat; seen from behind its margin is evenly rounded. Legs
short; the tibiae not compressed.

Subshining; very finely shagreened, the head and pronotum less so
than the rest. Mandibles sublucid, with fine punctures and few short
hairs. Front with a few coarse superficial punctures. Pubescence of
the head very minute, sparse, and scale-like; a few short erect hairs
on the front and occiput. Thorax and abdomen with silky pubes-
cence, which is most abundant on the gaster. Pile long and sparse
on the thorax, shorter and abundant on the gaster.

Color tes.taceous; mandibles, antennal scapes (except tip), tarsi,
and a narrow transverse band at the apex of each gastric segment dark
fuscous. Pile and pubescence yellow.

Described from two major workers from Porto Velho. Possibly
this should be considered a distinct species. The occipital angles of
the head are unusually narrow, the clypeus more deeply notched and
the antennal scapes are shorter than in the other forms of C. (M.)
maculatus.

476 bulletin: museum of compakative zoology.

200. Camponotus (Myrmoturba) mclanoticus Emery var. substitutus

Emery.

Numerous workers were found at Natal, Independencia, and in the
Maranguape Mountains. The species nests beneath stones.

201. Camponotus {Myrmothrix) nifipcs (Fabricius).

The typical form of this species was common, nesting in logs and
beneath bark, at Ceara-Mirim, Para, and Porto Velho.

202. Camponotus {Myrmothrix) abdominalis (Fabricius).

Very common at Natal, Ceara-lSIirira, Baixa Verde, Para, and
Madeira-Mamore Camps 39 and 46.

. 203. Camponotus (Mynnothrix) abdominalis var. atriceps F. Smith.
A single colony was taken at Abuna, Bolivia.

204. Camponotus {Myrmothrix) rapax (Fabricius).
Plate 5, fig. 38.

Worker major. (Plate 5, fig. 38). Length 12 mm.

Head one and a third times longer than broad, narrowed in front,
posterior angles narrowly rounded, occipital border narrowly and
rather deeply concave; the sides straight and subparallel until a little
in front of middle, then convergent. Mandibles large, elongate, with
six teeth on the blade. Clypeus a little broader than long, attaining
side margin of head, strongly carinate, the anterior border of middle
convex. Frontal area distinct, quadrangular. Frontal earinae evenly
curved from base to end, which is opposite the middle of eye. Eyes
small, rather flat. Antennae slender, the scapes bent at middle,
extending three eighths their length beyond the occipital corners;
funicular joints long and cylindrical, gradually decreasing in length
apically. Pronotum broader than long, rounded at sides and above,
in front finely margined. Mesonotum slightly longer than broad, nar-
rowed behind, the base slightly less than twice the breadth of posterior
portion. Epinotum at base divided by a distinct transverse suture;
evenly rounded in profile; narrow above, three times as long as liroad.

MANN: THE ANTS OF BRAZIL. 477

Petiolar node from above transverse, rounded at sides and in front;
in profile narrow, two and one half times higher than thick, rounded
at apex, the posterior surface nearly straight, the anterior slightly
more convex. Gaster elongate egg-shaped. Legs long and slender.

Subopaque, the whole body finely shagreened, somewhat more
coarsely on the head and more finely on the petiolar node. Mandibles
coarsely punctate and sparsely setose. Antennal funiculus coarsely
punctate.

Body with long, recumbent, silky, glistening hairs and very long
erect pile. The appressed pubescence is most abundant on the
gaster, thoracic pleurae, and head and absent from the petiolar node.
Legs with short semidepressed hairs, femora sparsely beset with long
stiff hairs. Antennal funiculus pubescent; scape with short semi-
depressed and sparse, longer, erect, stiff hairs.

Color black except a transverse reddish brown patch on the occiput
and the dorsum of the gaster, which is light ferruginous. Pubescence
yellow, pile brown.

]]'orker minor. Length 10 mm.

Head twice as long as broad, as broad in front as behind, with
slightly convex sides and narrowly rounded occipital border. Anten-
nal scapes extending nearly two thirds their length past the occipital
borders. Thorax shaped much as in worker major. Posterior surface
of petiolar node shallowly impressed at middle.

Sculpture, pilosity, and color much as in worker, but the head is
entirely black.

Described from one worker major and a series of minor workers
from Porto Velho and Madeira-Mamore Camps 39 and 4L This
distinct species is generally distributed throughout the Amazon region
and the northern parts of South America, but it is not common locally.

205. Camponotus {Myrmothrix) leydigi Forel.

Workers were found on tree trunks at Independencia and Manaos.

206. Camponotus (Myrmothrix) wheeleri, sp. nov.

Plate 6, fig. 49-5L

Worker major. (Plate 6, fig. 50). Length 12 mm.
Head very large, nearly as broad as long, narrowed in front, with
slightly convex sides, elongate, narrowly rounded occipital corners

478 bulletin: museum of comparative zoology.

and sharply truncate posterior border. Eyes small, slightly convex.
Mandibles small, short, with four coarse teeth. Clypeus about as
broad as long, carinate, the anterior border broadly concave. Frontal
area distinct. Frontal carinae more approximate in front than behind.
Antennae short and slender, the scapes curved, extending one eighth
their length past the occipital corners; funicular joints 1-3 subequal,
more than three times as long as broad, joints 4-6 subequal, slightly
shorter, than the first three, the rest of the joints still shorter and sub-
equal in length. Pronotum one half as broad as the head, the anterior
border and the front of sides with a rounded carina. Meso- and
epinotum in profile evenly arched above; from above, rounded at
sides. Petiolar node in profile about as thick as long, moderately con-
vex in front, nearly straight behind; from above as broad as long,
evenly rounded in front and behind. Legs very long, the tibiae and
metatarsi broad and very much depressed.

Sublucid throughout, the whole body finely shagreened, more
coarsely on the head; with a covering of silky, recumbent pubescence,
which is less abundant on the head, pro- and mesonotum, and the
petiolar node; with long, stiff, erect hairs on the head, thorax,
gaster, and legs; the head in addition bears shorter, subei-ect hairs.

Body and legs black, the gaster on account of the dense, silky
pubescence has a yellow sheen ; head reddish brown, the antennal
scapes black. Pubescence and short fine pile yellow; coarse hairs
black.

Worker minor (Plate 6, fig. 49, 51). Length 9 mm.

Head a little longer than broad, narrowed in front, the sides nearly
straight; occipital corners broadly rounded; posterior border straight.
Clypeus with straight, projecting anterior border. Antennae long,
the scapes extending three fourths their length past the occipital
corners of the head. Thorax and petiole similar to that of worker
major. Legs as in worker major.

Sculpture, color, pile, and pubescence as in worker major, but the
pubescence of the gaster is somewhat finer.

Described from a small series which was collected at Madeira-
Mamore Camp 41. The ants were nesting in a hollow palm tree and
it was necessary to smoke them out. Their movements in life were
very active.

The peculiar shape of the thorax, the strongly flattened long legs
and the color are very distinctive.

MANN: THE ANTS OF BRAZIL. 479

207. Camponotus {Mynnmnhlys) hurtoni, sp. nov.
Plate 6, fig. 45, 46.

Worker maior. Length 0 mm.

Head, excluding mandibles, longer than broad, broadest at occiput;
posterior corners narrowly rounded, the border slightly convex; sides
in front of eyes straight. Head in profile two thirds as thick as long.
Clypeus a little broader than long; anterior border narrowly rounded
at middle; surface very convex, with a strong carina at middle. Fron-
tal carinae weak, extending to opposite the anterior border of eyes.
Eyes small, convex, situated on sides back of posterior third of head.
Antennae long and slender, scape extending half its length past the
occipital corners, funicular joints long, cylindrical. Mandibles short
and thick, the blade with four teeth. Pronotum twice as broad as
long, the surface rather flat, sides nearly straight; anterior angles
evenly rounded; sides and anterior border with a rounded margin.
No mesoepinotal suture; the mesoepinotum rounded above, then
declivous to base, in profile twice as high as thick. Node wedge-
shaped, twice as deep as thick, the anterior and posterior surfaces
feebly convex; narrowly rounded above.

Sublucid, head, thorax, and abdomen very densely, transversely,
striolately punctate, with long silky pubescence, which is rather sparse
except on the front of head and the pronotum, and abundant, stiff,
erect pile, which is very long. Node without pubescence but bearing
several long hairs at the apex. Antennae pubescent, devoid of pile.
Femora and tibiae sparsely pilose.

Color black; mandibles and scape rufous. Pile and pubescence
white.

Described from three workers taken at Madeira-Mamore Camps
39 and 41.

208. Camponotus {Myrmavihlys) novagrenadensis Mayr.
Two major workers were taken at Abuna.

209. Camponotus {Myrmamblys) claviscapus Forel.

Many major and minor workers of this species, from Natal and
Ceara-Mirim, agree closely with Forel's description of the types, which
were from Trinidad.

480 bulletin: museum of comparative zoology.

210. Camponotus {Myrmohrachys) adpressisetosus Forel.

Plate 6, fig. 48.

Worker major. (Plate 6, fig. 48). Length 9 mm.

Head, excluding mandibles, longer than broad, with narrowly
rounded occipital corners and concave border; sides nearly straight
and subparallel for half their distance from occiput to anterior border,
then convex and convergent. Clypeus longer than broad, strongly
keeled at middle, the sides straight and parallel; anterior border
broadly concave at middle. Frontal area distinct, elongate, triangu-
lar. Frontal carinae extending to nearly opposite the posterior borders
of eyes. Eyes small, very flat, located in front of sides behind the
middle. Mandibles 4-dentate. Antennae short, slender, the scapes
curved, extending three fourths the distance to occipital angles, funi-
cular joints cylindrical, joints 2-10 subequal. Pronotum transverse,
sides evenly rounded, surface slightly convex. Mesonotum flat above,
a little broader than long, evenly rounded at sides. Mesoepinotal
suture faintly impressed. Epinotum from above three times as long
as broad; evenly rounded above and at sides, without distinct base
and declivity. Petiolar node twice as high as thick, rounded in front,
nearly straight behind, above narrowly rounded. Gaster short and
stout, oval. Legs long, tibiae moderately depressed.

Subopaque, very finely shagreened throughout, with a covering of
tine silky pubescence, which is a little more abundant on the gaster
and thorax than on the head. The head bears short, stiff suberect
hairs, in addition to much longer erect hairs. The latter are present
also on the thorax and gaster, where they form a long brush on the
epinotum and the apex of node. Antennae without pubescence or
pile. Mandibles subshining, finely punctate, with sparse, short hairs.
Legs with sparse, fine recumbent hairs; without erect hairs.

Color black; antennae and mandibles brown, recumbent pile and
pubescence golden.

Worker minor. Length 7 mm.

Head, excluding mandibles, longer than broad, narrowed in front,
with slightly convex sides, broadly rounded posterior corners and
straight occipital border. Clypeus as broad as long, strongly keeled
at middle. Mandibles 7-dentate. Antennal scapes extending half
their length past the occipital corners. Eyes large, moderately convex,
located at posterior third of head. Thorax much as in major worker,
but the mesoepinotal suture less impressed. Rest as in worker major.

MANN: THE ANTS OF BRAZIL. 481

Many workers were taken at Ceara-Mirim,' Natal, and Maranhao,
from nests beneath stones.

211. Camponotus (Myrmobrachys) crassus Mayr.

This was the commonest of the genus on the east coast, where many
workers, females, and males were taken at Natal, Ceara-Mirim,
Baturite Mountains, and Maranhao.

212. Camponotus {Mynnomalis) depressus (Fabricius).

Plate 5, fig. 42.

Taken at Abuna, Bolivia and Madeira-Mamore Camp 39. This
is one of the most singular ants, on account of its very elongate, flat
body and the extremely long legs. The few workers which I ob-
served were running about on leaves.

213. Camponotus (Myrmepomis) sericeiventris (Guerin).

Abundant on the Rio Madeira at Abuna and Camps 39 and 41.

Some years ago Mr. E. J. Newcomer gave me a live worker major
of this species which was found in a restaurant at Palo Alto, Cali-
fornia, having no doubt been imported with bananas from Central
America.

214. Camponotus (Myrmorhachis) latangulus Emery.

Workers from Para, ]\Ianaos, and Madeira-Mamore Camp 39 are
a trifle lighter in color than a series from Peru, but are otherwise
identical.

215. Camponotus (Myrnieurynota) heathi, sp. nov.
Plate 5, fig. 40, 41.

Worker major. (Plate 5, fig. 41). Length 5.5 mm.

Head nearly as broad as long; sides straight from occipital corners
to three fourths the distance to anterior border, then concave; rounded
anteriorly. Occipital corners broadly rounded, the margin nearly
straight. Frontal carinae narrow, not much elevated. Frontal area
broad, slightly convex, anteriorly as broad as the base of clypeus,

482 bulletin: museum of comparative zoology.

with faint longitudinal impression. Clypeus small, longer than broad,
anterior border truncate, strongly carinate at middle and depressed on
either side. Mandibles short and thick; bluntly dentate. Eyes
small, oval, rather flat, situated much posterior to middle of sides
of head. Antennae short, stout, the scape strongly inflexed, thickened
at apex, barely extending to occipital corners; funicular joints thick,
subequal. Pronotum as broad as long, constricted in front, the sides
margined, anterior border strongly margined. Sides at middle
straight, in front concave to apex, with an angle between ; narrowed
and rounded at base. Promesonotal impression distinct. INIesono-
tum triangular, rounded at sides, and slightly rounded above. Epino-
tum rounded at sides; at base with a large projection, which, seen
from the front, has straight sides and truncate apex; from the sides
this is seen to be composed of two tubercles, the one in front about
half as large as the other. From the apex of this spine the epinotum
in profile is concave to a point a little less than half the distance to
base, where it is armed with another projection. In profile the ante-
rior surface of this is convex, the posterior concave. This projection
is twice as high as the one anteriorly and at the apical end bifurcate
into two conical spines. From this to the marginate base the epino-
tum is concave. Petiolar node in profile thick at base and at about
one third the distance from base to apex strongly constricted; apex
and sides thinly marginate, the middle and the apex of sides with
conical projections; anterior surface rounded, posterior deeply con-
cave. Gaster elongate, sides slightly convex, the width of the first
three segments subequal. Legs long, rather stout.

Head and thorax subopaque, minutely and densely punctate,
the punctures on mesonotum and epinotum coarser. Head and
pronotum with very sparse pubescence and very sparse, scattered
semierect pile. Mandibles shining, with coarse punctures. An-
tennae finely pubescent, the apex of scape with a few erect hairs.
Posterior surface of node coarsely striolately punctate, surface granu-
lar in appearance. Gaster subopaque, densely punctate, with scat-
tered pubescence and a few erect hairs. Tibiae with sparse recumbent
pubescence, but legs otherwise without hairs.

Gaster above dark brown to black, each segment with a narrow
apical border of yellow; rest of body and legs ferruginous, tibiae,
tarsi, and antennae somewhat darker. Pile and pubescence gray.

Worker minor. (Plate 5, fig. 40). Length 4.5 mm.

Head narrowed anteriorly; sides in front of eyes slightly convex
to the anterior border, which is rounded; occipital corners very

MANN: THE ANTS OF BRAZIL. 483

broadly rounded; occipital margin straight. Frontal carinae well
elevated in front; frontal area convex. Clypeus rounded in front,
longitudinally carinate at middle. Mandibles thick, the blade
with five teeth. Eyes oval, convex, located at middle of sides of head.
Scape of antennae extending nearly half its length beyond occipital
corners, slightly bent, and thickened at apex; joints of funiculus
subecjual in length, each about twice as long as broad. Pronotum
convex, one and one half times as broad as long, broadest in front;
anterior border convex, posterior concave; sides in front broadly
margined, the anterior corners laminate. Mesonotum obliquely
flattened at sides, carinate at middle; the carinae, at junction of meso-
and epinotum extended into a long projection which is bent backward
and flattened at apex. i\t the l:)ase of the pronotum, on the sides is a
small, but distinct tubercle. Epinotum carinate at middle; the sides
very oblique. Slightly posterior to middle of epinotum is a second
projection, longer than the first, and thicker at base; seeipin profile
this is bent backward; it is tuberculate on the front surface at a little
less than half the distance to apex. The apex is deeply bifurcate for
nearly one half the length of spine. From the base of this projection
the epinotum is concave to base. Node of petiole, seen in profile,
longer than high; seen from behind about as thick as high. The
apex slightly margined, with three long, acuminate spines, one at
apex, and others at middle of sides. Posterior surface of node flat-
tened. Gaster elongate, sides slightly convex. Legs long, rather
stout.

Head, thorax, and epinotum subopaque, finely punctate. The
punctures are very dense on the head and pronotum, giving these
portions a velvety appearance. Head and pronotum with regular,
recumbent pubescence, the head with a very few short, erect hairs.
Antennae sparsely pubescent. Surface of posterior declivity of peti-
olar node granular. Gaster densely, minutely punctate, with thin
pubescence and sparse, short erect pile. Legs sparsely hairy.

Head above fuscous; each occipital corner with a ferruginous
blotch that extends in front of and beneath eye. Front of head,
from a short distance beneath insertion of antennae yellow. Mandi-
bles light ferruginous. Antennae fuscous, the scape at basal half
ferruginous. The apical border of first gastric segment, the remaining
segments of gaster and the apices of tibiae fuscous, rest of body and
legs ferruginous. Pile and pubescence white.

Porto Velho, Abuna, and Camp 39 on the Madeira -Mamore R. R.
The specimens were found running about on leaves.

484 bulletin: museum of comparative zoology.

The subgenus Myrmeuryota includes the species which have the
meso- and epinotum armed with spines. Of the nine known species,
all excepting C. (M.) hcathi and C. (J/.) curynohis Forel are West
Indian. The latter species, described from a worker minor from
Tonantins in Brazil has the pronotum much broader than in C. (J/.)
heal hi, the epinotum is different and the petiolar node is not spinose.
Like C. (M.) cristophei and C. (i/.) toussainti, which I found on leaves
and tree trunks in Hayti, C. (M.) hcathi is probably an arboreal
species.

I have much pleasure in dedicating this extraordinary species to
my former professor, Dr. Harold Heath, a member of the Stanford
Expedition and my companion on many collecting trips.

216. Dendromyrmex traili May. rufogastcr, var. nov.

Several-'specimens in the Wheeler collection from Bolivia (Stau-
dinger) differ from Mayr's description in having the gaster chestnut
red, instead of black. In other characters these agree well with the
description, and should, I think be considered a distinct variety of
D. traili.

217. Dendromyrmex nididans Smith.

One specimen, referable to this species, taken at Camp 39, Madeira-
Mamore R. R. Dendromyrmex nididans most closely approaches
D.fahricii Roger but has not a dense mat of pubescence on the gaster.

218. Dendromyrmex fabricii Roger.
One worker from Para.

219. Dendromyrmex chartifex Smith var. felis, var. nov.

The worker of this variety differs from the typical form in being
subopaque instead of shining. The whole body is more hairy. The
dense pubescence on the gaster gives it a velvety appearance. The
color is yellowish brown.

Female (dealated). Length 10.5 mm.

Head small, sides in front of eyes almost straight, curving slightly
towards base of clypeus; sides behind eyes convex to posterior margin,
which is slightly concave. Clypeus broad, carinate, the anterior

3\l\nn: the ants of brazil. 485

border truncate. Mandibles with six teeth. Eyes small, situated
posterior to middle of sides of head. Ocelli small. Frontal lamellae
elevated; enclosed area very slightly impressed at middle. Antennae
about two thirds as long as body. Pronotum narrow, anterior border
with raised margin which extends part way along the sides. Mesono-
tum much elevated, declivous in front, flattened above, with faint
indications of margin at sides. Epinotum rounded above and at
sides, margined at base. Node as in worker. Gaster ovate. Legs
long and slender.

Head, thorax, epinotum, and node subopaque, densely, finely
punctate, minutely striolate transversely, with very sparse pubescence,
but with abundant, semierect pile.

Gaster subopaque, densely punctate and striolate, both punctures
and striae being extremely minute. The pubescence is very sparse
and fine; the pile erect, not abundant.

Scape of antennae and legs with many erect hairs; funiculi pubes-
cent.

Color as in worker.

Described from several workers and a female taken at Tumatumari,
British Guiana (Amer. Mus. Nat. Hist, collection).

220. Dendroviyrmex chartifex subsp. mamoreensis, subsp. nov.

Worker. Body much more robust than in the typical form of the
species. Head constricted behind the eyes, sides convex, occipital
border with edge rounded, without trace of margination. Sides of
head in front of eyes evenly convex, narrowed toward base of clypeus.
Clypeus strongly carinate. Mandibles with six teeth on the blade.
Surface of frontal area convex, longitudinally carinate at middle.
Antennae nearly as long as body, rather thick. Pronotum nearly as
broad as long, depressed, and broadly margined anteriorly; seen in
profile, the surface is convex; sides distinctly margined for three
fourths the distance to promesonotal suture. Mesonotum slightly
convex above, rounded at sides. Epinotum flattened, sides marginate.
Petiolar node, seen in profile on anterior side straight and perpendicu-
lar at base, then sloping to apex, with an acute angle between the
two surfaces; posterior surface convex; from behind, evenly rounded
above. Gaster ovate, short and thick. Legs long and slender.

Body nearly opaque, gaster subshining; antennae and legs sub-
lucid.

486 bulletin: museum of comparative zoology.

Head, thorax, and abdomen finely, densely punctate and striolate,
with abundant closely appressed pubescence, which gives the insect
a velvety appearance. Body everywhere, except on funiculus, with
numerous erect or suberect hairs.

Color dark ferruginous, legs lighter. Pile and pubescence gray.

Camp 39 Madeira-Mamore R. R.

This subspecies differs from typical D. charfifcx in its more robust
form, proportionately larger head and the opacj^ue or semiopaque
structure of the integument. In a series of D. chariifcx from Bon
Lugar on the Rio Purus (Coll. Goeldi, det. Forel) before me, the whole
body is much more shining than in D. viamoreensis, and the pubes-
cence much more sparse. The head is proportionally smaller, and
the occipital margin is distinctly, though narrowly, margined.

221. Bendromyrmex madeirensis, sp. nov.

Worker. Close to D. cqnccdis.

Head with sides in front of eyes evenly convex, considerably nar-
rower at base of clypeus than in front of eyes; sides of occiput convex;
occipital border narrowly, but distinctly margined; the sides, seen
from above, angulate. Eye situated distinctly behind middle of
head. Frontal area broad, convex, faintly longitudinally impressed
at middle. Clypeus strongly carinate; the anterior border slightly
concave; mandibles wuth six teeth. Antennae long and slender.
Pronotum longer than broad, sides submarginate; strongly impressed
and margined at anterior border. Surface, except for anterior de-
pression, slightly convex. IVIesonotum rounded above and at sides.
Epinotum seen in profile, convex from apex to base; sides weakly
margined, surface flat. Apex of node, seen in profile acutely angulate,
from behind truncate, the sides near apical margin nearl}' straight,
forming an angle with the apical margin. Gaster ovate. Legs long
and slender.

Head shining, finely striolate and punctate, with rather sparse
pubescence and abundant semierect pile. Scape of antennae shining,
coarsely punctate, without pubescence, but with much long erect
pile; funiculus thickly pubescent, wuth short pile. Pronotum shining,
sparsely punctate, with sparse, long, recumbent and a few erect hairs.
Mesonotum more densely punctate and more thickly pubescent than
the pronotum. Epinotum and petiole semilucid, finely striate trans-
versely, the former with very sparse pubescence and a few long, erect

MANN: THE ANTS OF BRAZIL. 487

hairs, the latter without pubescence, but with a few erect hairs. The
first segment of gaster shining, minutely striolate, with very sparse
appressed hairs and a few" long ones. Rest orgaster more pubescent,
and pilose, striolate, semishining. Legs shining, femora and tibiae
pilose, tarsi with semiappressed pubescence.

Color black, except legs, which are dark "reddish brown. Pile and
pubescence gray.

Described from one worker taken at Abuna, Rio Madeira.

222. Dendroinyrmex apicalis, sp. no v.
Plate 6, fig. 43.

Worker. Length 8.5 mm.

Head one and one fourth times as broad as thorax, with sides in
front of eyes very slightly convex; behind eyes narrowed to near the
base, then reflexed outward, forming a short, but distinct neck.
Frontal laminae short and protruding. Clypeus longitudinally
carinate. Mandibles thick, the blades with six acute teeth. An-
tennae long and slender, the scape extending a little past the pro-
mesothoracic suture. Eyes hemispherical. Frontal area but little
longer than broad; with a strong longitudinal impression. Eyes
small, hemispherical, located only slightly behind and beneath middle
of sides of head. Pronotum noticeably longer than broad; sides
round, in front semimarginate; surface slightly convex posteriorly,
flattened anteriorly, but not disciform. Mesothorax as broad as
long, sides evenly rounded, seen in profile slightly convex, with faint
depression at middle. Epinotum a little over three times as long as
broad, sides slightly convex, evenly margined; in profile evenly
convex from apex to base, surface flattened. The perpendicular
anterior surface, and the slanting anterior surface of the petiolar node
form an acute angle at the apex. Postpetiole with slight depression
at middle; apex, seen from behind rounded. Gaster suboval. Legs
long and slender.

Body, except gaster and legs, sublucid throughout, gaster shining.
Head and thorax finely, transversely striolate, the striae more coarse
on the epinotum and petiolar node, but extremely delicate on the
gaster. Body finely punctate, each puncture with a short, recumbent
hair.

Scape of antennae with a thin pubescence; funiculus with thicker

488 bulletin: museum of comparative zoology.

pubescence. The head, coxae, and gaster with a few long hairs, other
parts of body without them.

Color black, apical four joints of antennae brown, tarsi brown.

Described from one worker taken at Madeira-Mamore R. R. Camp
39.

The elongate pronotum which is only faintly depressed anteriorly,
the shining black color throughout, the brown tarsi and tips of the
antennae and the nearly entire absence of pilosity, distinguish D.
apicalis from the other species of the genus. It approaches most
closely D. madeirensis from the same region, but differs from this in
the shape of the head, the rounded sides of pronotum, the structure
of the petiolar node, and in the extremely sparse pubescence and
pilosity.

223. Dcndromyrmex hranncri, sp. nov.
Plate 6, fig. 47.

Worker. Length 8 mm.

Head about one and one half times as long as broad, sides in front
of eyes slightly convex, slightly broader at base of clypeus than in
front of eyes, anterior corners angulate; occiput strongly contracted,
seen from above with almost straight margin to a point a little over
half the distance from eye to apex, then contracted into a narrow neck
which is longer than broad, and has the posterior edge strongly re-
flex^d. Clypeus slightly broader than long; sides straight, with a
strong carina for entire length; anterior border truncate. Frontal
area suboval in shape, wuth a longitudinal carina. Fi'ontal laminae
moderately elevated, approximating anteriorly. Mandibles rather
slender, blade with five teeth. Antennae long and slender, scape
extending a little beyond the mesoepinotal impression, joints of
fiagellum subequal in length, all longer than broad, the anterior ones
somewhat the thickest. Ej^e small, very convex, situated at posterior
fourth of head. Thorax long and slender, the width contained four
times in the length. Prothorax broadest at anterior tliird, constricted
in front, sides evenly rounded; in profile very slightly convex. Pro-
mesonotal impression not deep. Mesonotum evenly rounded to near
apex, where there is a strong transverse constriction; the sides from
above strongly concave. Epinotum slightly over twice as long as
broad, divided at anterior third by a broad transverse impression,
which gives the profile the shape of a saddle; the declivity short,

MANN: THE ANTS OF BRAZIL. 489

with flattened surface, reflexed at base. Petiolar node thick; rounded
above and at sides, a httle broader than long; in profile rounded in
front, nearly straight behind ; the postero ventral surface of the petiole
with a rounded projection. Ventral surface of mesothorax with an
angulate tooth at anterior third. Gaster elongate oval, about two
thirds as long as the thorax. Legs very long and slender.

Head sublucid, finely, densely, and evenly punctate. Thorax
shining, minutely punctate. Epinotum finely striate, sublucid.
Node sublucid, more coarsely striate transversely. Gaster shining,
finely striolate.

Head without pubescence but with sparse, erect pile, antennae more
coarsely pilose, funiculus pubescent, and with erect pile. Thorax,
petiole, and node without pubescence, and but sparsely pilose. Gaster
with scattered pubescence and sparse erect pile. Legs pilose, tibiae
with a little pubescence.

Color dark ferruginous, the thorax lighter in color than the head
or abdomen. Pile and pubescence white.

Described from several workers taken at Abuna. This is a very
aberrant form, resembling at first sight one of the elongate American
species of Dolichoderus. The long neck and the divided epinotum
widely separate this species from those others which have been in-
cluded in the genus Dendromyrmex.

Camponotus {Dinomyrme.r) agra Smith has the occiput drawn out
into a neck, and the head of the worker minor resembles that of
Dendromyrmex branneri, but the thoracic structure is entirely different.

The following key will serve for the identification of the species of
Dendromyrmex.

Form very long and slender ; occiput drawn out into a narrow neck
which is longer than broad; epinotum divided by broad trans-
verse depressions into two portions (saddle-shaped in profile).

branneri Mann

Form more robust; occiput not long; epinotum not divided. ... 1

1 . Thorax brown, or yellowish brown 2

Thorax black, or at least dark brown 4

2. Gaster shining chartifex Smith

Gaster opaque 3

3. Head smaller; occiput faintly margined; color yellow. (British

Guiana) chartifex var. felis Mann

Head larger; occiput not margined; color reddish. (Brazil)

chartifex var. mamoreensis Mann

490 bulletin: museum of comparative zoology.

4. More slender species; body black; thorax shining 5

More robust species partly brown ; thorax subopaque 6

5. With very sparse pile; tips of antennae brown; neck shorter;

anterior border of clypeus convex apicalis Mann

Abundantly pilose; tips of antennae black; neck longer.

madeirensis Mann

6. Gaster shining 7

Gaster opaque 8

7. Gaster black traili Mayr

Gaster red traili var. rufogaster Mann

8. Gaster brown, with distinct punctation nidulans (Smith)

Gaster black; punctation very fine fabricii Roger

PLATE 1.

Mann. — The Ants of Brazil.

PLATE 1.

Fig. 1. Eciton (Acamatus) legionis F. Smith subsp. cremilatum Mann.
Worker.

2. Platythyrea meinerti Forel. Worker. Dorsal view.

3. Platythyrea meinerti Forel. Lateral view.

4. Rhopalopone relicta Mann. Worker.

5. Head of same from front.

6. Ectatomma (Gnamptogenys) tortuolosum F. Smith. Worker.

7. Ectatomma (Gnamptogenys) concinnum (F. Smith). Worker.

8. Ectatomma (Ectatomma) confine Mayr. Worker.

9. Neoponera (Neoponera) bakeri Mann. Worker.

10. Neoponera (Neoponera) carinulata (Roger). Worker.

IL Anochetus (Anochetus) bispinosus (F. Smith). Worker.

BULL. MUS. COMP. ZOOL.

Ants of Brazil. Plate 1

PLATE 2.

Mann. — The Ants of Brazil.

PLATE 2.

Fig. 12. Belonopelta jeckylli Mann. Worker. Dorsal view.

13. Belonopelta jeckylli Mann. Worker. Lateral view.

14. Neoponera (Neoponera) cavinodis Mann. Worker.

15. Azteca schumamii Emery subsp. dubia Mann. ^ .

16. Azteca miilleri Emery subsp. terminalis Mann. ^ .

17. Dolichoderus (Monacis) tristis Mann. ^ .

18. Dolichoderus (Dolichoderus) imbecillus Mann, y .

BULL. MUS. COMP. ZOOL.

Ants of Brazil. Plate 2

Mann. — The Ants of Brazil.

PLATE 3.

Fig. 19. Dolichoderus (Hypoclinea) ghiliani Emery. § .

20. Dolichoderus (Hypoclinea) lutosus F. Smith. ^ .

21. Pseudomyrma arboris-sanctae Emery. Q .

22. Pheidole (Pheidole) carapuna Mann. Soldier.

23. Pheidole (Pheidole) wheeleri Mann. Soldier.

24. Pheidole (Pheidole) biconstricta Mayr subsp. burtoni Mann.

Soldier.

25. Pheidole (Pheidole) colobopsis Mann. 9 .

26. Head of same from front.

27. Crematogaster (Crematogaster) heathi Mann. 8 .

28. Megalomyrme.x wallacei Mann. ^ .

BULL. MUS. COMP. ZOOL.

Ants of Brazil. Plate 3

PLATE 4.

Mann. — The Ants of Brazil.

PLATE 4.

Fig. 29. Monomorium (Mitara) subterraneum Mann. ^ .

30. Head of same from front.

31. Solenopsis geminata (Fabr.) subsp. medusa Mann. Soldier.

32. Myrmicocrypta foreli Mann. ^ , head from front.

33. MyrmicocrjTDta foreli Mann. 9 .

34. Myrmicocrypta foreli Maun. ^ .

35. Crj-ptocerus (Cryptocerus) complanatus Guerin. Soldier.

36. Leptothorax (Goniothorax) echinatinodis Forel subsp. spininodis

Mayr. § .

BULL. MUS. COMP. ZOOL.

Ants of Brazil. Plate 4

PLATE 5.

Mann. — The Ants of Brazil.

PLATE 5.

Fig. 37. Apterostigma branneri Mann, cf .

38. Acanthognathus ocellatus Mayr. 9 .

39. Camponotus (Myrmothrix) rapax (Fabr.). ^ major,

40. Camponotus (Myrmeurynota) heathi Mann. S minor.

41. Camponotus (Myrmeurynota) heathi Mann. ^ major.

42. Camponotus (Myrmomahs) depressus Mayr. S •

BULL. MUS. COMP. ZOOL.

Ants of Brazil. Plate 5

PLATE 6.

Mann. — The Ants of Brazil.

PLATE 6.

Fig. 43. Dendromyrmex apicalis Mann. Worker.

44. Camponotus (Myrmoturba) maculatus Fabr. var. abunanus Mann.

Worker.

45. Camponotus (Myrmamblys) burtoni Mann. Worker major.

46. Camponotus (Myrmamblys) burtoni Mann. Head from front.

47. Dendromyrmex branneri Mann. Worker.

48. Camponotus (Myrmobrachys) adpressisetosus Forel. Worker

major.

49. Camponotus (Myrmothrix) wheeleri Mann. Worker minor.

50. Camponotus (Myrmothrix) wheeleri Mann. Worker major.

51. Camponotus (Myrmothrix) wheeleri Mann. Worker minor.

52. Camponotus (Myrmoturba) maculatus Fabr. subsp. fryi Mann.

Worker major.

Note. In the figures of Camponotus (Myrmothrix) wheeleri the points of
insertion of the antennae are shown too far from the clypeus.

BULL. MUS. COMP. ZOOL.

Ants of Brazil. Plate 6

PLATE 7.

Mann. — The Ants of Brazil.

PLATE 7.

Fig. 53. Plat3rthyrea meinerti Forel. Full-grown larva.

54. Ectatomma (Ectatomma) quadridens (Fabr.). Larva.

55. Dinoponera grandis Guerin subsp. mutica Emery. Immature (?)

larva.

56. Pseudomyrma rufa F. Smith. Female (dealated).

BULL. MUS. COMP. 200L.

Ants of Brazil. Plate 7

V

J

-y

55

v^sMr^?::

±

\J VJ

OCT 24 1916

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.
Vol. LX. No. 12.

THE FOSSIL ELATERIDAE OF FLORISSANT.

By H. F. Wickham.

With Seven Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.
October, 1916.

No. 12. — The Fossil Elateridae of Florissant.
By H. F. Wickham.

Elaterid beetles are fairly common as fossils. Some of the earli-
est Coleoptera known, occurring in the Triassic rocks, had the form
of an Elater more or less sketched out but, according to Handlirsch,
who has seen the specimens, none of them displayed characters which
would allow them to be placed in the modern family with any cer-
tainty. Again, in the Liassic beds, the elateriform Coleoptera ap-
peared, this time in rather greater abundance, but even yet they seem
to present no e\'idence of belonging to the family in a proper sense.
The lithographic chalk of Jurassic times has furnished insects which
have even been referred to the recent genus Elater but here, as before,
Handlirsch believes that the closeness of relationship has been over-
estimated, though he states that his Jurassic genus Malmelater
belongs at any rate to the Elateridae. This seems to be the earliest
well-supported record of the appearance of the family in geologic time.

Following the Jurassic, we have a period of immense duration in
which no large deposits of Coleoptera were made or, if they exist,
none have been discovered. No more Elateridae are recorded until
after the opening of the Tertiary, when they begin to be at least
moderately numerous. Menge is said to have had 130 specimens
from the Amber fauna. In the later deposits of Oeningen and other
European Miocene localities they seem to be quite abundant, Heer
having described many, some in fine preservation. By this time they
had become so much like our modern forms that generic identity
frequently seems quite well established though one cannot feel sure
that some important character may not have been carried away with
a missing member. Tarsal lobes and claw-teeth scarcely ever remain
intact, the mesosternum is often too distorted to study and in many
instances it is impossible to make out the limits of the metacoxal
plates which play so large a part in the classifications of systematists.

In the way of giving at a glance the published standing of Elateri-
dae in Tertiary strata, the following outline, compiled mostly from
Handlirsch and with his assignment of the age of each deposit, may
be useful. The records are given by localities in preference to arrange-
ment by generic sequence.

494

bulletin: aiuseum of comparative zoology.

Paludina Beds, Eiigland.

Elater sp.

Lower Eocene.

Baltic Amber.
Eucnemidae, (many),
Eucnemis sp.
Microrhagus sp.
Elateridae, (many).
Cardiophorus, (two ? sp.).
Cryptohypnus, (two sp.).

Lower Oligocene.
Elater, including Ampedus, (sevem

sp.).
Elater naumanni Giebel.
Agriotes sp.
Limonius, (two sp.).
Athous sp.

Aix. Lower Oligocene.

Elater, (two sp.).

Siebengebirge. Upper Oligocene.

Silicernius spectabilis Heyd.

Greith, Switzerland. Upper Oligocene.

Elaterites amissus Heer.

Spitzbergen.
Plater holmgreni Heer.

Kutschlin, Bohemia.
Campsosternus atavus Deichm.

Oeningen.
Adelocera granulata Heer.
Laeon primordialis Heer.
Alaus spectabilis Heer.
Cardiophorus brauni Heer. /
Cardiophorus sp. nov.
Elater, (five sp.).
Ampedus seyfriedi Heer.

Lower Miocene.
Elater ehrenwardi Heer.

Lower Miocene.
Elaterites dicrepidioides Deichm.

Upper Miocene.
Ischnodes gracilis Heer.
Limonius optabilis Heer.
Corymbites sutor Heer.
Elaterites lavateri Heer.
Elaterites obsoletus Heer.
Elaterites, (five sp.).

Myszyn, Galicia. Upper Miocene.

Elater wisniowskii Lomn.

From the above list, it will be seen that only eighteen species have
been specifically characterized from the European Tertiaries, scarcely
enough to make a comparison with the Florissant fauna of any value.
It should be noted, however, that several of the principal genera are
taken to be identical in the two areas. Too much confidence must
not be given the determinations in any case. Outside of the Floris-

WICKHAM: fossil ELATERIDAE of FLORISSANT. 495

sant district, the only North American Tertiary Elateridae thus far
made known are these: —

Green River, Wyo. Oligocene.

Corymbites velatus Scudd.

White River, Colorado-Utah. Oligocene.

Epiphanis deletus Scudd.

Fossil, Wyo. Oligocene.

Adocetus buprestoides Scudd.

Similkameen River, B. C. Miocene.

Limonius impunctus Scudd. Elaterites sp.

Nicola River, B. C. Miocene.

Cryptohypnus (?) terrestris Scudd.

These five records (since that of Elaterites cannot be considered as
having any special value) are even less illuminating than those of
Europe. Three of the genera are now recognized from Florissant.

Years ago, Scudder announced that he had about forty species of
Florissant Elateridae but he never gave them detailed study and it is
probable that the number was somewhat overestimated. At any
rate, when I looked through his collections in 1912 I was unable to
distinguish so many and of those in his cabinet a good proportion was
too poor for identification. Later explorations have brought in about
as many specimens as were known to Scudder and by a study of the
material belonging to the Museum of Comparative Zoology, the
United States National Museum, the Princeton University Geological
Museum, the Museum of the University of Colorado and the Peabody
Museum of Yale University, supplemented by a collection of my own,
I have separated forty-three species with some degree of certainty
as to their generic and specific affiliations. For convenience in making
comparisons, this list is appended. Excepting four which I have
described in earlier papers, all are new and are characterized in the
body of this article.

Eucneminae.

Eucnemis antiquatus. Microrhagus miocenicus.

Deltometopus fossilis. Microrhagus vulcanicus.

Fornax relictus.

496

bulletin: museum of comparative zoology.

Elaterinae.

Lacon exhumatus.
Cardiophorus lithographus.
Cardiophorus florissantensis.
Cardiophorus cockerelli.
Cardiophorus requiescens.
Cardiophorus (?) deprivatus.
Horistonotus coloradensis.
CryptohjV'pnus exterminatus.
Cryptohypnus hesperus.
Anchastus eruptus.
Anchastus diluviaHs.
Monocrepidius dubiosus.
Elater rohweri.
Elater scudderi.
Elater florissantensis.
Megapenthes primaevus.
Crj^ptagriotes minuseulus.
Agriotes comminutus.
Agriotes nearcticus.

Limonius aboriginalis.
Limonius florissantensis.
Limonius praecursor.
Limonius shoshonis.
Limonius volans.
Athous lethalis.
Athous contusus.
Athous fractus.
Paranomus exanimatus.
Paranomus heeri.
Paranomus laevissimus.
Ludiophanes haydeni.
Corymbites granuUcoUis.
Corymbites primitivus.
Corymbites submersus.
Corymbites restructus.
Corymbites prophetieus.
Oxygonus primus.
Melanaetes cockereUi.

Assuming the above species to be correctly referred to their re-
spective genera, analysis shows that five belong to the Eucneminae,
the remaining thirty-eight to the Elaterinae. Of those in the second
categor}^, one belongs to the Agrypnini, the other thirty-seven to the
Elaterini which holds today the great bulk of North American species
of the subfamily. According to the classification adopted by LeConte,
the Elaterini separates on the basis of the structure of the metacoxal
plates into two subtribes, the Elaterini (genuini) and the Corym-
bitini, dividing the North American species between them almost
exactly in the ratio of three to four. Of the fossils, fifteen are referred
to the first subtribe, twenty-two to the second, giving a ratio rather
startlingly similar. Of course the number of species involved is small
enough to allow a considerable percentage of error to creep in, should
the identifications turn out to be wrong in any case, but the conclu-
sions must be held to have some weight. On the face of the matter,
the figures would indicate that the relative percentages of Elaterini
and Corymbitini were almost the same at Florissant during the Mio-
cene as they are in North America in general today.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 497

Making some comparisons with the recent Elateridae of Colorado,
we find recorded in the catalogue of the beetles of that state about
seventy-three species, three of which are Eucneminae, leaving seventy
in the Elaterinae. Of these, three are Agr^-pnini, three Chalcole-
pidiini (a tribe not represented among oiu- fossils and containing large
species rather tropical than otherwise in their general range, though the
Colorado representatives are of the genus Alaus which runs well to the
north) while the rest, sixty-four in number, are Elaterini. These
Elaterini are divided into twenty-nine which belong to the subtribe
Elaterini proper and thirty-five to the Corymbitini, a moderate
divergence from the ratio shown at Florissant in the Miocene. If
we may depend upon these figures, the evidence indicates a rather
remarkable similarity in conditions then and now.

Two or three items of generic comparison deserve notice. In
looking over the fossils I was surprised to find such an apparent rich-
ness in Cardiophorus; but turning to the Colorado catalogue it will
be noted that Cardiophorus now has no less than eight representatives
in the state, against the five known as fossils. Here again, the two
ratios are remarkable for their similarity. Of the genus Corymbites,
we find sixteen recent Colorado species against five fossils — indi-
cating that this genus, relatively to the other Corymbitini, was only
half as numerously represented then as at present. Of the twenty-one
genera included in the entire list of fossils, nine are not now known
from Colorado ; two of these are erected as new, four others are fairly
distinctively northern and none of the three remaining can be con-
sidered southern types. In fact, there is nothing in the fossil Elateri-
dae to indicate tropical or subtropical conditions or origin.

Something should be said regarding the facies of this collection of
Elateridae. On looking through the list, one will be struck at once
by the fact that it is made up, in the main, of species belonging to
large and well-known genera, mostly those of wide distribution.
Even if we allow that the preservation of fossil beetles is practically
never good enough to permit absolute certainty in generic identifica-
tion, it remains true that these Florissant Elateridae do not, even in a
single instance, exhibit anything conspicuous or remarkable in size or
form. This family, today, is by no means without peculiar and
highly modified members, some of them reaching great size, others
displaying oddities in outline or in the development of various por-
tions of the body, as will be seen in glancing over the plates in the
extensive monograph of Candeze. We are forced here, to the same
conclusion as in so manv of the other families — that the Florissant

498 bulletin: museum of comparative zoology.

fauna, outside of the rhynchophorous series, comprised a rather
monotonous and little specialized lot of beetles.

This paper, with another now in press elsewhere, will bring the
number of published Coleoptera from these shales up to about 566
species, and may be considered as very nearly ending the task of
working up the available material. What remains consists of isolated
species in various families so scattered through the whole order as not
essentially to disturb the conclusions already reached in regard to the
number and nature of the representation of each group. The rich-
ness of this fauna remains absolutely unapproached by that of any
other known deposit, unless the many unworked collections of Amber
insects may yield a similar wealth.

Citation of catalogue numbers follows the plan of Scudder, in
joining by "and" those referring to the two halves of a single speci-
men with its counterpart. The drawings are made with the camera
lucida and will show the outlines, though not the sculpture, the latter
being carefully described in the specific diagnoses.

EUCNEMINAE.

EucNEMis ANTiQUATUS Wickham.

Described and figured in Bull. M. C. Z., 1914, 58, p. 437, pi. 2,
fig. 9. No other specimens have been met with.

Deltometopus fossilis, sp. nov.

Plate 1, fig. 1, 2.

Form fairly stout. Head rounded in front, surface finely, not very
deeply and moderately closely punctate, bearing a scant covering of
dark hairs. Antennae pl'actically complete on one side, in life evi-
dently reaching the prothoracic hind angles, first joint large, second
short, third not in good condition but apparently rather long, fourth
and following subequal, weakly serrate. Prothorax only a little
broader than long, (as preserved), the sides nearly straight and,
judging from the margins, about perfect, only a little sinuate in front
of the base, hind angles (only one of which remains), scarcely diver-

WICKHAM: fossil ELATERIDAE of FLORISSANT. 499

gent and but shortly if at all carinate, entire thoracic surface hairy
like the head, minutely and sparsely punctured. Elytra not tapering
until well behind the middle, striae very fine, their punctures sepa-
rated in general by considerably more than their own long diameters,
interspaces broad, flat, finely, sparsely punctured, each puncture
bearing a hair. Length, from front of head to elytral apex, 5.90 mm.;
of elytron, 3.75 mm.

Described from one specimen.

Type. — In the Museum of the University of Colorado. It was col-
lected at Station 14, Florissant, Colo., by S. A. Rohwer.

The generic reference is made with a good deal of doubt, but the
form, sculpture, and vestiture point, in general, to the Eucneminae
and the antennae and size are not unlike Deltometopus.

Fornax relictus, sp. nov.
Plate 1, fig. 3.

Outline rather fusiform. Head of moderate size. Antennae not pre-
served except the basal portion of one which is in too poor condition to
be described. Prothorax beneath with wide marginal groove, proster-
num not very well shown in front, but the lobe was evidently short,
sutures grooved and quite broad, probably nearly straight although
one of them is thrown out of line by pressure, spine not long, pointed,
its margin with a fine but distinct bead. The hind angle, shown on
one side only, is well developed, not strongly divergent, front angles
not completely preserved, sides, as far as shown, evidently convergent
anteriorly and slightly arcuate. Punctuation of the entire under-
side obscure, apparently minute, with marks of a covering of fine hairs.
Elytra with striato-punctate sculpture showing through. Length, from
front of head to abdominal apex, exclusive of sex organ, 6.85 mm.

Described from one specimen.

Type. — In the collection of H. F. Wickham. \Yilson Ranch,
Florissant, Colo.

In form and size, this beetle is not unlike the fossil Microrhagus
vulcanicus, described herein, but has stronger elytral sculpture.
The wide marginal prothoracic grooves and the type of the prosternal
sutures are much like those of the recent Fornax hornii of our eastern
states, which is said to be the female of F. calccatus. In general, the
form and size are also similar to that species.

500 bulletin: museum of comparative zoology.

MiCRORHAGUS VULCANICUS, Sp. nOV.

Plate 1, fig. 5.

Form moderately stout. Head finely but closely and rather deeply
punctured. Antennae not well preserved, about eight joints remaining
which are scarcely serrate and indicate that if entire the antennal apex
would pass well beyond the prothoracic hind angles. Prothorax
finely, sparsely punctate and strongly hairy, apex much narrowed,
sides rather pronouncedly arcuate, hind angles divergent. Scutel-
lum injured so that the exact shape is not definable. Elytra a httle
arcuate at sides and conjointly rounded at apex, hairy, slightly striate
near the base, the remainder of the surface finely punctulate. Length,
from front of head to elytral apex, 7.40 mm.; of elytron, 4.85 mm.

Described from one specimen.

Type — No. 2,775 M. C. Z. Florissant, Colo. (No. 13,034 S. H.
Scudder Coll.).

This insect has a type of sculpture and vestiture common in the
Eucneminae, and if assigned to that subfamily would go in Micro-
rhagus by the form of the coxal plates and the apparent structure of
the basal antennal joints. Compared with the recent M. triangularis,
the present species has finer sculpture throughout and is of larger size.

MiCRORHAGUS MIOCENICUS, Sp. nOV.

Plate 1, fig. 4.

Form fairly stout. Head quite large, strongly transverse, anterior
margin arcuate, surface obscurely but closely, and rather coarsely
punctured on the front, less strongly on the vertex, which becomes
nearly smooth posteriorly. Antennae poorly preserved, not strongly
serrate, reaching to or behind the prothoracic hind angles. Pro-
thorax about one fourth broader than long, apex narrower than the
base, front angles not well marked, sides regularly and moderately
arcuate, base nearly truncate, hind angles, (only one of which is
preserved), acute, slightly divergent and distinctly carinate. The
surface is finely, very obscurely and not closely punctate, with a thin
clothing of moderately long dark hairs. Elytra broad at humeri,
sinuately tapering behind them, hardly striate and with faint rows of
punctures, the vestiture like that of the prothoracic disk. Underside

WICKHAM: fossil ELATERIDAE of FLORISSANT. 501

not shown. Length, from front of head to elytral apex, 5.60 mm.;
of elytron, 3.60 mm.

Described from one specimen.

Type. — In the Museum of the University of Colorado. It was col-
lected by Mrs. W. P. Cockerell at Station 14, Florissant, Colo.

In general, this beetle is a good deal like the preceding, but is smaller
and has very diiferently shaped elytra.

ELATERINAE.

Lacon exhumatus, sp. nov.
Plate 1, fig. 6, 7.

Form stout. As the specimen shows the underside only, no details
of the sculpture of the upper surface can be given. Antennae with
only the middle portion well preserved, joints stout, moderately
serrate. Prothorax beneath punctured on the flanks but not very
closely nor strongly, prosternum somewhat smoother, lobe short,
blunt, sutures curved, excavate, more deeply in front. Elytra, as
shown from below, merely indicating that they were marked with
rows of coarse punctures. Abdominal punctuation obscure. Length,
from front of head to elytral apex, 7.75 mm.; of elytron, about 5.00
mm., the base being too obscure to locate exactly.

Described from one specimen.

Type.— ^o. 2,776 M. C. Z. Florissant, Colo. (No. 4,456 S. H.
Scudder Coll.).

This insect has the look of Lacon and agrees in the structure of the
underside of the prothorax, the short antennae and the elytral sculp-
tm'e. It is smaller than the average recent L. rectangular is, wide-
spread in North America, but is just the same size as some southern
specimens in my collection.

Cardiophorus lithographus, sp. nov.

Plate 2, fig. 1-3.

Form moderately stout. Head not well preserved, very minutely
punctulate. Prothorax nearly equal in length and breadth, apex
narrower than base, sides rather faintly arcuate, hind angles acute

502 bulletin: museum of comparative zoology.

and only a little divergent. Surface polished, with sparse, fine
punctuation and signs of delicate pubescence. Scutelhun cordiform,
impunctate. Elytra finel}^ regularly striate, the strial punctures
deep but not coarse, rounded or very slightly oblong, ordinarily
separated in each stria by about their own diameters or sometimes a
little less. Interstitial spaces flat, extremely minutely punctulate
and with fine pubescence. Underside almost perfectly smooth.
Length, 8.25 mm.; of elytron, 5.25 mm.

Described from one specimen, with counterpart.

Type. — In the collection of H. F. Wickliam. Wilson Ranch,
Florissant, Colo. With it are associated four others in my collection;
two in the collection of the U. S. National Museum; one in the Mu-
seum of the University of Colorado, found by Professor Cockerell's
party at Station 14; and No. 2,777-2,784 M..C. Z. (No. 4,746, 6,007,
7,650, 8,731, 11,174, 11,782, 12,423, 14,329, S. H. Scudder Coll.).

The generic reference is based on the form of the metacoxal plates,
the truncate prosternal spine, the cordiform scutellum and the carini-
form mark on the underside of the prothoracic flanks. The sculpture
is entirely that of Cardiophorus and the present species is not unlike
the common recent North American C. convexus in most of its char-
acters.

Cardiophorus florissantensis, sp. no v.

Plate 2, fig. 4, 5.

Form stout. Head minutely, closely punctulate and pubescent.
Antenna not complete, but in life evidently not quite reaching the
prothoracic basal angles, the articulations too obscure to allow of
comparisons of their lengths. Prothorax one fifth broader than long,
narrower apicall^', sides regularly arcuate to near the hind angles
which are acute and somewhat divergent, base sinuate each side,
rather prominent at middle, surface very minutely punctulate and
finely, not closely, pubescent. Scutellum cordiform. Elytra finely
striate, the striae with rounded or slightly elongate punctures which
are separated usually by less than their own diameters, interstitial
spaces broad, flat, not. visibly punctured, pubescent. Underside
nearly smooth. Length, from front of head to elytral apex, 10.70 mm. ;
of elytron, 6.60 mm.

Described from one specimen, with counterpart.

Type. — In the Museum of the University of Colorado, collected

WICKHAM: fossil ELATERIDAE of FLORISSANT. 503

by G. N. Rohwer at Station 13, Florissant, Colo.; with it are
associated a specimen, with counterpart, found by S. A. Rohwer and
one found by Mrs. W. P. Cockerell at the same place; five examples
in the collection of the U. S. National Museum; and No. 2,771-2,774
M. C. Z. (No. 84, 2,094, 12,421, 12,425 S. H. Scudder Coll.).

Larger than C. lithoc/rapJuis (p. 501), but otherwise similar. I do
not think there is any doubt of the specific distinctness of the two.
The coxal plates, as shown, look quite different but I am afraid to
depend entirely upon these as the edges may become broken and
change the apparent form.

Cardiophorus cockerelli, sp. nov.
Plate 2, fig. 6.

Form stout. Head minutely, obscurely punctured, antennae
wanting. Prothorax nearly one and one third times as broad as long,
not much narrowed anteriorly, sides strongly rounded, hind angles
short but prominent, carinate, basal sinuations pronounced, surface
rather densely and more sti'ongly and coarsely punctured than the
head. Scutellum apparently imperfect, pointed behind. Elytra
short, broad, somewhat obtusely conjointly rounded apically, surface
striate, the striae moderately deep, their punctures strong, rounded
or a little elongate, close-set, separated ordinarily by less than their
own diameters, interspaces broad, flat, roughened somewhat but not
distinctly punctured. The entire upper surface of the prothorax and
elytra shows signs of rather fuie pubescence. Length, from front of
head to elytral apex, 6.20 mm. ; of elytron, 3.85 mm.

Described from one specimen.

Type — No. 2,765 M. C. Z. Florissant, Colo. (No. 1,916 S. H.
Scudder Coll.). With it are associated No. 2,766-2,767 M. C. Z. (No.
6,379, 10,639 S. H. Scudder Coll.); and three specimens, two with
counterparts, in the Museum of the Lliiversity of Colorado, all from
Station 14, bearing the numbers 151, 178 and 179, 209 and 258.
No. 2,768-2,770 M. C. Z. (No. 7,476, 9,160, 12,041 S. H. Scudder
Coll.) probably belong here.

The underside is not shown in the type, but the specimen No. 2,767
M. C. Z. (No. 10,639 S. H. Scudder Coll.), displays it well, exhibiting
moderately curved prosternal sutures, truncate spine and nearly
smooth surface. This beetle is similar to the recent North American

504 bulletin: museum of comparative zoology.

C. cardisce, but seems to be a little more coarsely pimetured on the
prothorax. It is smaller than the Florissant species C. lithographvs
and C. florissantensis, and much less coarsely punctured than the
fossil Horistonotus coloradensis from the same shales.

Cardiophorus requiescens, sp. nov.
Plate 2, fig. 7, 8.

Form fairly elongate. Head minutely and closely punctulate.
Antennae wanting. Prothorax too badly damaged for description of
the form, punctuation sparse and fine. Scutellum cordiform. Elytral
striae extremely fine, the punctures shallow, somewhat elongate and, in
general, separated by several times their own diameters. Underside
nearly smooth, the visible punctures being sparse and small. Length,
from front of head to elytral apex, 6.65 mm.; of elytron, 4.15 mm.

Described from one specimen.

Type. — In the collection of H. F. Wickham. Wilson Ranch,
Florissant, Colo.

This seems to separate easily from the other Florissant fossil
Cardiophori by the finer sculpture and especially in the distant strial
punctures. The generic reference is plainly indicated by the form
of the prosternal spine and of the scutellum.

Cardiophorus (?) deprivatus, sp. nov.

Plate 2, fig. 9, 10.

Form stout. Head not visibly punctured. Antennae apparently
with the second joint hardly shortened, the third a trifle shorter than
the fourth, the following, up to and including the tenth, subequal and
only faintly serrate. Prothorax scarcely at all punctulate, about
one third wider than long, sides regularly and not strongly arcuate,
the angles not prominent. Scutellum a little elongate, pointed at
tip but not strictly cordiform. Elytra hardly three and one third
times as long as wide, rather strongly tapering, apices conjointly
rounding, surface without sculpture. Underside practically smooth
throughout, prosternal sutures curved, convex inwards, metacoxal
plates suddenly narrowed externally. Length, from front of head
to elytral apex, 5.15 mm.; of one elytron, 3.50 mm.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 505

Described from one specimen with its counterpart.

Type. — In the Museum of the University of Colorado. It was col-
lected at Station 13, Florissant, Colo., by Professor Cockerell's
expedition of 1906. The obverse bears his number 99, the reverse,
127.

Unfortunately the prosternal spine is damaged so that its form is
not shown and while the beetle is perhaps not a true Cardiophorus,
because of the form of the scutellum, it seems best to place it pro-
visionally in that genus on account of its general similarity to some
of the recent species with faint sculpture.

HORISTONOTUS COLORADENSIS, Sp. nov.

Plate 2, fig. 11, 12.

Form fairly stout. As the specimen shows from the underside, the
sculpture of the head is not visible. Prothorax, beneath, with the
flanks and prosternum distinctly but rather finely and only moder-
ately closely punctate, the punctures ordinarily separated by their
own diameters or a little less, not very regularly spaced. Meta-
sternum a little more finely punctured. Abdomen punctured in
general, like the prothorax, the proximal segments somewhat more
finely and sparsely than the distal, the terminal one with the punc-
tures crowded laterally. Scutellum not visible. Elytra displayed
in reverse, the punctuation showing through. It is arranged in
striae, the punctures coarse, rounded, deep, mostly separated by less
than their own diameters, those near the elytral apices nearly touching.
Length, 6.60 mm.; of elytron, 4.10 mm.

Described from one specimen.

Type. — In the collection of H. F. Wickliam. Florissant, Colo.
Possibly No. 2,763-2,764 M. C. Z. (No. 815, 6,384 S. H. Scudder
Coll.) may also belong here.

In this case, the generic reference is not made with much confidence.
However, the size, the truncate prosternal spine and the form of the
coxal plates point to the Cardiophori. The punctuation of the
underside is coarser than usual in Cardiophorus but is quite similar in
disposition to that of the recent Horistonotus simplex from the south-
western United States.

506 bulletin: museum of comparative zoology.

Cryptohypnus extermixatus, sp. nov.
Plate 1, fig. 8, 9.

Form short, broad, and stout. Surface hardly visibly punctate
anyAvhere, but this may possibly be due to the rather coarse texture
of the stone in which the specimen is preserved. Head rather large.
Antennae with the first joint long and thick, the second and third
scarcely shorter than those succeeding, none of which are much pro-
duced at the angles so that the organ is only weakly serrate. Pro-
thorax a little distorted, a trifle more than one third wider than long,
apex feebly emarginate, front angles pointed but obtuse, width great-
est in front of the middle, sides moderately strongly arcuate to a
point near the acute hind angles, slightly divergent and carinate, base
sinuate each side. Scutellum suborbicular. Elytra a little more
than twice the length of the prothoracic median line, pointed at apex.
Length, 4.55 mm.

Described from one specimen.

Tjjpe.— No. 2,762 M. C. Z. Florissant, Colo. (No. 11,280 S. H.
Scudder Coll.).

The form and antennal structure are those of Cryptohypnus. In
size, it approximates the recent C. nodurnus of Canada and our
northern states, but the sculpture, if properly shown on the stone, is
more like that of the much smaller C. pedoralis. This last species,
in its varieties, has a wide distribution in Xorth America today.

Cryptohypnus hesperus, sp. nov.

Plate 3, fig. 1.

Form fairly slender for this genus. Sculpture of head not definable.
Antennae not well shoTVTi, but one side is well enough preserved to
indicate that they reached at any rate to the hind angles of the pro-
thorax. Prothoracic width about one fourth greater than the length,
surface sculpture obscurely preserved, base not much broader than
the apex, sides rather strongly rounding, hind angles short, not diver-
gent, base deeply sinuate each side. Scutellum rounded or oblong.
Elytra tapering, arcuate at sides, apices conjointly rounded, sculp-
ture not well defined, showing only traces of faint striae. Length,
from front of head to abdominal apex, 4.30 mm.; of elytron, 2.60 ram.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 507

Described from one specimen.

Type.— No. 2,761 M. C. Z. Florissant, Colo. (No. 5,294 S. H.
Scudder Coll.).

This has the size and general appearance of Cryptohypnus and, as
far as shown, the sculpture seems to be like that of the recent C. pec-
toralis. In form, the present insect differs widely from the Florissant
fossil C. exterminatus.

Anchastus eruptus, sp. nov.
Plate 3, fig. 2, 3.

Form moderately stout. Sculpture of head not definable. Anten-
nae long, that on the right side, (as preserved), showing nine joints,
the terminal one of which, in life, would have reached well behind the
prothoracic basal angles, so it is likely that if complete the antennae
would have extended nearly to the elytral middle. The first joint is
large, the second small, third much larger than the second but a little
shorter than the fourth, fifth not in good condition, sLxth and fol-
lowing longer than the fourth, serrations, in general, well pronounced.
Prothorax with strongly curved prosternal sutures, lobe short, spine
stout and short, punctuation of flanks and sternum minute and incon-
spicuous. Elytra conjointly rounded apically, sculpture not showing
through. Abdominal sculpture very fine. Length, from front of
head to elytral apex, 4.80 mm. ; of elytron, about 3.00 mm.

Described from one specimen.

Type.— No. 2,760 M. C. Z. Florissant, Colo. (No. 11,281 S. H.
Scudder Coll.).

The coxal plates do not show up well, but seem to be very narrow
externally and broad internally, as in Anchastus. This genus has
similar antennal and prosternal characters, also.

Anchastus diluvialis, sp. nov.
Plate 3, fig. 4.

Form moderately elongate. Head not visibly punctured beneath.
Antennae poorly defined. Prothorax closely, distinctly, and regularly
but finely punctate on the flanks and sternum, the grooves strongly
double, curved. The prothoracic outline is obscured to some extent

508 bulletin: museum of comparative zoology.

by flattening, but as preserved the apex is very nearly as wide as the
base, the sides little curved, none of the angles plainly shown. Elytra
long, tapering, coarsely punctatostriate, the punctures separated by
about their own diameters, interstitial areas not wide nor visibly
punctured. Length, from front of head to elytral apex, 5.50 mm.;
of elytron, about 3.60 mm.

Described from one specimen.

Type.~^o. 2,759 M. C. Z. Florissant, Colo. (No. 11,277 S. H.
Scudder Coll.).

The generic reference is not certain, being based upon the sternal
grooves, the size, sculpture, and general form. This species easily
separates from the preceding by being much more coarsely sculptured.
Both are represented by undersides only.

MoNOCREPiDius DUBiosus, sp. nov.
Plate 4, fig. 1.

Form only moderately elongate. Head finely, sparsely, and in-
distinctly punctate. Prothorax about one fourth broader than long,
base and apex subequal, none of the angles very prominent, side
margin nearly regularly but not strongly arcuate, base broadly
emarginate in front of the scutellum, sinuate each side, surface not
well preserved but showing a few fine punctures. Scutellum sub-
quadrate. Elytra nearly three times the length of the prothoracic
median line, conjointly rounded at apex, surface finely and not deeply
striate, the striae with small, slightly elongate punctures, separated
in each row by approximately their own diameters, interstitial spaces
flat and not visibly punctured. Underside finely and feebly punctu-
late or nearly smooth. Length. 4.50 mm.; of elytron, 2.65 mm.

Described from one specimen, with counterpart.

Txjjpe. — In the collection of H. F. Wickham. Wilson Ranch,
Florissant, Colo.

This specimen is not especially well preserved nor does it offer any
striking characters. The underside shows the prosternal sutures to
be double, nearly straight, somewhat excavated anteriorly, the pro-
sternum rather narrow, the spine acuminate at tip. The hind coxal
plates are not well defined, but I think they are suddenly dilated
internally. Both legs and antennae are too poor for description.
No more suitable generic position can be suggested at present, though

WICKHAM: fossil ELATERIDAE of FLORISSANT. 509

all of the species of Monocrepidius that I know are more strongly
striate and somewhat differently proportioned.

Elater rohweri, sp. nov.
Plate 3, fig. 5, 6.

Form moderately elongate. Head rather short, distinctly but
finely punctured, somewhat sparsely on the sides and still more finely
and sparsely on the vertex, minutely hairy. Antennae quite slender,
scarcely at all serrate, not reaching the tips of the prothoracic hind
angles, basal joint large, second and third not well defined, the re-
mainder subequal, all finely hairy. Prothorax strongly narrowed
anteriorly, front margin arcuately emarginate, anterior angles not
very prominent, sides moderately arcuate, hind angles long, sharp,
a little divergent with distinct discal carina and possibly with an
external marginal one as well. The base is a little emarginate in
front of the scutellum. Thoracic disk with a well-impressed, smooth
median line on basal one fifth, the middle area finely and sparsely,
sides more strongly and closely punctate, entire surface hairy. Scutel-
lum oblong, punctured and hairy. Elytra bluntly pointed, tapering,
striae not deep, with rows of slightly elongate, rather fine punctures
separated in general by about their own long diameters, interstitial
spaces broad, flat, hairy, but not punctured excepting the small
depressions from which the hairs arise. Legs of moderate length,
finely hairy. Underside of body with most of the details not well
defined, but the prothoracic side-pieces are fairly strongly though not
densely nor coarsely punctured, the prosternum more finely. The
spine is pointed, the lobe broken at tip but apparently not long, the
sutures double, excavated, the metacoxal plates broad internally,
the abdomen finely hairy, scarcely punctulate. Length, from front
of head to elytral apex, 7.60 mm.; of elytron, 5.00 mm.

Described from one specimen, with counterpart.

Type. — In the Museum of the University of Colorado. Florissant,
Colo., collected at Station 14 by Professor Cockerell and bearing
his numbers 192 and 211.

The general features of this beetle point to Elater as a fairly exact
reference. It is most like some of the less strongly sculptured modern
North American species, such as E. sanguinipemiis or E. hehrensi,
but is rather smaller. Compared with the fossil E. scudderi, the

510 bulletin: museum of comparative zoology.

present species may easily be distinguished by the distinctly striato-
punetate elytra, while E. fiorissantensis, from these shales, is larger
and has a differently formed prothorax.

The specific name is given for Mr. S. A. Rohwer, a member of Profes-
sor Cockerell's expeditions and now of the U. S. National Museum
staff.

Elater florissantensis, sp. nov.

Plate 3, fig. 9.

Form moderately elongate. Head large, closely and rather strongly
but finely punctured. Antennae poorly defined. Prothorax about
one seventh broader than long, wider near the apex than at base,
finely, deeply, and closely punctured, sides regularly and somewhat
faintly arcuate anteriorly, nearly straight posteriorly, front angles
obtuse, hind angles long, a little divergent and apparently bicarinate.
Scutellum oval. Elytra rather strongly tapering, apices conjointly
a little rounded, surface clothed with rather long but not close hairs
of a darker color, striae fairly coarse, their punctures of moderate size,
usually a little elongate and separated by less than their own long
diameters, interstitial spaces flat, scarcely visibly punctulate. Length,
from front of head to elytral apex, 8.25 mm. ; of elytron, 5.40 mm. ;
of prothorax, along median line, 2.00 mm.

Described from one specimen.

Tijpe — No. 2,752 M. C. Z. Florissant, Colo. (No. 8,034 S. H.
Scudder Coll.). With it are associated No. 2,753-2,758 M. C. Z.
(No. 79 and 103, 6,869, 8,891, 9,200 and 11,732 S. H. Scudder Coll.).

The generic reference is only fairly exact. In thoracic outline the
fossil more nearly resembles our recent E. areolatus than any other
North American species known to me. The sculpture and vestitiu-e
of the elytra is something of the type seen in the living E. cordatus
from our Pacific coast. Unfortunately the outlines of the metacoxal
plates cannot be distinguished.

Elater scltdderi, sp. nov,

Plate 3, fig. 7, 8.

Form rather elongate. Head quite closely and coarsely punctured.
Antennae incomplete, but when entire probably not reaching the
prothoracic hind angles. Prothorax very little wider than long.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 511

sides, (judging by the better preserved one), gently and regularly
rounding, base and apex subequal or the latter a little narrower, front
angles obtuse, not at all prominent, hind angles moderately acute, a
little divergent and distinctly carinate. Thoracic punctuation rather
shallow and obscure, close and only fairly fine, surface strongly hairy.
Scutellum pointed at apex. Elytra moderately tapering to apex, not
striate and very obscurely punctate but hairy like the thorax. Length,
from front of head to elytral tip, 7.10 mm. ; of elytron, 4.75 mm.

Described from one specimen.

Type — No. 2,751 M. C. Z. Florissant, Colo. (No. 12,485 S. H.
Scudder Coll.).

Looks a good deal like E. florissantcnsis but is smaller and has non-
striate elytra. Probably it does not belong to Elater in the modern
sense.

Megapenthes primaevus, sp. nov.

Plate 4, fig. 2.

Form rather elongate. Head finely but extremely densely and
quite deeply punctured, somewhat less strongly upon the vertex
than upon the front. Antennae reaching well beyond the base of
the prothorax, heavier than in most of the fossil Elateridae but not
very strongly serrate, only the middle joints well defined. Prothorax
narrower at apex than at base, approximately equal in length and
breadth, surface finely and closely punctured though not quite so
strongly as the head, sides but little arcuate, front angles short,
hind ones not very long, a little divergent and distinctly carinate.
Scutellum oblong. Elytra conjointly rounded apically, finely striate,
strial punctures roimded, mostly separated by about their own diam-
eters or a little more, interspaces flat, wide, finely and not very closely
punctate, the punctures distinctly smaller than those of the striae.
Underside not shown. Length, 8.90 mm.; of elytron, 5.60 mm.;
of prothorax, 2.35 mm.

Described from one specimen.

Type.— No. 2,750 M. C. Z. Florissant, Colo. (No. 10,859 S. H.
Scudder Coll.). With it are associated two others, in the Museum of
the University of Colorado, one with counterpart (No. 226 and 249).
The only one with definite record is from Station 13B, collected
by S. A. Rohwer.

As seen under magnification, the general effect is much like that of

512 bulletin: museum of comparative zoology.

the recent M. aterrimus of the Pacific states. I am not sure of the
antennal structure, but it looks as if the second and third joints are
very short and what can dimly be seen of the hind coxal plate on one
side has the appearance of being broad internally and much narrowed
to the outer edge. Of course the generic reference is subject to
correction in the event of more material coming to hand.

Cryptagriotes, gen. nov.

Body form almost like that of Cryptohypnus. Coxal plates nearly
linear, scarcely narrowed externally, obtusely lobed over the thighs.
Prosternum short with a small, subtruncate lobe, sutures nearly
straight, apparently excavate anteriorly.

Type. — C. minusculus, sp. nov.

Cryptagriotes minusculus, sp. nov.
Plate 4, fig. 3.

Form moderately stout. Head large. Antennae not well pre-
served, slender, reaching beyond the pro thoracic hind angles. Pro-
thorax probably somewhat distorted, but, as preserved, wider in
front of the middle where the width is a little greater than the length.
Sides very gently arcuate, angles not well shown. Prosternal sutures
nearly straight, lobe rather short, entire under surface of the prothorax
finely punctulate and pubescent. Elytra two and one fourth times
the prothoracic length, conjointly rounded apically, the sides some-
what arcuate, sculpture not showing through. Abdomen very finely
punctulate and pubescent. Length, from front of head to elytral
^pex, 4.00 mm.; of elytron, about 2.30 mm.

Described from one specimen.

Tupe.— No. 2,749 M. C. Z. Florissant, Colo. (No. 8,653 S. H.
Scudder Coll.).

Judging from the form of the coxal plates, this little beetle should
go into the Corymbitini, possibly near x\griotes, but it does not agree
with any genus known to me.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 513

Agriotes nearcticus, sp. nov,
Plate 4, fig. 6.

Form elongate. Head minutely but very closely and distinctly
punctured. Antennae poorly preserved, but what remains of one
of them indicates that they were short and slender, scarcely sei^rate.
Prothorax broken along the side margins, obscuring the shape, punc-
tuation almost exactly like that of the head, the punctures finely
mamillate, nearly touching, even on the middle of the disk. Elytra
relatively rather elongate as compared with their width, striae appar-
ently very shallow but their rows of punctures are fairly deep and
strong, each puncture somewhat elongate, those in each series sepa-
rated by a little more or less than their own long diameters. Inter-
stitial spaces scarcely perceptibly punctulate, but with marks of a
fine, moderately short pubescence. Length, from front of head to
elytral apex, 8.00 mm. ; of elytron, 5.50 mm.

Described from one specimen.

Type.— No. 2,748 M. C. Z. Florissant, Colo. (No. 6,653 S. H.
Scudder Coll.). '

While resembling the Florissant fossils, Limonius fiorissantensis and
L. praccursor, this insect is more elongate and more delicately punc-
tured than the former and differs from the second in the closer punctu-
ation of the head and prothor^ as well as the almost complete lack of
it in the elytral interspaces. The coxal plates are not very clearly
shown, but I think they are correctly exhibited in the drawing. The
front seems to be higher than the labrum, as indicated by the distinct
line of demarcation in the fossil, and the slender antennae are like
those of Agriotes.

Agriotes comminutus, sp. nov.

Plate 4, fig. 4, 5.

Form rather elongate. Head coarsely, closely but not very deeply
punctured. Eyes and antennae not defined. Prothorax with the
margins badly broken, so that the exact shape is not discernible, but
it was evidently only a little wider than long, with a large discal
dark spot, similar to that of the recent North American A.fucosus, sur-
face sculpture obscure, the prosternal sutures, which show through,

514 bulletin: museum of comparative zoology.

nearly straight and apparently excavate anteriorly. Elytra not less
than two and one half times the prothoracic length, apices broken,
surface finely striate, the striae with distinctly elongate, well-im-
pressed punctures which are separated in each row by approximately
their own long diameters, interstitial spaces flat, broad, apparently
punctate. Abdominal dorsal sculpture, on the portion exposed by
the spreading of the elytra, obscure. Length, from front of head to
abdominal apex, 6.70 mm.; of portion of elytron as preserved, 3.60
mm.

Described from one specimen.

Type.— ^o. 2,747 M. C. Z. Florissant, Colo. (No. 11,800 S. H.
Scudder Coll.).

Since the general preservation of this beetle is poor, I should not
have ventured to describe it, had not the coxal plate been well shown
on one side. Taken into account with the form of the prosternal
sutures, the sculpture and the coloration, this seems to indicate a
probable affinity with Agriotes.

Limonius aboriginalis, sp. nov.
Plate 5, fig. 1-4.

Form stout for the genus. Head practically smooth. Antennae
just about reaching the hind prothoracic angles, moderately serrate,
second and third joints subequal, their united length about the same
as that of the fom-th which, however, is broader and begins the serra-
tion. Eyes normal. Prothorax about one fourth broader than long,
not visibly sculptured above, apex scarcely emarginate, narrower than
the base, sides arcuately broadening to about the middle thence
slightly sinuately narrowing posteriorly, hind angles well defined,
acute but not divergent. Scutellum oblong oval. Elytra with sides
less parallel than in most modern species of Limonius, apices con-
jointly rounded, surface rather finely but quite distinctly and regu-
larly striate, strial punctures becoming less distinct posteriorly,
rounded or slightly oblong, separated by about their o^\ti diameters.
Interstitial areas with signs of fine pubescence. Underside almost
smooth, only a few small, scattering punctures being visible. Legs not
displayed. Length, from front of head to elytral apex, 6.65 mm.

Described from one specimen, with counterpart.

Type, — In the collection of H. F. Wickham. Wilson Ranch,.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 515

Florissant, Colo. With it are associated No. 2,737-2,738 M. C. Z,,
(No. 7,971 and 10,952 S. H. Scudder Coll.). Most likely No. 2,739-
2,746 M. C. Z. (No. 2,870, 8,345, 8,549, 8,753, 12,766, 8,226, 8,842
and 11,788 S. H. Scudder Coll.), belong to the same species. There
are also three additional specimens in my collection.

While shorter and broader than most recent species of Limonius,
the essential characters, as shown by the coxal plates (which are
narrow and but little dilated internally, only moderately prominent
over the insertion of the thighs), the prosternal sutures, (double,
little curved), the short prosternal lobe and the blunt scutellum
correspond very well with this genus. The basal antennal structure
is similar to what we see in the recent L. crotchii of the western United
States, but the general aspect is more that of L. nitidulus from the
same district.

Limonius florissantensis, sp. nov.
Plate 5, fig. 5-7.

Form moderately elongate. Head rather finely and extremely
closely and deeply but regularly punctured. Antennae about reach-
ing the prothoracic base, faintly serrate. Prothorax approximately
one seventh broader than long, apex and base subec^ual, surface quite
evenly punctate, about as coarsely as the head but more sparsely,
apex nearly truncate, front angles obtuse, sides regularly arcuate to
about the middle, which is the broadest part, thence narrowing to
near the base, hind angles acute, carinate, but hardly divergent.
Scutellum oblong oval. Elytra about two and two thirds times the
length of the prothoracic median line, finely, sharply striate, strial
punctures fine, somewhat elongate, separated by approximately
their own long diameters, interstitial spaces flat, broad, confusedly
and sparsely punctate, the punctures of varying sizes, the largest
distinctly smaller than those of the striae. Underside punctured
throughout, rather coarsely and closely on the prosternum and flanks,
more finely on the meso- and metasternal sclerites, the abdomen
finely punctate except on the last segment and along the sides, where
the sculpture is coarser. Length, 8.40 mm.; of elytron, 5.50 mm.

Described from two specimens, one with counterpart.
Type. — In the collection of H. F. Wickham. Wilson Ranch,
Florissant, Colo. With it are associated another specimen, with
counterpart, in my own collection; two from Station 14 in the Mu-

516 bulletin: museum of comparative zoology.

seum of the University of Colorado; No. 6,572 of the Princeton col-
lection; and No. 2,734 M. C. Z. (No. 11,664 S. H. Scudder Coll.). It
is possible that No. 2,735, 2,736 M. C. Z. (No. 8,340, 10,492 S. H.
Scudder Coll.), represent the same species.

Characters pointing to Limonius are seen in the coxal plates, hind
tarsi, prosternal sutures, and antennae. The short prosternal lobe
is more like Nothodes. Compared with L. aboriginalis, the present
species is much more roughly sculptured.

Limonius praecursor, sp. nov.
Plate 5, fig. 8, 9.

Form elongate, slender. Head finely but very regularly and dis-
tinctly punctured, the punctures separated by their own diameters
or less. Antennae short, showing only a few of the joints well enough
for description, but these are about one half longer than wide and
weakly serrate. Prothorax punctured similarly to the head but a
little more finely, the sides not in very good condition but evidently
subparallel, length and width subequal. Scutellum obscure. Elytra
about two and two fifths times the prothoracic length, subparallel,
finely but very distinctly striate, the striae with decidedly elongate
punctures which are separated in each row by their own diameters,
a little more or less, interstitial spaces broad and flat, very minutely
pimctulate. Underside of prothorax finely but clearly punctured,
rather closely on the flanks, less so on the prosternum, sutures double,
a little curved in front, nearl}^ straight behind, broader anteriorly,
lobe moderate. Length, from front of head to elytral apex, 8.50 mm.;
of elytron, 5.75 mm.

Described from one specimen, with counterpart.

Type.— No. 2,730 and 2,731 M. C. Z. Florissant, Colo. (No.
9,417 and 10,558 S. H. Scudder Coll.). With it is doubtfully asso-
ciated another specimen, also with counterpart. No. 2,732 and 2,733
M. C. Z. (No. 12,049 and 12,762 S. H. Scudder Coll.).

In general form and sculpture, this approaches closely to L. floris-
santensis. However; the present insect has relatively shorter elytra,
with markedly finer and sparser cephalic and prothoracic punctuation.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 517

LiMONius SHOSHONis, sp. nov.
Plate 5, fig. 10.

Form fairly elongate. Head finely and rather vaguely punctate.
Antennae lacking. Prothorax almost exactly equal in length and
breadth, surface finely, not deeply, but fairly closely punctate, apex
not much narrower than the base, sides feebly arcuate, front angles
nearly rectangular, hind angles carinate, sharp, only a little divergent.
Scutellum nearly triangular. Elytra moderately tapering, finely
striate, strial punctures a little elongate, quite fine, not very deep and
separated from each other in the same row by approximately their
own diameters, interstitial spaces flat, wide, scarcely visibly punctu-
late, Qnely hairy. Length, from front of head to elytral apex, 7.25
mm.; of elytron, 4.35 mm.

Described from one specimen.

Type. — In the Museum of the University of Colorado. It was
collected by Professor Cockerell at Station 14, Florissant, Colo., in
1906 and bears his number 58.

The beetle is provisionally placed in Limonius chiefly on account of
its form and sculpture. It is smoother than the other Florissant
species referred to this genus.

Limonius volans, sp. nov.

Plate 5, fig. 11.

Form rather elongate. Head deeply punctured, closely and rela-
tively coarsely on the front, more finely and sparsely on the vertex.
Antennae wanting. Prothorax a trifle broader than long, more
finely and much more sparsely pvmctured than the head except near
the side margins where the sculpture is much coarser and closer than
on the disk. Base somewhat broader than the apex, sides very little
arcuate, front angles short, hind ones of moderate length and but
slightly divergent. Elytra two and three fourths times the prothora-
cic length, conjointly rounded at apex, finely striate, the strial punc-
tures not very close nor well defined, interstitial spaces flat, with
strong, fairly sparse punctures, more pronounced, though probably
not larger, than those of the striae. Length, from front of head to
elytral apex, 9.00 mm.; of elytron, 5.60 mm.

r

518 bulletin: museum of comparative zoology.

Described from one specimen.

Type. — In the Museum of the University of Colorado. It was col-
lected at Station 14, Florissant, Colo., by G. N. Rohwer, while a
member of one of Professor Cockerell's parties.

Separates from L. florissantensis, with which it agrees in the rela-
tively coarse interstitial punctuation, by having a much finer and
sparser sculpture of the head and particularly of the thorax. The
generic reference is provisional, being based mostly on facies.

Athous lethalis, sp. no v.
Plate 6, fig. 1, 2.

Form elongate, parallel. Head finely and extremely densely punc-
tured and with a short pubescence. Antennae long, slender, faintly
serrate, apparently not entire but reaching far beyond the prothoracic
hind angles, basal joints too poor to allow of their definition. Pro-
thorax punctured similarly to the head but a trifle more coarsely and
less deeply, length and breadth equal, front angles slightly prominent,
sides nearly straight to the hind angles which are acute and a little
divergent, base sinuate each side. Scutellum oblong oval. Elytra
a little over three times the length of the prothoracic median line,
apices conjointly rounded, finely striate and pubescent, the striae with
small, deep, nearly circular or slightly elongate punctures which are
separated in the series by their own diameters or something more.
I'nderside of prosternum closely and finely punctured, the prothoracic
flanks less strongly, sculpture of the remainder of the thoracic scle-
rites and abdomen very obscure. Length, from front margin of pro-
thorax to elytral tip, 8.40 mm.; of elytron, 5.50 mm.

Described from one specimen, with counterpart.

Type.— No. 2,728 and 2,729 M. C. Z. Florissant, Colo. (No. 8,464
and 8,713 S. H. Scudder Coll.).

The prothorax is ornamented with a broad brown stripe, about one
third of the discal width, occupying the median area from base to
apex, similar to that seen in the recent A. excavatus, from California.
The latter insect, however, is much more coarsely sculptured. The
coxal plates are not well displayed in the fossil, but the prosternal
lobe and sutures, as well as the general form, correspond well with
the genus in which I have placed it.

WICKHAM: fossil ELATERIDAE of FLORISSANT. 519

Athous contusus, sp. nov.
Plate 6, fig. 3, 4.

Form very elongate, subparallel. Head moderately coarsely and
fairly closely punctured. Antennae slender, and, when complete,
probably reaching or passing the pro thoracic hind angles. Only a
few of the joints are well defined and these are scarcely serrate. Pro-
thorax long, narrow, the sides not in good condition, apparently Avider
at base than at apex, hind angles only moderately pronounced, flanks
rather closely but not coarsely punctured, prosternum more strongly.
Elytra long, conjointly rounded at apex, strongly sculptured, the
punctures of the striae rounded, separated longitudinally by their
own diameters or a little more. Abdominal punctuation fine, moder-
ately close. Length, from front of head to elytral apex, 11.15 mm.;
of elytron, 7.65 mm.

Described from one specimen.

Type.— No. 2,727 M. C. Z. Florissant, Colo. (No. 8,346 S. H.
Scudder Coll.).

The specimen is preserved in such a way as to show the upper side
of the head and elytra and the details of the underside of the pro-
thorax, due to the manner of splitting the stone. It retains a portion
of the raised frontal margin and the aspect is quite that of Athous.

Athous fractus, sp. nov,

Plate 6, fig. 5.

Form elongate. Head with a rather well-pronounced frontal mar-
gin, surface finely punctate and pubescent. Antennae weakly serrate,
slender, the basal and apical joints poorly defined, but in life the
antennal tip evidently attained or passed the prothoracic hind angles.
Prothorax very little broader than long, surface finely, not closely
punctate and pubescent. Sides very little arcuate, angles small, the
hind ones not in good preservation but evidently carinate and at least
moderately prominent. The notch in front of the angle, as shown
on the figure, is perhaps adventitious. Scutellum oblong. Elytra
broken at tip but apparently, if complete, not much, if any, less than
three times as long as the prothorax, finely striate, the striae with
irregularly spaced, round, or often elongate or elliptical, punctures

520 bulletin: museum of comparative zoology.

separated by more or less than their own long diameters. Interstitial
spaces flat, broad, pubescent but scarcely visibly punctulate. Only
one leg shows, which is of moderate size. - Length of fragment, 13.40
mm.; of prothorax, along median line, 3.50 mm.

Described from one specimen.

Type. — In the Museum of the University of Colorado. It was
collected at Station 14, Florissant, Colo., by S, A. Rohwer.

Placed in Athous because of the form, the coxal plates (only in-
distinctly seen), the frontal margin and the very long prosternal lobe
which shows through as indicated, Plate 6, fig. 5. In this figure, the
dotted lines will show the courses of the elytral striae, but the punc-
tures are actually somewhat smaller and more numerous than the dots
which might be taken to represent them. It seems smoother than
the recent North American species known to me.

Paranomus exanbl^tus, sp. nov.
Plate 6, fig. 6, 7.

Form only moderately elongate. Head practically smooth. An-
tennae not well enough preserved to show the relative sizes of most
of the joints, but they are quite weakly or scarcely serrate, reaching,
in life, beyond the prothoracic hind angles. Prothorax in poor
condition and probably somewhat distorted, but as shown it is a little
more than one fifth broader than long, wider in front of the middle,
front angles a little acute, sides moderately arcuate in anterior three
fourths, thence sinuate, in reverse curve, to the hind angles which
are sharp and slightly divergent, base broadly emarginate in front
of the scutellum, sinuate each side, surface minutely, sparsely punc-
tured. Scutellum suborbicular. Elytra three times the length of the
prothoracic median line, conjointly rounded apically, not striate nor
visibly punctured but finely pubescent. Underside nearly smooth.
Length, 7.00 mm. ; of elytron, 4.30 mm.

Described from one specimen, with counterpart.

Type. — In the collection of H. F. Wickham. Wilson Ranch,
Florissant, Colo. With it is associated, somewhat doubtfully, an-
other from the same source.

Most probably a Paranomus, but more finely sculptured than P.
costalis or P. estriatus, the only recent species know to me.
The prosternal sutures are moderately curved, the hind coxal

WICKHAM: fossil ELATERIDAE of FLORISSANT. 521

plates but slightly dilated externally and without a distinct tooth
over the thighs. The beetle differs from P. laevissimus in the
proportions of the elytra and prothorax, as well as in some minor
details which may be gathered from the descriptions.

Paranomus LAEVISSIMUS, sp. nov.
Plate 6, fig. 10.

Form fairly stout. Head \-ery finely punctulate, a little more
coarsely anteriorly. Antennae poorly preserved, slender, scarcely
at all serrate. Prothorax almost absolutely smooth, neither side
completely preserved but from a combination of the two it is evident
that the base and apex were subequal, the width about one third
greater than the length, sides arcuate, sinuate in front of the hind
angles which are somewhat divergent, acute and carinate, basal
margin strongly sinuate each side. Scutellum obscure, apparently
oblong. Elytra two and three fifths times the length of the pro-
thoracic median line, without sculpture except two lines of faint
elongate punctures near the outer edges and some still weaker ones
on the disk, no visible hair marks. Underside almost perfectly
smooth. Length, from front of head to elytral apex, 8.10 mm.; of
elytron, 4.75 mm.

Described from one specimen, with counterpart.

Type. — In the Museum of the University of Colorado, collected at
Station 14, Florissant, Colo., by Mrs. W. P. Cockerell.

Perhaps the most striking characteristic of this beetle is the almost
total lack of sculpture. The form is like that of Cardiophorus and
the curved prosternal sutures are similar to those found in that genus,
but the spine is not truncate nor is the scutellum cordiform. The
nature of the prosternal sutures forbids reference to Cr^'ptohypnus,
and while the coxal plates are not distinctly shown I think they are
gradually smaller externally as in the Corymbitini. The practical
lack of elytral striation leads me to refer the insect to Paranomus.

Paranomus heeri, sp. nov.

Plate 6, fig. 8, 9.

Form only fairly elongate. Head minutely, sparsely pimctulate
on the vertex, more closely at front and sides. Antennae bent under

522 bulletin: museum of comparative zoology.

the body, lying along the breast near the prosternal sutures, not well
enough defined for description. Prothorax punetulate, finely and
sparsely, the outline incomplete on one side but evidently the width
is about one half greater than the median length, base and apex
subequal, front angles obtuse, sides regularly rounding to the hind
angles which are obscure and probably short. Scutellum oval.
Elytra a little over three times the prothoracic length, apices con-
jointly rounding, surface distinctly and rather deeply but finely and
sparsely punctured without any definite strial arrangement. Under-
side obscurely, finely punctate. Length, from front of head to elytral
apex, 4.65 mm.; of elytron, 3.25 mm.

Described from one specimen, with counterpart.

Type. — In the Museum of the University of Colorado. It was
collected at Station 14, Florissant, Colo., by Dr. W. M. Wheeler,
while a member of one of Professor Cockerell's expeditions.

Probably not a true Paranomus, but I can find no better place for
it and the assemblage of visible characters points in that direction.
The metacoxal plates are not suddenly dilated, the prosternal lobe
is moderate, the sutures nearly straight, apparently slightly excavate
anteriorly, the elvtra not striate. The size is somewhat less than that
of the recent P. estriatus, from Mt. Washington.

LuDiOPHANES, gen. nov.

Form of Ludius. Elytra confusedly punctate, not striate. Coxal
plates gradualh' narrowed externally and not toothed over the inser-
tion of the thighs. Scutellum ogival,

Ti/pe. — L. haydeni, sp. nov. ,

LUDIOPHANES HAYDENI, Sp. nOV.

Plate 4, fig. 7-9.

Form moderately elongate, tapering a little to both ends. Head
short, closely, deeply, and coarsely punctured, except on the extreme
frontal region where the sculpture is more shallow. Antennae very
slightly longer than the prothoracic median line but not reaching the
tips of the hind angles, eleven jointed, feebly serrate, first joint large,
second shorter than the third, third and fourth subequal in length,

WICKHAM: fossil ELATERIDAE of FLORISSANT. 523

eleventh much longer than the tenth. Eyes not strongly convex.
Prothorax a little broader than long, slightly narrower at apex than
at base, front angles acute, sides gently and almost regularly arcuate,
faintly sinuate in front of the hind angles which are acute and feebly
divergent. The thoracic apex seems hardly emarginate, the base is
notched in front of the scutellum and sinuate each side. Punctuation
of pronotum close, deep, and rather coarse over the entire surface,
the punctures ever;^^iere separated by much less than their own
diameters and but slightly less crowded along the median line than
at sides, each with a central mark which looks as if it may have been
the point of insertion of a hair or scale. Scutellum oval, much longer
than wide, coarsely punctured. Elytra moderately tapering, not
pointed at apices, confusedly but in general evenly punctured except
that the punctures become somewhat more sparse posteriorly where
they are separated by spaces about equal to or a little more than their
own diameters. Each puncture carried a moderately long, curved
dark hair, giving a somewhat shaggy appearance to the surface.
Underside of body well preserved, showing the following features : —
prothoracic flanks finely and densely punctured, prosternum, includ-
ing the spine, more coarsely and deeply; lobe strong, roimded;
sutures double, nearly straight, excavate anteriorly; meso- and
metasternal areas similarly but in general less closely punctate, coxal
plates narrow, little dilated externally, with a rounded lobe over the
insertion of the thighs; abdominal punctuation rather fine but deep,
closer externally but everywhere well separated. Legs not well
displayed. Length, from front of head to tip of abdomen, excluding
sex organ, 14.25 mm.; of prothorax, along median line, 3.35 mm.;
of elytron, 9.00 mm.; of antenna, 3.65 mm.; width of prothorax,
3.75 mm.

Described from one specimen, %vith coimterpart.

Tyye. — In the collection of H. F. Wickham. Wilson Ranch,
Florissant, Colo.

This is probably the finest and best preserved specimen I have
seen among the Elateridae of the Florissant shales. The aspect is
much like that of a Ludius or of a Megapenthes, like the recent west-
ern North American M. aterrimus, but the form of the coxal plates
indicates a position with the Coryrabitini.^ The punctuation does
not agree with that of any species of the group known to me and
serves at once to differentiate it from all the Florissant fossil Elateridae
of similar size. The antenna and sex organ are omitted (Plate 4,
fig. 7) but the former is shown (Plate 4, fig. 8).

524 bulletin: museum of comparative zoology.

CoRYMBiTES PRiMiTivus Wickham.

Described in American journal of science, 1908, ser. 4, 26, p. 77,
fig. 2. It is a large species, about 22 mm. in length, and seems not
to have been particularly rare. The type is in the Peabody Museum
of Yale University and was found at Station 14, Florissant, Colo.,
by G. N. Rohwer. None are in the collections of the Museum of
Comparative Zoology, but I have a good example, with counterpart,
and the prothorax of another from the Wilson Ranch.

CORYMBITES GRANULICOLLIS Wicldiam.

This was described in the same article as the preceding. It is still
larger, 24 mm. long, and is the most striking of all the Florissant
Elateridae as far as size is concerned. The type is with the Peabody
Museum, at Yale University. I have a very good specimen obtained
at Florissant from a local collector who claimed to have found it in
the railroad cut that runs through the Corixa bed. I doubt the
accuracy of his statement. The original locality was Station 14,
which has yielded many beautiful insects of various families.

CORYMBITES SUBMERSUS, Sp. UOV.

Plate 7, fig. 1-3.

Form fairly elongate. Head moderately coarsely punctured on the
front, vertex becoming only faintly sculptured. Antennae broken,
but enough remains to show that the second joint is shorter than the
third and, judging from the portions preserved, the organ, when com-
plete, reached slightly beyond the points of the prothoracic hind
angles. The antennal serrations are faint. Mandibles a little promi-
nent. Prothorax short, narrow anteriorly, broadest across the base,
the sides arcuate near the front angles, which are obtuse, but becoming
nearly straight to the hind angles which are long, sharp, divergent,
and carinate. The sculptiu"e of the pronotum consists of an extremely
fine punctuation, with rather sparse pubescence. There is a fine
lateral marginal bead the full length of each side. Scutellum oblong.
Elytra striate, the striae with fine, deep, rounded or somewhat elon-

WICKHAM: fossil ELATERIDAE of FLORISSANT. 525

gate punctures separated in the series by something more or less than
their own diameters. Underside showing that the prosternum is
strongly, closely and rather coarsely punctured around the anterior
portion of the lobe, nearly smooth at middle, side pieces vaguely
punctate, abdomen thinly pubescent, the punctuation fine, shallow
and sparse. Legs poorly preserved, of moderate size. Length, from
front of head to abdominal apex, 13.50 mm.; of elytron, 9.45 mm.;
of prothorax, along median line, 2.45 mm. ; width of prothorax across
base, just in front of hind angles, 3.35 ram.

Described from one specimen, with counterpart.

Type. — In the Museum of the University of Colorado. It was col-
lected by G. N. Rohwer, at Station 14, Florissant, Colo.

Easily distinguished from the other Florissant fossil Corymbites
by the short prothorax and widely divergent hind angles which give
the appearance of the recent C. appressus from the northern United
States and Canada. That species, however, has the prosternum and
elytra differently sculptured. The prosternal, coxal, and antennal
characters of the fossil all agree well with the genus in which it is
placed.

Corymbites restructus, sp. nov.

Plate 7, fig. 4.

Form fairly elongate, tapering to the ends as in the recent C. hiero-
glyphicus. Since the specimen is exposed in ventral view, no de-
scription of the upper surface can be given. Head poorly preserved,
antennae not shown. Prothorax rather closely and fairly coarsely
punctured beneath, more deeply and strongly on the prosternum,
spine margined at sides, lobe much rounded, hind angles long and
acute, slightly diverging. Metasternal and abdominal punctuation
much shallower and more vague than that of the prothorax. Hind
leg moderately long, the first tarsal joint not much lengthened, the
fourth and fifth obscured. Elytra pointed at apex, sides rather
strongly rounding. Length, from front of head to abdominal apex,
but without extruded sex organ, 16.30 mm.

Described from one specimen.

Type. — In the Museum of the University of Colorado. It was col-
lected by Mrs. W. P. Cockerell at Station 14, Florissant, Colo.

Undoubtedly a Corymbites and easily distinguished from C. primi- "
thus or C. granulicollis by its smaller size and more fusiform outline.

526 bulletin: museum of comparative zoology.

CORYMBITES PROPHETICUS, sp. nOV.

Plate 7, fig. 5.

Form stout. Head in very poor preservation, finely, regularly,
and closely punctate. Antennae wanting. Prothorax at sides punc-
tured almost exactly like the head, a little more finely and sparsel\- on
the disk, about one fourth broader than long, sides nearly regularly
arcuate, somewhat more suddenly in front, anterior angles slightly
acute, hind ones quite strongly so, feebly divergent and carinate'.
Scutellum oblong. Elytra quite strongly arcuate at sides, apices
conjointly rounded, finely striate, striae with distinct but not coarse
punctures which are circular or slightly elongate and separated in each
series by their own diameters or less. Interspaces broad, flattened
or nearly so, finely punctulate and distinctly pubescent. Legs
wanting. Length, from front of head to elytral tip, 10.90 mm.;
of elytron, 6.55 mm. ; of prothorax along median line, 3.00 mm.

Described from one specimen.

Type — No. 2,724 M. C. Z. Florissant, Colo. (No. 13,657 S. H.
Scudder Coll.). With it is associated, somewhat rloubtfully. No.
2,275 M. C. Z. (No. 11,282 S. H. Scudder Coll.). Two poor specimens
from Station 13 and 13B are in the Museum of the L^niversity of
Colorado.

This is quite surely a Corymbites and is of the same general form
as the recent C. aereipennis, common in the northern and mountain
regions of this continent. The sculpture seems not to have been very
different. The size and outline will distinguish it from all the other
fossil Florissant species.

OXYGONUS PRIMUS, Sp. nOV.

Plate 7, fig. 6.

Form fairly stout. Head not well preserved, showing no sculpture.
Antennae moderately long but the individual joints are not definable.
Prothorax stout, suborbicular, the sides strongly roimded, apex and
base subequal, prosternal grooves double, somewhat curved, lobe
long, front edge quite arcuate and strongly advanced, sculpture fine
or nearly wanting. Elytron about four times as long as wide, moder-

WICKHAM: fossil ELATERIDAE of FLORISSANT. 527

ately pointed apically, punctatostriate, the punctures rounded or
somewhat elongate and separated in each row by more than their
own diameters. Abdomen and legs wanting. Length, from front
of head to elytral apex, 6.00 mm. ; of elytron, 3.85 mm.

Described from one specimen.

Type.— No. 2,726 M. C. Z. Florissant, Colo. (No. 6,381 S. H.
Scudder Coll.).

A small species, about the size of the recent Californian 0. ater.
The form of the prothorax will separate it at once from all the other
fossil Florissant Elateridae.

Melanactes cockerelli Wickham.

Originally described in the American journal of science, 1908, ser. 4,
26, p. 77, fig. 3. The type is in the Peabody Museum of Yale Uni-
versity and was collected at Station 14, Florissant, Colo. No other
specimens have come to light. It is a large insect, 23.50 mm. in length
and similar in general appearance to the Florissant fossil Corymbites
granulicollis.

PLATE I.

WicKHAM. — The Fossil Elateridae of Florissant.

PLATE 1.

Fig. 1. Deltometopus fossilis.

2. Deltometopus fossilis, antenna.

3. Fornax relictus.

4. Microrhagus miocenicus.

5. Microrhagus vulcanicus.

6. Lacon exhumatus.

7. Lacon exhumatus, antenna.

8. Cr>i3tohj7)nus exterminatus.

9. Crj'ptohjTJnus exterminatus, antenna.

BULL. MUS. COMP. ZOOL.

Florissant Elateridae. Plate 1

'DC:

PLATE 2.

WiCKHAM. — The Fossil Elateridae of Florissant.

PLATE 2.

Fig. 1. Cardiophorus lithographus.

2. Cardiophorus lithographus, underside of prothorax.

3. Cardiophorus lithographus, hind coxal plate.

4. Cardiophorus florissantensis.

5. Cardiophorus florissantensis, hind coxal plate.

6. Cardiophorus cockerelh.

7. Cardiophorus requiescens.

8. Cardiophorus requiescens, prosternal spine.

9. Cardiophorus (?) deprivatus.

10. Cardiophorus (?) deprivatus, antenna.

11. Horistonotus coloradensis.

12. Horistonotus coloradensis, hind coxal plate.

BULL. MUS. COMP. ZOOL.

Florissant Elateridae. Plate 2

PLATE 3.

WiCKHAM.— The Fossil Elateridae of Floriasant.

PLATE 3.

Fig. 1. Cryptohypnus hesperus.

2. Anchastus eruptus.

3. Anchastus eruptus, antenna.

4. Anchastus diluviaUs.

5. Elater rohweri.

6. Elater rohweri, underside of prothorax.

7. Elater scudderi.

8. Elater scudderi, antenna.

9. Elater florissantensis.

BULL. MUS. COMP. ZOOL.

Florissant Eumeridae. Plates

/>■

^^^^

PIA

WicKHAM.— The FoasU Elateridae of Floriaaant.

PLATE 4.

Fig. 1. Monocrepidius dubiosus.

2. Megapenthes primaevxis.

3. Cryptagriotes minusculus.

4. Agriotes comminutus.

5. Agriotes comminutus, hind coxal plate.

6. Agriotes nearcticus.

7. Ludiophanes haydeni.

8. Ludiophanes haydeni, antenna.

9. Ludiophanes haydeni, hind coxal plate.

BULL. MUS. COMP. ZOOL.

Florissant Elateridae. Plate 4

PLATE 5.

WiCKHAM.— The Fossil Elateridae of Florissant.

PLATE 5.

Fig. 1. Limonius aboriginalis.

2. Limonius aboriginalis, underside of prothorax.

3. Limonius aboriginalis, hind coxal plate.

4. Limonius aboriginalis, antenna,

5. Limonius florissantensis.

6. Limonius florissantensis, hind coxal plate.

7. Limonius florissantensis, hind tarsus.

8. Limonius praecursor.

9. Limonius praecursor, antenna.

10. Limonius shoshonis.

11. Limonius volans.

BULL. MUS. COMP. ZOOL.

Florissant Elateridae. Plate 5

PLATE 6.

WiCKHAM.— Tlie Fossil Klaleridae of Florissant.

PLATE 6.

Fig. 1. Athous lethalis.

2. Athous lethalis, underside of prothorax.

3. Athous contusus.

4. Athous contusus, antenna.

5. Athous fractus.

6. Paranomus exanimatus.

7. Paranomus exanimatus, hind coxal plate.

8. Paranomus heeri.

9. Paranomus heeri, hind coxal plate.
10. Paranomus laevissimus.

BULL. MUS. COMP. ZOOL.

Florissant Elateridae. Plate 6

/

PLATE 7.

WiCKHAM.— 'I'he Fossil i:iuleridae of Florissant.

PLATE 7.

Fig. 1. Corymbites submersus.

2. Corymbites submersus, underside of prothorax.

3. Corymbites submersus, hind coxal plate.

4. Corymbites restructus.

5. Corymbites propheticus.

6. Oxygonus jn-imus.

BULL. MUS. COMP. ZOOL.

Florissant Elateridae. Plate 7

Ji"^

II

S

ACME
BOOKSIW'JiNG CO.. INC.

NOV 2 9 1983

100 CAMCRIDC-!-: STREET
CHARLESTOWN. MASS.

Harvard MCZ Libra

3 2044 066 303 280

Date Due

jW*§==«96