pfnaenst agate pert taeeayateaee ctor acyeererr aes Tes

(
.
ee

is

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

VOLUME XIII.

LONDON:

PRINTED FOR THE SOCIETY :
SOLD AT THEIR HOUSE IN HANOVER-SQUARE ;
AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.

1895.

My ; “eh ereb: eae a Uh ay AG a

“ Yosrahe i airett.
aS

t

wikaturalty Fe ah) Til
ee eran

he Reyes Ara) rc é os
; at ae

“ i % ora

TH way a ta
jit agent

i acité } ,

Nok eee Nae or
PRINTED BY TAYLOR AND FRANCIS,

¢ Rey A aks ee
_ RED LION COURT, FLEET STREET. ~~

CONTENTS.

I. On the Genus Urothoe and a new Genus Urothoides. By the Rev. THomas R. R.
STEBBING cA pmo oot coh ay nye: Gk» sig tepe isthe. san as) We Pach es ie cele pOeene

II. On four new British Amphipoda. By the Rev. Tuomas R. R. Srespine, M_A., and
IDA) JRoriomnsO, JEG, JGR 6 5 een be ee ol Vs Jo go SI

III. On the Morphology of a Reptilian Bird, Opisthocomus cristatus. By W. K.
] AT 2 ial Ca an ale gem ne sh Gd ne SIRE Vd is A ER «Sateen WAR Saag ete Re CN ete}

IV. Contributions to our Knowledge of the Antipatharian Corals. By F.Jurrrey BELL,
M.A., Sec. R.M.S., Corr. Mem. Linn. Soc. N.S.W., F.Z.S., Professor of
Comparative Anatomy and Zoology in King’s College, London. . . . . 87

V. Cataloque of the Reptiles and Batrachians of Barbary (Morocco, Algeria, Tunisia),
based chiefly upon the Notes and Collections made in 1880-1884 by M. Fernand
Matasteume bin Gee Ave OULENGE Rial. ys.) eset alone eg Fgincemincn remem mn

VI. On a Skull of Trogontherium cuvieri from the Forest Bed of East Runton, near
Cromer. By E. T. Newton, F.G.S., P.Z.S., Geological Survey . . . . 165

VII. Contributions to the Anatomy of the Anthropoid Apes. By Frank EK. Bepparp,
M.A., Prosector to the Society, and Lecturer on Biology at Guy's Hospital . \77

VIII. On the British Paleogene Bryozoa. By J. W. Grucory, B.Sc., P.Z.S., British
NUOSCUTD (NGG. LECT) os Se ae Re is ernie oasetn ac ie 8711!)

IX. On additional Bones of the Dodo and other Extinct Birds of Mauritius obtained by
Mr. Tufopore Sauzter. By Sir Epwarp Newton, A.C.M.G., P.L.S., CMZS.,
and ANSI GAD OWay Rhu) ae Ar mE AS EMREK ZS) See ene) a Ol

1V CONTENTS.

X. Description of a remarkable new Sea-urchin of the Genus Cidaris from Mauritius.
By ¥. Jurvrey Buu, MA., Sec. R.M.S., Pr ee a Comparative Anatomy and
Zoology in King’s College. . . . « by ee 2 aye eespases03

XI. On Remains of an Extinct Gigantic Tor toise from Madagascar (‘Testudo grandidietri,
Vaillant). eBy Gr AG BOULENGER |e) ¢. ¢-.) 2) (2) ices eee SOe)

XIL. On the Remains of some Gigantic Land-Tortoises, and of an extinct Lizard,
recently discovered in Mauritius. By Hans Gavow, PhD. M.A, FRS.,
Lecturer on Advanced Morphology of Vertebrata, and Strickland Curator,
Onawensiian Oy CGD s 6 2 8 8 8 8 ee og eg aD

XII. A Revision of the Genera of the Alcyonaria Stolonifera, with a Description of one
new Genus and several new Species. By SypNey J. Hickson, M.A. Cantab.,

DSc. Lond., F.Z.S., Fellow of Downing College, Cambridge. . . . . 325

XIV. Descriptions of nine new Species of Amphipodous Crustaceans from the Tropical
Atlantic. By the Rev. Toomas R. R. Sveppine, MA. . . . . . 849

XV. On the Cranial Osteology, Classification, and Phylogeny of the Dinornithide. By
T. Jerrery Parker, D.Sc., PAS. ree He noe in the University of
Otago, Dunedin, New Zealand . . . . : Be eee ae OMS,

list otathesPaners:contaimedsin| Viol-eXelies eo at ee lS

ImdextoniSpecies: Occ... We BY AR a SE ea) at kia an ao

TRAINS AL Cue 1. ON Ss

OF

THE ZOOLOGICAL SOCIETY

1852.
1852.
1853.
1856.
1857.
1857.

1857.
1857.

1857.
1860.
1862.
1862.
1865.
1869.

1869.

OF LONDON.

I. On the Genus Urothoe and a new Genus Urothoides.
By the Rev. THomas R. R. Srrpsine, J/.A.

Received July 16th, 1889, read November 5th, 1889.

[Puates I.-IV.]

Genus Urornor, Dana, 1852.

Urothoe’, Dana, American Journ. Sci. and Arts, ser. 2, vol. xiv. p. 311.
iy Dana, United-States Explor. Exped. vol. xiii. pt. 2, pp. 908, 920.
Egidia, Costa, Rendiconto d. Soc. r. Borb. Acad. d. Scienze, p. 165.
Gammarus (pars), Spence Bate, On the British Edriophthalma, Brit. Assoc. Report for 1855.
ELgidia, Costa, Ricerche sui Crost. Amfip. Nap. pp. 174, 190.
Sulcator (pars), Spence Bate, Synopsis of British Edr. Crust., Ann, & Mag. Nat. Hist.
ser. 2, vol. xix. p. 140.
» (pars), White, Popular Hist. Brit. Crust. p. 175.
Urothoe, Spence Bate, Synopsis of British Edr. Crust., Ann. & Mag. Nat. Hist. ser. 2, vol. xix.
p. 145.
White, Popular Hist. Brit. Crust. p. 186.
Boeck, Forhandlinger ved de Skand. Naturf. 8de Méde, p. 646.
5 Spence Bate, Brit. Mus. Catal. Amph. Crust. p. 114.
is Bate & Westwood, British Sess. Crust. vol. i. p. 192.
5 Lilljeborg, On Lysianassa magellanica, p. 18.
a Grube, Mitth. titber St. Vaast-la-Hougue, Abhandlungen der schlesischen Gesell-
schaft fiir vaterlindische Cultur, 1868-9, p. 119.
v Norman, Last Report on Dredging among the Shetland Isles, Brit. Assoc. Report
for 1868, p. 279.

‘ The name is variously printed in different works as Urothoe, Urothdée, and Urothoe.

vol. XI1.—PART I. No. 1.—January, 1891. B

2 REV. T. R. R. STEBBING ON THE

1870. Urothoe, Boeck, Crust. amph. bor. et arct. p. 57.

1876. 3 Boeck, De Skand. og Arkt. Amph. p. 224’.

1876. 55 Giard, Comptes Rendus, Jan. 3, p. 76; Ann. & Mag. Nat. Hist. ser. 4, vol. xvii.
p- 261.

1876. Bs Stebbing, Ann. & Mag. Nat. ‘Hist. ser. 4, vol. xvii. p. 344.

Meinert, Crust. Isop. Amph. et Decap. Danie, p. 107.

1878. oy Spence Bate, the Crustacea in Couch’s Cornish Fauna revised, Journ. Roy. Inst.
Cornwall, no. xix. pt. i. p. 48.

1879. es Sars, Crust. et Pyen. nova, p. 446.

1879. y Stebbing, Sessile-eyed Crustacea of Devonshire, Suppl. List, Trans. Devon. Assoc.
vol. xi. p. 519.

1882. . Haswell, Catal. Australian Stalk- and Sessile-eyed Crustacea, p. 240.

1882. 0 Sars, Oversigt af Norges Crustaceer, p. 22.

1884. AS Chevreux, Assoc. pour l’ay. des Sciences, Blois.

1885. Egidia, Carus, Prodromus Faun Mediterranez, pars u1. p. 419.

1885. Urothoe, Sars, Den norske Nordhavys-Exped. p. 164.

1886. * Gerstaecker, Bronn’s Klassen und Ordnungen, Bd. v. Abth. ii. p. 501.

1887. 2 Barrois, Note sur Morph. des Orchesties et liste Amph. du Boulonnais, p. 16.

1887. ifs Bonnier, Catal. Crust. Malac. Concarneau, p. 79.

1887. Bs Chevreux, Crust. Amph. dragués par |’ Hirondelle, Bull. Soc. Zool. de France, t. xii.

extr. pp. 13, 15.

1887. a Chevreux, Catal. Crust. Amph. Bretagne, Bull. Soc. Zool. de France, t. xii. extr. p. 10.

1888. » Chevreux, Distr. géogr. Amph. sur les cdtes de France, Bull. Soc. d’études Sci. de
Paris, 11° année, 1% sem. 1888, extr. pp. 2, 7.

1888. ,, Chevreux, Amph. du littoral des Acores, Bull. Soc. Zool. de France, t. xiii. p. 34.

1888. BS Chevreux, Dragage de l’Hirondelle au large de Lorient, Bull. Soc. Zool. de France,
t. xiii. séance du 28 février.
1888. . Chevreux, Amph. réc. aux environs de Cherchell, Assoc. pour l’ay. des Sci, Oran,

séance du 3 avril.
1888. 33 Robertson, Catal. Amph. and Isop. Clyde, p. 29.
1888. ep Stebbing, Challenger Reports, vol. xxix. p. 824.

Upper lip distally rounded, strongly attached to the mandibles.

Mandibles having the cutting-edge scarcely denticulate, the secondary plate small,
spine-row wanting, the molar tubercle strong, all three joints of the palp slender and
rather elongate.

First maxille with inner plate slender, curved, the two joints of the palp subequal.

Mavillipeds with inner and outer plates of only moderate size, the second joint of the
palp large and dilated, the third joint apically dilated, the fourth narrow.

Upper antenne with a secondary flagellum; the three joints of the peduncle
moderately elongate, not very unequal in length, together longer than the few-jointed
flagellum.

’ In the first part of this work (1872) at p. 56, Boeck refers to Costa’s Hgidia as Aegidia.

GENERA UROTHOE AND UROTHOIDES. 3}

Lower antenne. Peduncle spinous; the flagellum in the male very long, and multi-
articulate, in the female very short, two-jointed; calceoli present in the adult male
both on the flagellum and on the last joint of the peduncle.

First and second gnathopods similar, subchelate, the wrist larger than the hand; the
finger of the first gnathopod longer than that of the second, each having a little
transparent cap over the tip.

First and second pereopods having the third joint longer than the strongly spined
fourth or fifth. 5

Third perwopods with the first, third, and fourth joints dilated, the third, fourth,
and fifth strongly spined, and furnished with long plumose sete; the fourth joint
as wide as or wider than the third.

Fourth pereopods the longest, with long plumose setze on the first and third joints.
In all the perwopods the finger has the inner margin more or less nodulous or serrulate
and has a cap over the tip.

Pleopods having the inner ramus conspicuously shorter than the outer, and in the
female the peduncle distally widened.

Third uropods. The outer ramus furnished with a small second joint.

Telson in general cleft nearly to the base.

Body generally obese, not much either depressed or compressed ; the side-plates not
very large. Gland-cells are distributed in great numbers over different parts of the
animal.

In this genus the eyes are very variable; in some, if not in all, species the eyes,
though very large in the male, are of moderate size in the female, and very small in
the young. In Urothoe abbreviata, Sars, no eyes were observed ; but the specimen being
very small and probably young, it cannot be inferred with certainty that this is a blind
species. ‘The second joint of the mandibular paip forms an angle or bend near its
base, owing to which its full length is sometimes not perceived. The inner plate of
the first maxillze appears to be always slender and slightly curved ; the first joint of the
palp seems to vary in its proportion to the second, to which it is sometimes equal,
while in other cases it may be longer or shorter than it. Boeck has remarked that,
when Dana attributed to this genus long and narrow maxillipeds with very small inner
lamelle, he was thinking only of his Urothoe rostratus, which belongs elsewhere.
Spence Bate was perhaps influenced by Dana’s account when describing the outer
plates as small, and the inner as rudimentary; they are, in fact, both of moderate
size. In most, and perhaps in all, species the wrist of the first gnathopod is distin-
guished from that of the second by having a row of short feathered sete planted just
within the distal margin. As arule, there is an oblique row of sete on the outer
surface of the third and fourth joints of the first and second pereopods; long feathered
setee are attached to the inner distal margin of the third and fourth joints and to the

inner surface of the fifth joint in the third perwopods, also to the inner surface of the
B2

4 REV. T. R. R. STEBBING ON THE

first joint and the hind margin of the third joint in the fourth perwopods, also to the
inner surface on the lower part of the second pleon-segment, and to the rami of the
third uropods.

There have been at various times assigned to this genus twelve named species, one
unnamed species, and one named variety. Four of the species have their proper
places in other genera. Uvrothoe rostratus, Dana, 1852, from the Sooloo Sea, was
transferred by Boeck in 1876 to Phoxus (now Phoxocephalus) ; and whether it belongs
to that genus or not, it is certainly excluded by the maxillipeds from Urothoe. In 1874
Professor 8. I. Smith recorded “ Urothoé, species,” as “apparently belonging to this
genus,” from Vineyard Sound, N.E. America, ‘This, however, proves to be a species of
the genus Harpinia, Boeck. Urothoé pinguis, Haswell, 1880, from Bondi, New South
Wales, also approaches the genus Harpinia, but is clearly removed from Urothoe by
the upper antenne, the maxillipeds, the gnathopods, and some other details. Urothoe
lachneéssa, Stebbing, 1888, from Kerguelen, should be transferred to a new genus
Urothoides, on account of the character of the fourth and fifth pereeopods and the third
uropods, as well as the absence of the plumose sete, which are so conspicuous in
the European species of Urothoe.

The type species, Urothoe irrostratus, Dana, 1852, had scarcely been published, when
another species of the same genus appeared under the name Egidia pulchella, Costa,
1853, from the Bay of Naples. The fuller description of this species was given in
1857, and in 1872 Boeck pointed out that Hgidia was a synonym of Urothoe. In the
generic description Costa says that the lower antenne have the first joint unarmed,
doubtless meaning that there is no gland-cone ; but this is a mistake, as the gland-cone,
though inconspicuous, is present. It is clear both from the generic and_ specific
characters that Costa confused some of the limbs, overlooking one of the first two pairs
of pereeopods altogether, and regarding the third pair as the second. In describing
the third pereopods he also evidently overlooked the second joint, so that, although
he gives a recognizable description and figure of the last four joints of the limb, he
supposes the last of the four, the finger, to be entirely wanting. It is clear from the
account of the lower antenne that Urothoe pulchella was described from a male
specimen. ‘The eyes are stated to be large, rounded-triangular, nearly meeting on the
back. The colour of the living animal was, according to Costa, pale green (verdiccio
pallido). ‘The prevailing colour in the genus is light buff, sometimes mottled with
pink or rose-colour. Rose-coloured specimens procured from Naples by Dr. Norman
must probably be assigned to Urothoé elegans, Bate. Other specimens from the same
locality, not so coloured, agree well with Costa’s species, though not having the hand
of the second gnathopods produced into a tooth confronting the finger. This tooth
may have been described and figured by Costa under a misapprehension owing to his
haying only had an oblique view of the little convex palm.

Cr

GENERA UROTHOE AND UROTHOIDES.

In 1856 Spence Bate published a third species of the genus under the name
frammarus elegans, which in 1857 he changed to Urothoe elegans. ‘The type
specimen from Plymouth was described in detail, and figured in the British-
Museum Catalogue in 1862. Many of the characters given are common to other
species. The long flagellum of the lower antenne shows that the specimen was
amale. The eyes are described as “nearly horizontal, long ovate,” the palms of the
gnathopods as “ oblique, imperfectly defined, ciliated,” the fingers of the pereeopods as
straight and sharp. In the fifth pereeopods, it is said, “a few long plumose cilia
mixed with short simple ones occur upon the posterior margins of the carpus and
propodos.” ‘The first and second uropods are said to be short, with the “rami very
short, shorter than their respective bases, subequal,” while the third pair are, as usual,
“long; rami longer than base, plumosely ciliated.” In the figure the postero-lateral
angles of the third pleon-segment are decidedly acute. In the ‘ British Sessile-eyed
Crustacea’ (1862) a figure of the species similar to that in the Catalogue is given, with
a much shorter description. Here the authors think two “important points,” the
shape of the eyes and the size of the antenne, sufficient to distinguish it from Urothoe
marinus. ‘The rest of the animal,” they say, “scarcely offers any specific variation
from U. marinus.” As the size of the antenne is dependent on age and sex, the only
character left for distinction is that of the eyes, which, the authors say, “are uniform.”
Since this has no meaning, it becomes a question whether the word intended may have
been reniform or oviform. An explanation is given that the specific name was suggested
by the extremely beautiful colouring,—whitish-buff, and parts mottled with pink,—but
no stress is laid upon this as a specific character. The Scotch specimens of Urothoe,
from the Clyde and the Shetland Isles, which make the nearest approach to this species,
agree with Spence Bate’s figure of it in not having the joints below the first in the third
pereeopods much dilated, and in haying the corresponding joints in the two following
pairs rather elongate; but no specimens that have come under my observation have the
postero-lateral angles of the third pleon-segment acutely produced as in Spence Bate’s
figure of Urothoe elegans, or the second uropods with rami so very short. ‘The dis-
crepancies may be explained by the fact that no dissection of the specimen was made.
This will account for the circumstance that the first and second pereeopods are drawn
with seven joints, the third only with five, and that no regard is paid to the long
plumose setee on the third and fourth pereeopods.

In 1857, under the name of Sulcator marinus, Spence Bate published a species which
in 1862 he figured and more fully described as Urothoe marinus, having received it from
the Moray Firth, the Shetlands, and the Clyde. It was pointed out by Giard in 1876 that
the brevity of the lower antenne relied on as a specific character only indicated that
the specimen examined was a female. In the description of these antenne it is stated
that the first joint of the peduncle is “furnished with three longitudinal rows of obtuse-
pointed spines,” and that the flagellum consists of a single joint; but that which by a

6 REV. T. R. R. STEBBING ON THE

slip of the pen is called the first, is in reality the fourth joint of the peduncle, and it is
surmounted by two, not three, rows of spines, although sometimes the spines of one
row by lying over in opposite directions give the appearance of a double row. The
flagellum consists of two joints, a long anda short one. ‘The first pair of gnathopods
are said to be much smaller than the second; but, while this would be unusual in the
genus, there is a fairly certain proof that the gnathopods have been interchanged in
the description, since the larger gnathopods are figured as having on the distal margin
of the wrist the little row of spines which is distinctive of the first pair. It is true that
both in the Museum Catalogue and in the ‘ British Sessile-eyed Crustacea’ these
enathopods are figured with the branchial vesicle attached, which of course cannot
belong to the first pair, and no doubt it was this apparent attachment of the breathing-
organ which led to the confusion. The segments carrying the two pairs of gnathopods
are in this genus so closely tied together, that the accident might easily arise of the
second gnathopod being detached, while leaving its branchial vesicle in apparent
connection with the first gnathopod. Of the crooked finger ascribed to the fifth
pereopods I can give no account, except that it appears to be abnormal or accidental.
The character most easy to seize for this species is to be found in the uropods, of which
the first pair have a short peduncle and short unequal rami, the inner ramus being the
shorter and much curved. In Spence Bate’s figure the tips of these rami do not reach
to the end of the short peduncle of the second pair, but in their natural position they
reach much further than this, since the fourth pleon-segment much overlaps the fifth
at the sides. The telson is accurately defined as “‘subapically furnished with a short
spine and several fine hairs,” but in no specimen that I have seen is it cleft quite to the
base as in the figure. The length of the animal is said in the Museum Catalogue to
be “y',ths of an inch;” but in the ‘British Sessile-eyed Crustacea’ we read “length
yeths of an inch,” without any explanation of the change. In the latter work a line
presumably indicating the natural size measures 3°; of an inch, and there can in fact
be no doubt that $3 is a misprint. Nevertheless Urothoe marinus is a much larger
and more robust species than Urothoe elegans.

Urothoé norvegica, Boeck, 1860, from Norway, was the species next established.
Figures and detailed description of it did not appear till 1876, and even with these the
species is still left in some obscurity. It may be presumed from the small size of the
lower antenne that the specimen examined was a female’; for, though Boeck states
that the flagellum has four joints, an inspection of the figure makes it tolerably certain
that there were actually no more than the usual two. Of the third pleon-segment
Boeck makes two statements—one, that the postero-lateral angle is produced upwards,
the other, that it is acute. His figure does not correspond with either particular. Of
the third pereeopods he remarks that the first joint is not much dilated. Specimens
obtained by Canon Norman from “ Sleat Sound, 1866,” and “ Shetland, 1867,” agree with

* For a different opinion, see Chevreux on this species, 1887.

GENERA UROTHOE AND UROTHOIDES. a

Boeck’s description in having the angles of the third pleon-segment acute (though not
produced upwards), and in having the first joint of the third pereopods rather less
dilated than is usual in the genus, and with the lower hinder angle of the joint very
much rounded. In Boeck’s work, De Skand. og Arkt. Amphip. plate vii., fig. 4 pro-
bably refers to these limbs, not, as the lettering would indicate, to the fifth pereopods.
In describing the fifth pereeopods he says that the fourth joint is longer and thicker than
the first. First is obviously a misprint for ///th, a correction which brings the statement
into agreement with Norman’s specimens above referred to, and with Boeck’s own figure
of the limb, plate vi. fig. 9m, where 9 m would indicate the third or the fourth pero-
pods, and is beyond doubt a mistake for 9”. Plate vii. has two figures marked 4 h, asif
representing the maxillipeds of Urothoe norvegica, but the upper one should probably
be 5h, referring to the species “ Phovus Holbélli ;”
the palp has been accidentally omitted. The figures 4g and 4m near the right-hand
lower corner of the plate are most likely also wrongly numbered.

In 1862 Spence Bate described and figured under the title * Urothoé Bairdii, n. s.,”
a specimen obtained, like Urothoe marinus, from the Moray Firth. The lower antenne
show that the specimen was a male, but not a male which had attained its fullest
development, so as to have calceoli and a very prolonged flagellum with very slender
joints. Both in the Museum ‘ Catalogue’ and in the ‘ British Sessile-eyed Crustacea’
the maxillipeds are figured, evidently by mistake, without a finger; in the latter, but
not in the former, work the lettering of the first and second pairs of uropods is
transposed. The latter work affirms that the third pereopods ‘‘ terminate in a knife-
shaped finger, the anterior margin of which is entire,” and the margin is so represented
in both works, owing, I believe, to defective observation, since the denticulation of this
margin may easily escape notice from some points of view, but has nevertheless proved
to be present in all specimens of the genus that I have examined. It seems to me
scarcely doubtful that Urothoe Bairdii is a synonym of Urothoe marinus. The pecu-
liarity of having the outer ramus of the third uropods devoid of plumose sete can
scarcely be relied on of itself to constitute specific distinction. It can derive no validity
from the sanction of a single, not fully developed, specimen. Boeck supposes Urothoe
Bairdii to be a synonym of his own Urothoe norvegica; but in the former species the
first joint of the third pereeopods is very broad instead of being, as in Boeck’s species,
comparatively narrow, and the rami of the first uropods are very unequal, while in
Boeck’s species they are equal, so that there is really no question of uniting these
two names.

In the same year (1862) Spence Bate established another species, upon a specimen
one-tenth of an inch in length from Tenby, under the name Urothoé brevicornis, n. s.
The specific name is unfortunate, since it is now known that the young and females
in all species of the genus have the lower antenne short, while they are long in the
adult males. As in the case of Urothoe marinus, so here it is evident that the first and

in the lower one the first joint of

8 REY. T. R. R. STEBBING ON THE

second gnathopods have been confused, the larger first gnathopods with the spines on
the apex of the wrist being figured and described as the second.

Professor Grube in 1869 described and figured a specimen from St, Vaast-la-Hougue
under the title “ Urothée marinus, Sp. Bate, t var. pectinatus, Gr.” (in the explanation
of the plate “ Urothoe marinus, Sp. B. (?var. pectina, Gr.).” The account of the lower
antenne shows that the specimen was a female. Grube says that these antenne have five
joints, the second much longer than the first, armed aboye with a subtriple row of spines,
the third simple, and armed with an outer row of rather long sete, the flagellum very
short, equalling the length of the third joint. The so-called second joint with the
subtriple row of spines is evidently the fourth joint of the peduncle, the so-called third
being the fifth, and the remaining two joints constituting the flagellum. He notices
that the finger of the second gnathopods is shorter than that of the first. He would
have been well content to assign the specimen to Urothoe marinus, Bate & Westwood,
but for certain differences in the pereopods, especially the third pair, and in the
uropods and telson. The stripes and patches of yellow-ochre which he notices on the
three hinder pairs of pereopods are not peculiar to any one species of Urothoe, being
due to the gland-cells which are found in almost every part of the animal. The spines
on the third, fourth, and fifth joints of the third pereeopods are not fully represented
in Spence Bate’s figures, but such details often escape notice, or are only perfunctorily
represented before their importance has been brought into special prominence, so that
it is rash of Grube to infer that had they been present they could not have been
neglected. In regard to the plumose sete on these limbs Grube himself has fallen
into some misapprehension. He says that Spence Bate speaks only of one row of
simple hairs on the hind rim of the third pereopods, while in Grube’s specimen they
are as decidedly plumose as Spence Bate figures them on the fifth pair’; the first joint,
too, appears in Spence Bate noticeably narrowed below, whereas he (Grube) found it
equally broad above and below. But Spence Bate speaks only of the first joint as
having the hind margin “ fringed with a row of simple hairs,” just as Grube himself
figures it. ‘The rest of the leg is remarkable for long plumose cilia,” according to
the account in the ‘ British Sessile-eyed Crustacea.’ In the full figure given in that
work of Urothoe marinus, the first joint of the third pereeopods answers to Grube’s
description of it; but in a full figure the limbs being seen at all sorts of angles often
give but a very rough idea of their actual details, and in the separate figure of the limb
in question it is at once seen that the first joint is, as is usual throughout the genus,
broader below than above. ‘The first uropods are omitted from Grube’s figure of his
specimen; but in the Latin description he says that the first and second uropods reach
almost equally far back, rarely beyond the peduncle of the third pair, having their
rami slightly curved. By this account the position of the species is left obscure. The

* It is in Vrothoe elegans, not in Urothoe marinus, that Spence Bate figures the plumose sete on the fifth
pereopods.

GENERA UROTHOE AND UROTHOIDES. 9

telson is described in the Latin as “latius lanceolatum, usque ultra dimidium fissum,”
and in the German as “cleft only to the middle, much longer than broad, running out
into two narrow points, each with a spinule and two sete.” In the species of Urothoe
in general the telson is cleft much beyond the centre.

In 1879 Sars described, under the name Urothoé abbreviata, n. sp., a single specimen;
3 millim. in length, taken in the sea north-west of Finmark, lat. 71° 25! N., long.
15° 40"5 E., from a depth of 620 fathoms. Of this in 1885 he gave a somewhat fuller
description and a figure. In the earlier account he distinguishes it from other species
by the abbreviate form of the body, the absence of eyes, and the rudimentary accessory
flagellum of the upper antenne ; in the later, “‘ by its remarkably short and thickset
body, the peculiar form distinguishing the first pair of antenn, the absence of eyes,
and by the short last pair of caudal stylets.” In , >xint of fact these characters are not
of much service for the recognition of the species. ‘he short and thickset body is not
the exception but the rule in this genus. In the young the eyes are also, as a rule,
very inconspicuous. ‘The first pair of antenne,” according to Sars, “are rather
elongate, and unlike those in all other known species.” Yet according to his descrip-
tion and figure of these organs they agree very well with the prevalent form, the
smallness of the flagellum in all probability only indicating that the specimen was a
very young one. Its diminutive size and the shortness of the last uropods are in accord
with this supposition. In the earlier account Sars states that the flagellum of the lower
antennee has four joints, just as Boeck does for his Urothoe norvegica; but in the later
account Sars omits all mention of this flagellum, only saying that these antenne from
their position are difficult to examine without dissection. ‘The circumstance that the
almost rectangular corners of the third pleon-segment are not produced upwards is
probably mentioned as a point of distinction from Boeck’s species, but, judging by
Boeck’s figure, the upward production is wrongly attributed to that species itself. On
the whole, with our present information, it is difficult to avoid the conclusion that
Urothoe abbreviata is the young of Urothoe norvegica, a deep-water form, of which
fuller and more definite details are still to be desired.

In 1888 M. Chevreux described, under the name “Urothoe Poucheti, nov. sp.,” a male
specimen taken at the surface, off Ponta Delgada, at the island of San Miguel, in the
Azores. ‘This species,” he says, “while tolerably near to Urothoe elegans, Sp. Bate,
differs from it by its less obese form, the size and peculiar appearance of the eyes, and
above all by the first two pairs of uropods, which are more developed than in other
species of the genus.” ‘The eyes prominent, very large, rounded, meeting one another
at the apex of the head,” would naturally be a very distinctive character to any one
judging only by the published figures of species in this genus. It is, however, common
to the fully developed males in several species, and agrees closely with Costa’s description
of the eyes in Egidia pulchella. ‘The uropods, of which M. Chevreux has kindly sent

VOL. XI1I.—Part I. No. 2.—January, 1891. ©

10 REV. T. R. R. STEBBING ON THE

me a drawing, correspond in general shape and proportions with those of Urothoe brevi-
cornis, Sp. Bate; but the peduncle of the first pair shows a series of twenty-four spines,
and the outer ramus has four spines, the outer ramus of the second pair carrying three
spines. Urothoe brevicornis, moreover, is not less, but more, bulky than Uvrothoe
elegans.

I have been most kindly assisted with specimens for this paper by M. E. Chevreux,
D. Robertson, Esq., and the Rev. Canon Norman. Dr. Norman had already taken
up the subject himself, but as soon as he heard that my work was further advanced
than his own he at once, more suo, placed at my disposal both his preliminary notes
and his collection of species from numerous localities.

UROTHOE IRROSTRATUS, Dana.

1852. Urothoe irrostratus, Dana, United-States Explor. Exped. vol. xiii. pt. 2, p. 922, pl. 62.

figs. 6 a-f.
1862. 5 ? Spence Bate, Brit. Mus. Catal. Amph. Crust. p. 117, pl. xx. figs. 3, 3e.
1876. oh 33 Boeck, De Skand. og Arkt. Amph. p. 225.

This, which has become the type species of this genus, was only partially figured by
Dana, under the impression that this and his Urothoe rostratus might be male and
female of one species. His description of Urothoe irrostratus is: —* Near the rostratus.
Front not rostrate. Flagellum of the superior antenne six- or seven-jointed, shorter
than the base; appendage very short, two- or three-jointed. Tarsi of feet of fourth and
fifth pairs nodulose along inner side, this side somewhat arcuate.’ Spence Bate inter-
prets the tarsi to mean the carpus; but in the special figures which Dana devotes to the
extremities of the fourth and fifth pairs of feet, the carpus is not included, while the
finger shows a nodulose convex inner margin in accordance with the description of the
so-called tarsi. It is unfortunate that Dana gives neither figure nor description of the
sixth and seventh pairs of feet (the fourth and fifth pereeopods). In his notes on the
genus he says “the six posterior legs [third, fourth, and fifth pairs of pereeopods] are
broad lamellar, especially the first, third, and fourth joints.” But just as he evidently
inferred without examination that the maxillipeds of Urothoe irrostratus were similar
to those of Urothoe rostratus, so he may have inferred without observation that the
fourth and fifth pereeopods would be alike in the two species. In the European species
the expression “broad lamellar” would not be especially appropriate to the third and
fourth joints of the last two pairs of peraeopods.

Dana figures the lower antenne with a long, slender flagellum carrying calceoli, so
that there can be no doubt that his specimen was a male. The three joints of the
peduncle of the upper antenne are figured as nearly equal in length.

The exact relationship of this species to its kindred in Europe cannot well be
determined until fresh specimens have been obtained from the Sooloo Sea. Should
there prove to be any very striking diversity in regard to the fourth and fifth pereopods,

GiNERA UROTHOE AND UROTHOIDES. 11

amounting to generic difference, it might be necessary to revive Costa’s name Egidia for
the European species; but it is not very probable that any such necessity will arise.

UROTHOE PULCHELLA (Costa). (Plate IV. A.)

1853. Egidia pulchella, Costa, Ricerche sw Crostacei Amfipodi del Regno di Napoli, Rend. d. Soc.
r. Borbon. Acad. d. Sci. n. ser. 1853.

1857. a 3 Costa, Ricerche sui Crostacei Amfipodi del Regno di Napoli, Mem. d. R.
Accad. d. Sei. di Napoli, vol. i. p. 190, tav. iv. fig. 38, a—g.

1876. Urothoé pulchella, Boeck, De Skand. og Arkt. Amph. p. 225.

1885. Egidia pulchella, Carus, Prodromus Faune Mediterraneze, pars ii. p. 419.

1888. Urothoé pulchella, Stebbing, ‘ Challenger? Amphipoda, p. 297.

Rostrum little developed, lower front corners of the head fully rounded; third
pleon-segment with the postero-lateral angles not acute.

yes reniform, except in the smaller stages, seemingly never very large.

Upper antenne. First joint thicker but by no means longer than the second, each
having two lines of setze on the surface; third joint shorter than the first, and much
more slender, nearly as long as the five- or six-jointed flagellum ; secondary flagellum
three- to four-jointed, longer than half the principal.

Lower antenne. In the male specimen examined the fourth joint of the peduncle had
twenty-one spines in the outer row, five in the inner, and several small sete ; the fifth
joint had a row of five spines near the outer margin, and eight long serrate sete near
the inner; the flagellum consisted of twenty-three joints, and was more than twice as
long as the peduncle. There being no brush of hairs on the fourth joint of the
peduncle, and there being spines but no calceoli on the fifth joint, this specimen may
not represent the fully adult male. In the female from Naples the fourth joint has
twenty-one spines in one row, two in the other, about a dozen long sete along one
margin, and seven at one corner; the fifth joint shorter than the fourth, with a row of
eleven spines, ten long sete near the convex border, and six at the opposite distal
corner; the flagellum scarcely as long as the fifth joint of the peduncle, the first joint
carrying three rather slender spines, and the second six long sete and three shorter
ones; the second joint is about one third the length of the first, its two apical
sete long.

The triturating organs have on the confronting margins each twelve or fourteen stout
spines toothed on two edges, the series being continued along the distal border by six-
and-twenty slender serrulate spines or sete.

Mandibles. The cutting-plate forming a strong blunt tooth; the secondary plate
small, on the left mandible denticulate, on the right more slender, strap-like, slightly
bifid; the molar tubercle powerful; the second joint of the palp carrying nine sete,
three or four of which are very small; the third joint about as long as the second,
nearly acute at the apex, carrying eleven spines, four or five at the apex seta-like.

c2

12 REV. T. R. R. STEBBING ON THE

Lower lip. The outer and inner lobes have the margins well furred ; the mandibular
processes are divergent.

First maxille. The slender, slightly curved, inner plate has on the convex outer
margin and apex six plumose sete; the outer plate has eleven spines on the truncate
distal margin, some straight, some curved, all, except perhaps the outermost, more or
less serrate, or with a subapical tooth ; the palp not reaching so far as the outer plate’s
apex, having at its base a little lobe of the trunk, the first joint rather broader and
longer than the second, which on the truncate apex carries three sete, the inner
serrate, the other two plumose.

Second maxiliw. Inner plate rather shorter than the outer, tending to oval in shape,
the apical end narrow, bordered with spines, plumose sete fringing both it and the
distal half of the inner margin; the outer plate with parallel sides, only the truncate
distal margin occupied with spines.

Mazillipeds. The inner plates with three plumose sete on the inner margin, two or
three teeth and several plumose sete or spines on the truncate distal margin ; the outer
plates bordered with seven stout spines and seventeen plumose sete of various sizes; the
second joint of the palp with the broad distal margin naked, the inner margin and
adjacent surface tnickly fringed with sete ; the third joint pear-shaped, with a very narrow
neck, the broad distal half of the joint set about with groups of sete, and having the
short and narrow slightly furred terminal joint attached to the middle of its distal margin ;
at the narrowed apex of the fourth joint there are two sete and three or four very small
ones near the apex.

First qnathopods. Side-plates not very narrow, the lower hinder angle forming a
well-produced rounded point, with a single spinule. Limb as in Urothoe marinus, but
with only seven spines in the spine-row adjoining the broad distal margin of the wrist,
and the hand oval.

Second gnathopods. The narrow hand is slightly distinguished from that of other
species by the prominence of the small convex palm, and is about four fifths as long as
the wrist.

First and second pereopods nearly as in Urothoe marinus, but the fourth joint has six
or seven spines, the fifth has nine or ten, and the finger has six or seven small tubercles,
which in the second pair are set very closely together. In the first pair the side-plates
have eight or nine spinules at the lower hind corner. The marsupial plates of the
female, here and apparently in the kindred species, long and slender, and when fully
developed having the distal half frmged with moderately long sete.

Third pereopods. The hind rim of the side-plates almost completely smooth. The
limb distinguished from that of Urothoe brevicornis by having the fourth joint very
much broader than the third, this joint (although belonging to a smaller species) having
more numerous spines in the spine-rows, and also on its distal margin a great number
of very long densely plumose sete ; the fifth joint is broad, and the finger not gently

GENERA UROTHOE AND UROTHOIDES. 1

os)

tapering as in Urothoe brevicornis, but narrowing rather abruptly as in Urothoe marinus,
from which, however, it differs in having on the front margin many little sharp denticles
rather closely set, even that nearest the tip being of no great size, nor does the tapering
part form a concave margin as in the last-named species.

Fourth pereopods nearly as in Urothoe brevicornis, the third joint not shorter than
the fourth, the fourth with four or five groups of spines on the front and two on the
hind margin ; the finger more than half the length of the fifth joint, very narrowly
tapering, with numerous little teeth or tubercles along the front margin.

Fifth perwopods nearly as in Urothoe marinus, but the side-plates less crenate, the
first joint scarcely at all more widened above than below, the fourth joint with four or
five groups of spines on the front and two on the hind margin, the slender tapering
finger more than half the length of the fifth joint, with from twenty to thirty little
tubercles along its front margin; the first joint in the female having the front margin
slightly more convex than in the male.

Pleopods nearly as in Urothoe marinus, but none of the peduncles elongate, and the
joints of the inner ramus not exceeding thirteen, those of the outer not exceeding
nineteen in number.

Uropods nearly as in Urothoe marinus, but outer branch of first pair with only two
spines, branches of second pair without armature, the branches of the third pair with
fewer sete, and the inner branch only reaching the base of the second joint of the
outer.

Telson short, with very convex sides, otherwise nearly as in Urothoe brevicornis.

Length about a fifth of an inch.

Localities. Naples (specimens obtained by Canon Norman) and off the west coast of
France (specimens obtained by M. E. Chevreux).

URorHoE ELEGANS, Sp. Bate. (Plate I.)

1856. Gammarus elegans, Sp. Bate, On the British Edriophthalma, Brit. Assoc. Report for 1855.

1857. Urothoé elegans, Sp. Bate, Synopsis of Brit. Edr. Crust., Ann. & Mag, Nat. Hist. ser. 2,
vol. xix. p. 145.

1857. 55 y White, Popular History of British Crustacea, p. 186.

1862. 99 a Sp. Bate, Brit. Mus. Catal. Amph. Crust. p. 117, pl. xx. fig. 2.

1862. 35 35 Bate & Westwood, British Sess. Crust. vol. i. p. 200, woodcut.

1869. s 5p Norman, Last Report on Dredging among the Shetland Isles, Brit, Assoc.
Report for 1868, p. 279.

1876. a es Giard, Comptes Rendus, Jan. 3, p. 76; Ann. & Mag. Nat. Hist. ser. 4,
vol. xvii. p. 261.

1876. bp op Stebbing, Ann. & Mag. Nat. Hist. ser. 4, vol. xvii. p, 344.

1879. 5 a Sp. Bate, The Crustacea in Couch’s Cornish Fauna reyised and added to
Journ, Roy. Inst. Cornwall, no. xix. pt. ii. p. 48 (sep. copy).

14 REV. T. R. BR. STEBBING ON THE

1884. Urothve elegans, Chevreux, Assoc. pour lav. des Sciences, Congrés de Blois, Amph. du
Croisic, p. 2 (sep. copy).

1887. 5 by Chevreux, Crust. Amph. Bretagne, Bull. Soc. Zool. de France, t. xil. extr.
p. ll.

1888. 5 3 Chevreux, Amph. du litt. des Acores, p. 5.

Rostral point obtuse, very slightly produced. Postero-lateral angles of the first
pleon-segment having a minutely produced point, those of the second more sharply
produced, those of the third slightly rounded.

Eyes moderately large in the adult male.

Upper antenne. Peduncle as in Urothoe marinus; flagellum six-jointed; secondary
flagellum slender, three-jointed, decidedly less than half as long as the principal.

Lower antenne. In the adult male peduncle very nearly as in Urothoe marinus, the
fourth joint, however, having sixteen or eighteen spines in one row, and two or three
in the other, the fifth joint carrying a row of nine calceoli on one edge, and four or
five rather small setee on the other; the flagellum between two and three times as long
as the peduncle, consisting of forty joints. In one specimen on one antenna each of
the first five joints, on the other each of the first three had a calceolus, the remaining
joints being alternately without and with these appendages. In the female the fourth
joint longer than the fifth, carrying fourteen spines in one row and two in the other,
with a few sete, the fifth joint having six spines and four of the long sete; the
flagellum full as long as the fifth joint of the peduncle, with one spine and four sete
on the first joint, the slender second joint being half as long as the first, the two sete
at its tip not very elongate.

Mandibiles. Cutting-plate like a broad tooth; the little secondary plate on the left
mandible (when unworn) cut into five little unequal teeth, on the right strap-
shaped; the third joint of the palp longer than the second, with five sete in a series
beginning low down on the front margin, and two accompanied by two spines at
the apex.

Lower lip. Mandibular processes well developed.

First maazille. Ynner plate slender, curved, with two plumose sete on the rounded
apex ; outer plate as in Urothoe pulchella; second joint of the palp about equal in
length and thickness to the first, fully reaching the apex of the outer plate, having
three sete on its distal margin.

Second maxille as in Urothoe pulchella, but with the plumose setz not fringing all
the distal half of the inner margin of the inner plate.

Mazillipeds differing but little from those of Urothoe pulchella; the outer plate with
five spines ; the second and third joints of the palp not so much distally widened.

First gnathopods. Side-plates distally widened, with two spinules at the lower hind
corner. The wrist with seven or eight spines in the row on the very sloping distal
margin; the hand narrowly oval.

GENERA UROTHOE AND UROTHOIDES. 15

Second gnathopods as well as the first, in general character, like those of Urothoe
marinus, but with narrower wrists and hands.

First and second perwopods as in Urothoe marinus, but comparatively slender, with
only four stout spines on the fourth joint, eight on the fifth, the finger with three or
four small and distant pointed tubercles. The side-plates of the first pair have five
spinules at the indented hind corner.

Third pereopods. The side-plates with the hind margin only slightly indented. The
limb strikingly distinguished from that of Urothoe pulchella by having the fourth joint
not broader than the third, with about seven spines in each of its spine-rows; the finger
is comparatively narrow, the distal half slenderly tapering, with about seven small
tubercles on the front margin.

Fourth pereopods differing little from those of Urothoe marinus and other species,
but having the third joint shorter than the fourth, with five or six plumose sete on the
hind margin, the fourth joint shorter or not longer than the fifth, with three groups of
spines in front and two behind, the finger very slender and tapering, with about seven
denticles on the front margin.

Fifth pereopods in general character like those of Urothoe marinus, but the first joint
not widened above, and with no conspicuous row of spinules on the upper part of the
hind margin; the third joint with two groups of spinules on each margin, the fourth
and fifth joints each with three groups on the front and two on the hind margin; the
finger very slender, more than half as long as the fifth joint, with some seven or eight
little nodules on the front margin. The female has the front margin of the first joint
convex instead of nearly straight.

Pleopods as in Urothoe marinus, but with fewer joints to the rami, the inner having
eleven or twelve, the outer thirteen or fourteen.

Uropods similar to those of Urothoe brevicornis, but the rami of the two first pairs
smooth, the peduncle of the second pair having only one spine on the inner margin ;
the inner ramus of the third pair is sometimes very decidedly shorter than the outer ;
the number of plumose sete on the rami of the third pair appears to be too variable to
afford a character.

Telson not very broad, nearly as in Urothoe brevicornis.

Length less than a fifth of an inch.

Localities. The specimens figured were dredged in February 1889 from a depth of
20 fathoms, off Fairlie Perch, in the Clyde, near Cumbrae, by Mr. David Robertson.
Other specimens examined were taken by Canon Norman in dredging among the
Shetland Isles. The original specimen, to which the name was given by Mr. Spence
Bate, came from the neighbourhood of the Eddystone Lighthouse.

16 REV. T. R. R. STEBBING ON THE

URoTHor MARINUS, Sp. Bate. (Plate II.)

1857. Sulcator marinus, Sp. Bate, Synopsis of Brit. Edr. Crust., Ann, & Mag. Nat. Hist. ser. 2,
vol. xix. p. 140.

1857. 5 % White, Popular History of British Crustacea, p. 175.

1862. Urothoé marinus, Sp. Bate, Brit. Mus. Catal. Amph. Crust. p. 115, pl. xix. fig. 2.

1862. Ss Bate & Westwood, British Sess. Crust. vol. i, p. 195, woodcuts.

1869. ae 3 ? var. pectinatus, Grube, Mitth. tiber St. Vaast-la-Hougue, Abh. der schles.
Gesellsch. fiir vaterl. Cultur, 1868-9, p. 119.

1869. i Norman, Last Report on Dredging among the Shetland Isles, p. 279.

1876. » marina, Giard, Comptes Rendus, Jan. 8, p. 76; Ann. & Mag. Nat. Hist. ser. 4,
vol. xvi. p. 261.

1876. » marinus, Stebbing, Ann. & Mag. Nat. Hist. ser. 4, vol. xvii. p. 344.

1877. » marina, Meinert, Crust. Isop. Amph. et Decap. Dani, p. 107.

1884. » marinus, Chevreux. Assoc. pour lav. des Sciences, Congrés de Blois, Amph. du
Croisic, p. 313.

1887. Ms Barrois, Morph. des Orchesties et liste Amph. du Boulonnais, p. 16.

1887. » marina, Bonnier, Catal. Crust. Malac. Concarneau, p. 79.

1887. oh 5 Chevreux, Crust. Amph. Bretagne, Bull. Soc. Zool. de France, t. xii. extr.
pp. 10, 34, 36.

1888. 53 45 Chevreux, Dragage de l’Hirondelle au large de Lorient, p. 1.

1888. Be 5 Chevreux, Amph. réc. aux env. de Cherchell, extr. p. 5.

1888. 5» marinus, Robertson, Catal. Amph. and Isop. of Clyde, p. 30.

The sides of the rostrum not forming an obtuse angle, but the apex not acute. ‘The
postero-lateral angles of the second pleon-segment acutely produced, but not those of
the first or the third segment.

Hyes very large in the adult male, nearly meeting on the top of the head.

Upper antenne. First joint thicker, but scarcely longer, than the second, each with
an elongate group of sete on the upper or outer side, and (in the adult male) a brush-
like fringe of hairs on the lower or inner; the third joint thinner than the second,
two thirds as long, with a few hairs; flagellum nine-jointed ; secondary flagellum, little
more than half as long as the principal, five-jointed.

Lower antenne. In the adult male first three joints short, gland-cone very inconspicuous,
the third joint having a tuft of hairs near the inner distal angle; the fourth joint longer
than the three preceding united, closely fringed on one side with a brush of hairs, which
also pass round the distal margin, on the other side carrying numerous unequal spines,
eighteen in one row, four in the other, and also some sete; the fifth joint nearly as
long as the fourth, carrying twelve sete on one edge, and eight calceoli, each with an
accompanying tuft of hairs, on the other; the flagellum very long and slender, with
fifty joints, each of the first six having a calceolus, and of the next forty-two each
alternate one, the calceoli being smaller towards the end of the flagellum. In a
specimen with no fringe of hairs and no calceoli the fifth joint of the peduncle has a

GENERA UROTHOE AND UROTHOIDES. nig

row of ten unequal spines, and the flagellum consists of twenty-five joints. In a female
specimen from the same locality the fourth joint has seventeen spines in one row and
three in the other; the fifth joint is decidedly shorter than the fourth, and has nine or
ten spines, the long setze of this and the preceding joint being fewer than in the female
specimen described of Urothoe brevicornis. The flagellum is as long as the last joint of
the peduncle, the first joint carrying two spines near the distal end, and eight long sete
round it; the little terminal joint is scarcely more than a third of the breadth of the
preceding.

Mandibles as in Urothoe elegans, but the long sete on the second joint of the palp
more numerous, and the long narrow third joint having ten spines along the front

margin, and two at the apex, together with four others that are long and more or
less setiform.

Lower lip. Mandibular processes strongly developed.

First maailie. The curved inner plate with four or five setee on and near the apex ;
outer plate and palp as in Urothoe elegans.

Second maxille with plumose sete passing round the apex and the distal half of the
inner margin of the inner plate, which also has a row of plumose spines on the apical
margin, these maxille not sensibly differing from those of Urothoe pulchella.

Masxillipeds nearly as in Urothoe pulchella, the inner plates rather broader, the outer
with six stout spines and twenty seta, thus arranged—a short seta, three rather large
ones, a group of three small ones, of which the middle is the longest, such a group
followed by a stout spine occurring four times in succession, then a single seta, two
stout spines, and (on the apical margin) three plumose sete. ‘The number of spines
and sete, however, cannot be depended on as constant in this species, or probably in
any of the others. Here the third joint of the palp is more elongate than in Urothoe
pulchella.

First gnathopods. Side-plates narrow, with a group of three spinules at the lower
hind corner. The first joint slender, with a few spinules on the front margin, many
long pectinate sete on the hind margin, and four groups of them on the inner surface ;
the second joint short, with one group of sete on the hind margin; the third joint
with one or two sete on the hind margin, its apex acute; the wrist almost as long as
the first joint and much broader, the convex hind margin or breast closely fringed with
unequal setiform spines, the inner surface carrying five or six groups, the row of
pectinate spines at the distal margin twelve in number; overlapping this row and
running to the front margin is a row of microscopic spinules, such a series being found
in all the species examined; the hand about three quarters as long as the wrist, much
narrower, with setiform spines in two rows on the inner surface, in one row on the
outer, and fringing more than the distal half of the hind margin; the palm oblique,
microscopically pectinate, fringed with various setules, and defined by a stout spine,

Om)

VoL. XII.—Part I. No, 3.—January, 1891. D

18 REY. T. R. R. STEBBING ON THE

beyond which the capped tip of the finger reaches; the finger carries two or three
setules.

Second gnathopods. Side-plates distally widened, the lower distal angle not having,
as in the first pair, a produced point, carrying a group of seven setules. ‘The sack-like
branchial vesicle as long as or longer than the first joint. The limb differs from that
of the first pair in having the first joint more elongate, with two instead of four groups
of setee on the inner surface; the long sete on the second and third joints are more
numerous; the wrist is narrower and without the special armature of the distal
margin ; the hand has less of the hind margin furnished with sete, and the palm a little
less oblique; the finger is decidedly shorter, its tip reaching very little beyond the
palmar spine.

First perwopods, Side-plates widened below, with six or seven spinules on the upward
curved hinder part of the lower margin. Branchial vesicle longer than the first joint,
nearly uniform in breadth, except at the point of attachment, First joint straight, with
setze at the apex of the hind margin, and near that of the front; second joint with a
group of sete about the apex of the hind margin; third joint as long as the fourth and
fifth joints together, having groups of sete along the hind margin, and near it a trans-
verse group high up on the outer surface, which also carries an oblique series near the
apex of the convex front margin; the fourth joint nearly as broad as the third, with an
oblique group of setee approaching the distal end of the front margin, several bent
plumose sete and four graduated stout spines along the undulating hind margin, one
stout spine and four set on the distal border, which projects strongly behind the fifth
joint; the fifth joint a little shorter than the preceding, and scarcely half its width,
carrying seven or eight unequal short spines directed towards the finger, the distal part
of the hind margin undulating; the finger straight, except at the tip, with five tubercles
along the hind margin, which, with the spines of the fourth and fifth joints, must give
it a prchensile character; the slightly bent tip has a transparent cap.

Second perwopods. Side-plates broader than in the preceding pair, with the spinules
not grouped at the corner, but spread along the lower margin. Branchial vesicle much
dilated below the long neck. Limb scarcely differing from that of first pair, except
that the first and third joints are longer, and the first is a little sinuous.

Third pereopods. Side-plates with the hind lobe deeper than the front, its margin
crenulate with spinules in the indents. Branchial vesicles narrowed at the two
extremities. The first joint very broad, distally widened, the wing often darkened
by the crowd of gland-cells, the upper corners rounded, the lower hinder angle not
very acute, sometimes rounded, the front margin rather sinuous, with two or three
spinules above, two groups of sete below, and a few spines at the apex, the hind
margin crenulate, and having about a dozen sete, the straight lower margin of the
wing carrying two or three setules; the second joint not half the width of the first, but
broader than long, with a group of sete and a spine at the apex of the front; the third

GENERA UROTHOE AND UROTHOIDES. UG.

joint rather longer, scarcely broader, with a group of sets and a spine at the indent of
the front margin, the apex of which has five spines: a row of seven spines arms the
lower margin of the outer surface and the rounded hinder apex, the corresponding
margin of the inner surface being fringed with some sixteen immensely long and
densely plumose sete, some of which reach to the extremity of the limb; there are
sometimes short plumose sete along the hind margin; the fourth joint is nearly as
long as the two preceding joints together, and decidedly, yet not very greatly, wider than
either; on the outer surface there is a series of eight unequal spines reaching the
indent of the front margin, and another reaching that of the hind margin; below these
there are two distal series, respectively of six and eight spines, and on the hinder part
of the distal margin there are also on the inner surface some long plumose sete; the
fifth joint is longer, but very much narrower than the fourth, having at the middle of
the hind margin a group of five unequal spines, and of three at the apex, on the
strongly indented front margin three groups of seven, six, and five respectively, and
on the inner surface below the middle a row of five long densely plumose sete; the
finger is not much shorter than the fifth joint, comparatively broad, except near the
cap-bearing tip, containing, like almost all other parts of the animal, numerous gland-
cells with their ducts and minute circular openings; the hind margin almost straight,
carrying near the base a small plumose cilium, the front margin at first smooth, but for
two thirds of its length armed with nodules, about thirteen in number, the first eight
or nine successively larger, the next one, two, or three small and sharper than the rest,
followed by one near the tip longer than any of the others.

Fourth perwopods. Side-plates narrower than in the preceding pair, the hind margin
crenate and fringed with spinules. ‘The branchial vesicle smaller than in the preceding
pair. The first joint large and oval, rather broader below than above, the front margin
a little sinuous, with two or three spinules above, and three or four groups of pectinate
setiform spines below; the convex hind margin very shallowly crenulate, with spinules
in the indents; the wing has numerous gland-cells, and sometimes as many as sixteen
long and very plumose sete, planted in a slight curve a great way from the margin on
the inner surface ; the short second joint has a group of pectinate spines at the apex of
the front margin; the third joint, which is more than twice as long as broad, has three
groups of spines along the front, nine long, much flattened, sete along the hind margin,
and a spine at its apex; the fourth joint is longer than the third, and has four groups
of rather stout spines along the front, and three mixed groups along the hind margin ;
the fifth joint is narrower than the fourth, scarcely longer than the third, with three
groups of stout spines along the front, one at the sharply produced apex of the hind
margin, and a little higher up a slender group; the finger is half the length of the
fifth joint, straight, very slender, with a plumose cilium on the hind margin near the
base, several little nodules along the front, of these that near the curved tip of the
finger being as usual the largest.

D2

20 REY. T. R. R. STEBBING ON THE

Fifth pereopods. Side-plates small, with the hind margin crenulate. The branchial
yesicle directed forward, not much longer than broad, very much shorter and narrower
than the first joint of the limb. The first joint broader than in the preceding pair and
as long, the widest part not far from the base, the broad wing containing numerous
gland-cells, its irregularly convex border shallowly indented and fringed with spinules,
of which, as in the preceding pair, those at the upper part are much larger than those
below; the front margin nearly straight or a little convex, with three or four groups of
spines; the short second joint having a group of spines at the apex of the front; the
third joint decidedly shorter than the fourth, with three groups of spines in front and
two or three behind; the fourth joint with four or five groups of spines on the front,
and three or four on the hind margin; the fifth joint subequal in length to the fourth,
but narrower, with three or four groups of spines in front and two behind; the finger
straight, slender, tapering, about half the length of the fifth joint, with seven or eight
little nodules along the front margin, and the usual plumose cilium near the base.
The whole limb is a little shorter than that of the fourth pair, the difference being
chiefly owing to the shorter third joint.

Pleopods. The peduncles of the first pair the longest, in all the pairs carrying one or
two groups of sete on the lower margin. ‘The coupling-spines (re¢/nacula’) long and
slender, with five or six teeth, besides the strongly bent apex; in general there are two
only of these organs on the peduncle, but in one instance four were counted ; there is
always a plumose spine with them; the cleft spines on the first joint of the inner ramus
two or three in number; the spoon-shaped arm much shorter than the serrate one; the
inner ramus sometimes with sixteen, the longer outer one with twenty joints.

Uropods. The stout peduncle of the first pair subequal in length to the outer ramus,
having on the outer margin one or two groups of setiform spines and a row of about
seven small spines, and a row of three on the inner margin; the rami are not very long,
smooth, the inner much smaller than the outer, both strongly curved, although in some
positions the curvature may not be very distinctly seen ; peduncles of the second pair
rather shorter and narrower than those of the first, similarly armed, but with rather
stronger spines; the rami subequal, the outer almost imperceptibly the longer, both
smooth, slightly curved, a little longer than the peduncle ; peduncles of the third pair
stouter than the preceding pairs, not longer than broad, having groups of spines about
the distal margin; the rami long and moderately broad, extending for nearly their whole
length beyond the telson and the other uropods, subequal, fringed with numerous long
plumose sete; in one specimen the outer ramus had on its outer margin twenty, two at
the apex of the little second joint, and eight on the inner margin; the inner ramus had

* In regard to the occurrence of the reténacula in various groups of Crustacea, and their phylogenetic import-
ance, see the interesting footnote in ‘ Carcinologische Mittheilungen,’ ix. p. 220, by Paul Mayer, 1880. By Carl
Bovallius they had been recognized in his account of “ Pterygocera arenaria.” Dr. Boas calls them in German
“ Hefthaken.” When I wrote my ‘Challenger’ report these notices had escaped my attention.

GENERA UROTHOE AND UROTHOIDES. 21

nine on the outer margin, two at the apex, and fourteen on the inner margin, In
another specimen, also a male, and from the same locality, there were only fifteen sete
on the outer and seven on the inner margin of the outer ramus, while in this specimen
the second joint was more elongate than in the other. Very little constancy is to be
expected in the number of these ornaments, which undoubtedly varies with the age of
the animal, as well as probably between individuals of the same age.

Telson. Cleft nearly to the base on the upper side and quite to the base on the under
side, the cleft gaping a little distally, the convex outer margins having each two cilia
near the centre, and a spinule not far from the apex, the distal margin of each half
oblique, carrying a feathered cilium, a short stout spine, and three sete.

Length of a good-sized specimen one third of an inch.

Localities. The specimen figured was taken at a low tide in February 1889, at
Cumbrae, in the Clyde, by Mr. David Robertson. Other specimens examined and
considered to belong to this species were obtained by M. E. Chevreux, of le Croisic,
Loire-Inférieure, off the French coast, and in Balta Sound, Shetland, by Canon Norman.
M. Chevreux in recording Urothoe marinus from “ les environs de Cherchell,” says, “le
genre Urothoe w était pas représenté jusqu'ici en Méditerranée,” but this is overlooking

Urothoe pulchella (Costa).

UrorHokr norvecica, Boeck. (Plate IV. B.)
1860. Urothoé norvegica, Boeck, Forhandl. ved de Skand. Naturf. 8de Méde, p. 647.

NEV ADS: op Boeck, Crust. Amph. bor. et arct. p. 58.

WSS) 55 60 Boeck, De Skand. og Arkt. Amph. p. 226, pl. vi. fig. 9, pl. vii. fig. 4.

NSS2i ees, fe Sars, Oversigt af Norges Crustaceer, p. 22.

WEI gp cs Chevreux, Crust. Amph. dragués par l’Hirondelle, Bull. Soc. Zool. de
France, extr. pp. 18, 15.

1887. e 39 Chevreux, Crust. Amph. Bretagne, Bull. Soc. Zool. de France, t. yi.
extr. p. 10.

1888. _,, ys Robertson, Catal. Amph. and Isop. of Clyde, p. 29.

Postero-lateral angles of the first three pleon-segments acute, those of the first two
only minutely produced.

Upper antenne. The first joint of the peduncle stout, the third joint decidedly
shorter than the first or second; the flagellum of five joints, together little longer than
the first or second joint of the peduncle; the secondary flagellum three-jointed, about
half the length of the principal.

Lower antenne. Fourth joint of the peduncle longer and much stouter than the fifth,
carrying a single row of about sixteen unequal stout spines near the convex margin, and
a row of sete apparently unmixed with stout spines; the fifth joint carrying five stout
spines and some sete; the flagellum consisting of eighteen naked joints, the structure
being indicative of a male not fully adult.

Lo
bo

REY. T. R. R. STEBBING ON THE

Mouth-organs as in Urothoe elegans. The palp of the first maaille consists of two
equal joints, and has three sete on the apex.

First gnathopods. The side-plates with a row of six sete on the truncate distal margin.
The limb nearly as in Urothoe elegans, the spine-row at the apex of the wrist consisting
of only three or four spines.

Second gnathopods nearly as in Urothoe elegans.

First and second pereopods scarcely to be distinguished from those of Urothoe elegans,
except that the inner margin of the finger, which in that species has very few nodules,
is here closely serrate, of the series of projecting points only the large one nearest the
tip of the finger being blunt enough to deserve the name of a nodule.

Third pereopods distinguished from those of Urothoe elegans by the much rounded
lower hind corner of the first joint, fewer plumose sete on the third and fourth joints,
and by the close serration of the distal half of the slender tapering finger.

Fourth pereopods. The front margin of the first joint has some spinules at four points
of the upper part, and at intervals on the lower part four long plumose sete; the third
joint is longer than the fourth, the plumose sete on the hind margin missing, but not
originally more than three or, at most, four in number; the finger slender, closely
serrate for two-thirds of its length.

Fifth percwopods nearly as in the female of Urothoe elegans, but here the third and
fifth joints are nearly equal in length, the fourth joint being decidedly longer than
either, with spines at four points of the front and at three of the hind margin.

Pleopods. The peduncles short; the cleft spines two or three; the inner ramus with
ten or eleven joints, the outer with thirteen to fifteen.

Uropods. Peduncles of the first pair a little longer than the straight slender rami,
the outer ramus with one spinule just above the middle, the inner equal in length,
smooth ; the peduncle of the second pair not longer than the rami, which are slender,
straight, subequal to one another, but shorter than those of the first pair; they are
smooth, or possibly there is a spinule on the outer ramus; the peduncles of the third
pair are shorter than the rami; the inner ramus is considerably shorter than the outer,
with a single seta near the extremity of its outer margin, five or six plumose sete on
the inner margin; the outer ramus has a spinule and seta at three points on the lower
half of each margin; the second joint is a third the length of the first.

Telson cleft nearly to the base so as to form two halves almost oval, but straight at
the base, and with the adjoining margins slightly compressed along the upper part ;
below the centre of each outer margin there are two cilia, and from an incision in
each rounded and rather narrow apex projects a moderately long spine, with a minute
cilium adjoining on the outer side.

Length about ene fifth of an inch or less.

Locality. The Shetland Isles, taken by the Rev. A. M. Norman in 1867.

GENERA UROTHOE AND UROTHOIDES, 23

UROTHOE BREVICORNIS, Sp. Bate. (Plate III.; Plate IV. C.)

1862. Urothoé brevicornis, Sp. Bate, Brit. Mus. Catal. Amph. Crust. p. 116, pl. xx. fig. 1.
W625 - 5; He Bate & Westwood, British Sess. Crust. vol. 1. p. 198, woodcuts.
1879. ,, marinus, Stebbing, Sess. Crust. Devon., Trans. Devon. Assoe, vol. xi. p. 519.

Rostrum forming an obtuse angle, but with a little acute point between the upper
antenne. Second pleon-segment with the plumose sete large and numerous, the
produced point of the postero-lateral angles minute.

Hyes very conspicuous in the male, approaching one another very closely.

Upper antenne nearly as in Urothoe marinus. In the adult male specimen the first
joint of the peduncle was decidedly longer than the second, and the principal flagellum
had seven joints, the accessory flagellum being more than half its length, with six joints,
of which the last was minute.

Lower antenne, In the adult male peduncle nearly as in Urothoe marinus, but in the
fourth joint the brush of hairs not passing round the distal margin; the spines twenty-
two in one row, five in the other, the setee more numerous; the fifth joint rather longer
than the fourth, with nine calceoli on one side, and eight long plumose sete on the
other; the flagellum not once and a half as long as the peduncle, having twenty-three
or twenty-four joints, with a calceolus to each of the first three or four, and then on
alternate joints for some distance along. In the female the fourth joint is longer than
the fifth, and has twenty-five spines in one row, seven in the other, with numerous long
serrate setee near the straight inner border, and a group of about a dozen near the
produced distal corner of the convex side; the fifth joint has thirteen spines along the
convex border, with a group of eight long setie near its distal end, and along the other
border some fifteen long sete; the two-jointed flagellum is shorter than the last joint
of the peduncle ; the first joint has ten long sete round the distal half, and three spines
on one margin; the little second joint is about a quarter the length and not half the
breadth of the first, and is tipped with two slender sete several times its own length.
Both in male and female the fourth and fifth joints of the peduncle have some little
stiff setze or cilia, which are strongly plumose, and appear as if twisted at the centre.

Upper lip with a smoothly rounded margin to the principal plate, except that in the
centre a little space is marked off by a small notch on either side, the tract so marked
off being distally smooth, but furry on the sides. The inner plate appears to have a
quite smooth margin not reaching quite to the distal margin of the outer plate. ‘The
structure of this lip seems to be tolerably uniform in all the species.

Mandibles as in Urothoe marinus, except that the third joint of the palp has nearly
two thirds of the inner margin clear of spines.

Lower lip as in Urothoe marinus.

First maaille. The inner plate with three plumose sete on the apex ; outer plate as
in the other species ; palp with the first joint rather longer than the second, the second
with the usual three sete on the apex.

24 REY. T. R. R. STEBBING ON THE

Second maxille. The plumose sete occupying two thirds of the inner margin of the
inner plate.

Mazillipeds, probably not to be distinguished from those of Urothoe marinus by any
constant character; thus in one specimen one of the outer plates had six, and the other
seven, stout spines, while in another specimen each of these plates had eight stout
spines.

First and second gnathopods not distinguishable from those of Urothoe marinus. In
a male specimen from North Wales there were eight spines in the distal spine-row of
the wrist, in a female from the same locality ten, and in another female from South
Devon fourteen.

First and second pereopods scarcely, if at all, distinguishable from those of Urothoe
marinus. In the specimens examined the fourth joint had six spines instead of five,
and the fifth joint eight spines instead of seven, while the tubercles on the finger were
fewer and less pronounced.

Third perwopods closely resembling those of Urothoe marinus, but the large first joint
with fewer indents on the hind margin, and its apex rather more acute; the third joint
has nine or ten spines in each group of its distal margin; the fourth joint ten or eleven
in each of its four groups; the fifth joint also has larger groups of spines, and is rather
more widened, whereas the distal narrowing of the finger is much less abrupt than in
the species compared, and the marginal nodules are smaller.

Fourth perwopods nearly resembling those of Urothoe marinus, but the third joint is
here rather longer than the fourth, having almost the whole of the hind margin fringed
with the long plumose sete.

Fifth perwopods closely resembling those of Urothoe marinus, but with the first joint
of the limb not wider above than below. In the female the first joint appears to have
the front margin more convex than in the male.

Pleopods as in Urothoe marinus, but with no more than two cleft spines observed on
the ramus of any specimen.

Uropods. Peduncle of the first pair much more elongate than in Urothoe marinus,
with two groups of setiform spines and five stouter spines on the outer and four spines
on the inner margin; the rami slender, subequal in length to one another and to the
peduncle, nearly straight, reaching beyond the second pair, the outer having three
spines upon it, the inner having a little seta on its inner margin; the peduncle of the
second pair much shorter than that of the preceding pair, shorter than the rami, with
four spines on the outer and two on the inner margin; the rami subequal, shorter than
the preceding pair, straight, the outer with two spines, the inner with a little seta; the
peduncle of the third pair rather longer than broad, but otherwise this pair is scarcely,
if at all, to be distinguished from the corresponding pair in Urothoe marinus.

Telson differing very little from that of Urothoe marinus, the convex margins unbroken

GENERA UROTHOE AND UROTHOIDES.

bo
Or

by the insertion of a spinule, the distal margin less broad, carrying a feathered cilium,
a spine, and one or two setules.

Length. A quarter to a third of an inch.

Localities. 'The specimen figured was taken with others of both sexes at Llanfairfechan
in North Wales, from the banks of little streams or pools left in the sands at low tide.
The species has also been obtained at Goodrington, near Torquay, from the sand left
bare at low tide. The specimen to which Spence Bate gave the name drevicornis was
taken at Tenby, a locality intermediate between the two just named, and presents such
differences from the description of the adult here given as might be expected in an
example only a tenth of an inch long. Recently the species has been met with in
North Devon.

[URorHor Poucueti, Chevreux.

1888. Urothoe Poucheti, Chevreux, Crust. Amph. du littoral des Acores, Bull. Soc. Zool. de France,
t. xl. janvier 1888, p. 34.

When this paper was sent to the Zoological Society, the type specimen of Urothoe
poucheti, which is at present unique, was being exhibited in the Pavillon de Monaco at
the Paris Exhibition. It has since been lent me through the kind instrumentality of
M. Chevreux, naturally, however, without being available for dissection, which is almost
essential for a full and accurate account of a species in this genus. Under this restric-
tion the following brief notes were made upon it :—

Eyes strikingly large and black, the specimen being a male.

First gnathopods. The finger longer than in the second pair.

Third pereopods. The lower hinder angle of the first joint well rounded. The third
joint not broader than the second, and not strongly spined, but with long plumose
sete; the fourth joint not broader than the third; the finger long and slender, the
distal half serrate.

Fourth perwopods. The first joint having slender spines on the lower part of the
front margin, and seven plumose sete within the wing; the third joint longer than the
fourth, instead of shorter as in Urothoe elegans.

Fifth perwopods. Vront margin of the first joint straight.

Uropods. The first pair with the rami equal, slender, not quite so long as the
peduncle; the second pair considerably smaller than the first, the rami equal, slender,
shorter than the peduncle; the third pair with long rami reaching back beyond those
of the first pair.

The postero-lateral corners of the third pleon-segment are slightly rounded, almost
rectangular.

The species approaches Urothoe elegans, Spence Bate, and in regard to the third
pereopods it much resembles Urothoe norvegica, Boeck ; but taking all the characters

VOL. X1I.—Part I. No. 4.—January, 1891. 10

26 REV. T. R. R. STEBBING ON THE

together (see page 9), it must be regarded as distinct from all the earlier known
members of the genus.—T. R. R. S., Sept. 1890.]

UROTHOIDES, nov. gen.
1888. Urothoe, Stebbing, ‘Challenger’ Amphipoda, Zool. Reports, vol. xxix. p. 824.

Nearly resembling Urothoe, Dana, in regard to the antenn, mouth-organs, gnathopods,
first and second persopods, and the pleon.

Third and fourth perwopods having the first, third, and fourth joints much expanded,
the third joint more widely than the fourth ; these limbs not armed with long plumose
setae as in Urothoe.

Fifth perwopods having the much expanded first joint strongly produced downwards
behind, and with a strongly serrate hind margin.

The fingers of the perwopods not nodulous on the inner margin.

The name is derived from Urothoe, a closely related genus, and eidoe, likeness.

UROTHOIDES LACHNERSSA.

1888. Urothoé lachneéssa, Stebbing, ‘ Challenger’ Amphipoda, Zool. Reports, vol. xxix. p. 825,
pl. lvu.

It is with some hesitation that I now propose a new genus for this recently published
species. When originally including it in the genus Urothoe I was not aware how
singularly compact a group the existing species of that genus formed, and how
intimately connected with one another they were in many minute details.

Since the ‘ Challenger’ Report was published I have examined an additional specimen
of this species, in which the upper antennz proved to be abnormal. ‘The entire peduncle
is stout, the second joint not longer than broad, the third longer than either the first or
second, conically produced along two thirds of the first joint of the secondary flagellum.
The principal flagellum consists of four joints, the secondary flagellum of three, the first
of which is as long as the other two together, and the three together are as long as the
principal flagellum. It may be assumed that the malformation results from a coalescence
of the third joint of the peduncle with the first one or two joints of the principal
flagellum, or it might be more correct to say that the articulations have not been
developed so as to produce the usual distinction of these joints.

The second segment of the pleon is not armed with long plumose sete as in the
genus Urothoe, and the rami of the third uropods are also devoid of these ornaments.
In the species of Urothoe the outer of these rami is perhaps invariably longer than the
inner, but sometimes the difference is scarcely perceptible, whereas in the present
species the difference between the two rami is very great.

GENERA UROTHOE AND UROTHOIDES. 27

SUMMARY AND INDEX.

The history of the genus may be shown in brief by the following list :—

Pages
1852. Urothoe rostratus, Dana: since called Phowocephalus rostratus. . . . . . . . 8,4
TRE ORONO CARO EHO | Dany Wess a. lob Ciol oo en Guat yds Gunes tou nso Jew ce Zid)
1853. Hgidia pulchella, Costa: since called Urothoe pulchella. . . . . ... . . 411
1856. Gammarus elegans, Bate: __,, i WRAL AGTS 55 2 6 oo 5 0 BRB) Op 8
1857. Sulcator marinus, Bate: i oy WROD TRERUOUS) <3 H IIO
1860. Urothoe norvegica, Boeck . . . . . . ww, 6, 7,9, 21
1862. Urothoe bairdii, Bate: since called Urothoe marinus .

ou
—

SCS
tw)
cn

1862. Urothoe brevicornis, Bate . . . . .. .
1869. Urothoe marinus, Bate, ? var. pectinatus, Grube

: 8
1874. Urothoe, species, 8. 1. Smith: since called Harpinia, species . 4,
1879. Urothoe abbreviata, G. O. Sars PCR Vay Ment Lars iene nes 3,9
1880. Urothoe pinguis, Haswell : since called Harpinia(?) pinguis . A.
1888. Urothoe poucheti, Chevreux . EH aw MEA, hoe Myo teesr me Pe 9, 25
1888. Urothoe lachneéssa, Stebbing: since called Urothoides lachneéssa 4, 26

The genus Urothoe is therefore at present composed of the following eight species :—

Urothoe abbreviata, Sars. Urothoe marinus, Bate.
» Orevicornis, Bate. » norvegica, Boeck.
5, elegans, Bate. 5, poucheti, Chevreux.
» trrostratus, Dana. », pulchella (Costa).

Should any reader object to finding some of the specific names masculine and others
feminine, he is respectfully reminded that among the specimens examined some were
males and others females.

Of the eight species enumerated, it must be observed that they are more remarkable
for their likeness to one another than for any differences that can be discerned. The
magnitude of the eyes and the structure of the lower antenne vary greatly with the age
and sex of the animal, the most constant feature being that the lower antenne in the
female have a two-jointed flagellum. Among the details that appear to prevail
throughout the genus may be noticed the vast number of gland-celis over all parts
of the body, the transparent caps to the tips of all the fingers, a peculiar spine-row on
the wrist of the first gnathopods, and the long plumose seta on the third, fourth, and
fifth joints of the third pereopods, on the first joint of the fourth peropods, and on
the second segment of the pleon. The species for the most part are to be distinguished
from one another only by groups of small differences. Among these, however, a single
feature may here and there make itself moderately conspicuous: thus, only Urothoe

abbreviata, Sars, is said to be blind; only Urothoe elegans, Sp. Bate, is described as
E 2

28 REV. T. R. R. STEBBING ON THE

ornamented with rose-coloured markings’. Urothoe pulchella (Costa) has the fourth
joint of the third pereopods wider than in any other species. Urothoe marinus, Sp.
Bate, alone has the rami of the first and second uropods strongly curved. Urothoe
poucheti, Chevreux, appears to be distinguished from all other species by the greater
length and stronger armature of the first uropods. Urothoe norvegica, Boeck, shows no
very salient difference from Urothoe elegans, unless it may do so in colouring. Urothoe
brevicornis, Sp. Bate, makes a near approach to Urothoe marinus, except in regard to
its longer and straighter uropods. Dana’s Urothoe irrostratus is the only species at
present known from the Pacific. As Spence Bate has pointed out, it makes a near
approach to Urothoe elegans; but as the figures and description are incomplete, it is not
at present possible to decide whether it is identical with any European species or
otherwise. ‘The exact position of Grube’s Urothoe marinus, var. pectinatus, is also
doubtful.

EXPLANATION OF THE PLATES.

PLATE I.

Urothoe elegans.

The full figure is given in lateral view, the three lines above it indicating the
natural size of a female, a male, and a young specimen, respectively.

a.s. Upper antenna of the male; @.s, v, of the young.
ai. Lower antenna of the male; a.i,@?, of the female; a7, v, of the
young.
12,9. Lower lip of the female.
gn. 1, First gnathopod of the male.
gn. 2. Second gnathopod of the male,
prp. 1, 2, 8, 4, 5. The first, second, third, fourth, and fifth pereopods respectively, of
the male.
prp. 2, %. Second pereeopod of the young.
prp. 5,%. Fifth pereeopod of a female; prp. 5, v, of the young.
ur. 1, 2, 3. The first, second, and third uropods respectively, of the male.
ur. 2, ur. 8, 2. Second and third uropods of the female.
t. Telson of the male.

1 Yet such markings are still visible in a specimen from Balta Sound, Shetlands, taken by the Rev. A. M.
Norman in 1860, and labelled (probably for that very reason) as Urothoe elegans, though otherwise it does not
seem to be distinguishable from Urothoe marinus. Hence it is doubtful whether reliance can be placed upon
colouring as a distinguishing mark of Urothoe elegans.

GENERA UROTHOE AND UROTHOIDES. 29

PLATE iT:

Urothoe marinus.
The full figure is given in lateral view, with a line above it indicating the natural size.
too) too)

a.s. Upper antenna.
a.t. Lower antenna; only a part of the flagellum drawn.
m.m. The pair of mandibles,
ma. 1. First maxilla.
ma. 2. Second maxilla.
mxp. One half of the maxillipeds, and on a larger scale the inner and outer
plates of the other half.
gu. 1. First gnathopod.
gn. 2. Second gnathopod.
prp. 1, 2, 3, 4, 5. The first, second, third, fourth, and fifth pereeopods respectively ; the
figure prp. 3 to the left showing the outer side, and the figure
prp. 3 to the right showing the inner side of the limb.
plp, sp. Coupling-spines and a cleft spine of the pleopods.
ur. 1, 2, 3. First, second, and third uropods respectively.
t. Telson.

PLATE III.

Urothoe brevicornis.

The full figure is given in lateral view from a small specimen, the line above
it to the right indicating the natural size, the line above it to the left showing the
length of a female specimen. The figures of the separate parts were all taken from
the male.

a.s. Upper antenna. mx. 1, First maxilla.
a.i. Lower antenna. gn. 1, First gnathopod.
m. Left mandible. gn. 2. Second gnathopod.

1c. Lower lip.
prp. 1, 2, 3, 4, 5. First, second, third, fourth, and fifth perseopods respectively, both
sides of the third pereeopod being shown.
Pi.s. 2, Pl.s. 3. The lower portions respectively of the second and third segments of
the pleon.
ur. 1, 2, 8. First, second, and third uropods respectively.
t. Telson.

30

ur.

or. tr.

ON THE GENERA UROTHOE AND UROTHOIDES.

PLATE IV.
A. Urothoe pulchella, ¢ .

Triturating organs 77 situ in the stomach.
qn. 1. First gnathopod. prp. 3. Third pereopod.
gn. 2. Second gnathopod. plp, A pair of pleopods.

. First, second, and third uropods respectively.

t. Telson upturned, and consequently viewed from the underside.

I Py Bs
. Telson.

B. Urothoe norvegica.

gn. 1. First gnathopod. prp. 3. Third pereeopod.
gn. 2. Second gnathopod. plp. A pair of pleopods.
First, second, and third uropods respectively.

C. Urothoe brevicornis, 2 .

a.t. Lower antenna. i.s. Upper lip.

. First, second, and third uropods respectively.

tT. Telson.

JTRennie Reid .Lith.Edin’

Del. T.R.R.Stebbing.

UROTHOE ELEGANS. SP. BATE.

iy diay

Fras, Lo0€. foe. YoU MU FEM

Zz MM
init MUG
tl Mt i

ZA CW fel
oe SE ee

ur.3
Del. T.RRSte bbing. J.T.Rennie Reid .Lith Edin®™

UROTHOE MARINUS. SP. BATE.

Trans, Lovk, doc. Vok, MU He IT

Del. TRR.Stebbing. J.T. Rennie Reid LithEdin?

UROTHOE BREVICORNIS. SP BATE.

7 ay * We

mtn

ae) LOL, voc. Yoh SUSE LY

Cc

‘ 5 c ea
LS
La Ne Ba \
(Mii

KOS

! e ‘

Del TRR.Stebbinp. ; J.T.Rennie Reid Lith. Edin?

UROTHOE PULCHELLA ?(COSTA). UROTHOE NORVEGICA,BOECK. UROTHOE BREVICORNIS ¥SP BATE.

yew

oa

Il. On four new British Amphipoda. By the Rev. Tuomas R. R. Srespine, 1.A., and
Davip Rosertson, .L.8., GS.

Received September 19th, 1889, read November 19th, 1889.
[Puates V., VI.]

1. SOPHROSYNE ROBERTSONI, n. sp. (Plate V. a.)

Rostrum minute; the body moderately compressed, with rounded back ; the hinder
angles of the first pleon-segment slightly rounded, those of the second squared, those of
the third strongly produced upwards, so that a deep cavity is formed on either side
between these and the arched dorsal surface of the segment ; the much shallower fourth
segment forms dorsally three little humps, two median and one distal; the fifth and
sixth segments are very small. The integument is closely speckled.

Eyes not perceived.

Upper antenne. First joint shorter than the head, once and a half as long as broad,
as long as the six-jointed flagellum, the second joint not half as long as the first, and
about twice as long as the third; the flagellum tapering, the first joint narrower but a
little longer than the last of the peduncle, each joint with one or two filaments; the
secondary flagellum slender, three-jointed, the last joint minute.

Lower antenne. Gland-cone very prominent; third joint longer than broad, fourth
joint longer than the fifth; the flagellum eight-jointed, slender, tapering, the first six
joints together being about as long as the fourth joint of the peduncle.

Mandibles. The trunk slender, the cutting-edge narrow, apparently with a denticle
at either end; molar tubercle wanting; palp set far forward, much longer than the
trunk, first joint short, second elongate, slightly curved near the base, with one spine
near the distal end, the third joint as long as the second, distally truncate, with four
unequal spines on the extremity and one a little below it.

First maxille. The inner plate small and low down; the outer plate forming a single
apical tooth, with a little denticle on its side; the palp was broken in dissection.

Second maxille. ‘The inner plate much shorter and narrower than the outer, with
three minute setules near or at the rounded apex, the much larger outer plate having
four setules similarly situated.

Maxillipeds, The inner plate very small, not reaching to the base of the outer plate,
carrying a single spinule on the apex; one of the pair of plates was in the specimen
much smaller than the other; the outer plate longer but narrower than the first joint
of the peduncle, having two or three spinules on the inner margin, and four somewhat
larger on the apex; the first joint of the peduncle rather longer than broad, the second
much longer, with two groups of spines on the inner margin and one at the outer apex ;

32 REV. T. R. R. STEBBING AND MR. D. ROBERTSON ON

the third joint rather shorter and narrower than the second, with several spines about
the distal two thirds; the curved finger nearly as long as the third joint, with the dorsal
setule not very near the base.

Triturating organ. One margin fringed with ten graduated spines, which are short
and moderately stout, not serrate.

First gnathopods. The side-plates broad above, so as to project much beyond the
upper part of the segment. The first joint broad, not very long, deeply channelled,
with some spinules on the front margins, and a group of spines at the hinder apex ; the
second joint as broad as long, channelled, with spines along the hind margin; the third
joint much longer than broad, with about a dozen spines and half a dozen sete along
the serrate hind margin, which is pointed at the apex; the wrist triangular, distally
broader than the length, the front margin convex, with an apical spine, the hinder apex
having a graduated series of five spines with several sete; the hand much longer than
the wrist and broad, but much longer than broad, with the front margin convex, the
hinder concave, armed with two spines, and ending in a long slender tooth, the palm
smooth, sinuous, the surface of the hand and the apex of the hind margin having a few
setee or spines; the finger closely fits the palm, except that its tip overlaps the apical
tooth. One of this pair of gnathopods was in the specimen rather smaller than the
other.

Second gnathopods. Side-plates longer but narrower than those of the first pair.
Branchial vesicles broader but a little shorter than the side-plates, apically almost
pointed. Marsupial plates very narrow, with three sete at the apex. First joint of
the limb elongate, narrow, a little widened distally ; second joint longer than the third
or hand; the third joint having much of the convex hind margin furred with setules ;
the wrist narrow, elongate, about half the length of the first joint, the front and hind
margins furred with setules; the hand less than two thirds the length of the wrist, the
margins nearly parallel, except near the base, both furred with setules, the hind margin
distally carrying scale-like spinules, the apical margin having the usual spines; the
finger minute, broad at the base, apically narrow and hooked, the dorsal setule median.

First pereopods. The side-plates like the preceding pair, but larger. The first joint
not reaching below the side-plate, with setules along the front margin, and a spine or
two at the hinder apex; the second joint is armed at three points of the hind margin ;
the third joint at eight points of the hind margin and the produced apex of the front;
the fourth joint, much narrower and shorter than the third, has sete at four points of
the hind margin ; the fifth joint is longer but narrower than the fourth, with sete at two
or three points; the finger is slightly curved, a little shorter than the fifth joint.

Second perwopods. ‘The side-plates excavate behind to more than half the depth
of the plate. The limb similar to the preceding pair, but with the third joint a
little shorter.

Third pereopods. Side-plates of equal breadth and length, the front margin convex,

NEW BRITISH AMPHIPODA. 33

the hinder nearly straight. The first joint a little longer than the side-plates, but not
so broad, pear-shaped, with spines at eleven points of the convex front margin, and
with a little serration at the top of the hind margin; the second joint almost embedded
in the first; the third much shorter than in the two preceding pairs, with spines at two
points of the hinder and four of the front margin; the fourth joint shorter and narrower
than the third, with spines at three points of the front margin; the fifth joint narrower
but longer than the fourth, with spines at two or three points of the front; the finger
slightly curved, nearly as long as the fifth joint.

Fourth perwopods. The side-plates narrower and a little shorter than the preceding
pair, more strongly bilobed below. ‘he first joint much longer than in the preceding
pair, elongate oval, the front margin with spines at eleven points, the hind margin
slightly serrate; the second joint rather longer than broad; the third joint similar to
that in the preceding pair, but larger; the fourth nearly as long, with spines at four
points in front; the fifth narrower, but as long, with spines at four points; the finger
a little shorter than the fifth joint.

Fifth pereopods. Side-plates much smaller than the preceding pair, not bilobed, the
hind margin very convex, the front straight. The first joint a little longer than broad,
as long as in the preceding pair and much wider, almost all the front margin fringed
with spines, the hind margin roughly serrate; the other joints nearly as in the third
pair, but the third and fourth joints smaller, and the finger as long as the smooth
fifth joint.

Pleopods. Vhe coupling-spines small, two in number, with two lateral teeth near the
reflexed apex ; the cleft spines two in number, the arms nearly equal; the joints of the
inner ramus seven or eight, those of the outer eight or nine in number.

Uropods. The peduncles of the first pair longer than the rami, with several spines on
the margins, the rami slender, tapering, subequal, the outer carrying three spines, the
inner a single one; the peduncles of the second pair shorter than the rami, which are
nearly as long as those of the first pair, the outer a little the longer, armed with two
spines, the inner being unarmed; the peduncles of the third pair scarcely longer than
broad, the rami smooth, slender, subequal, smaller than those of the preceding pair.

Telson flanked for more than half its length by the produced sides of the sixth
segment, divided for more than a third of the length, the slightly indented apices not
quite reaching the distal end of the peduncles of the third urepods.

Length one quarter of an inch.

Locality. The Clyde. One specimen, female, containing eight or nine rather large
eggs in a forward state of development.

From Sophrosyne murray, taken at Kerguelen Island, the new species’ differs in

1 It is proper to mention that Mr. David Robertson, my colleague in this paper, by whom this and the other
specimens described were captured, only assented to the adoption of the specific name robertsont at my particular
request.—T. R. R. 8.

VoL. X11.—PartT 1. No. 5.—January, 1891. F

34 REY. T. R. R. STEBBING AND MR. D. ROBERTSON ON

several small details; the second joint of the upper antenne is less elongate, the
armature of the first gnathopods is slighter, the hand of the second gnathopods is
differently shaped, the third segment of the pleon is dorsally curved downwards instead
of being posteriorly squared, the telson is longer than broad and less deeply cleft. In
general features the two species are remarkably alike, and now that this rather striking
genus is found to have a representative in one of our own estuaries, it seems singular
that it should have first been made known to us from Antarctic waters.

2. SYRRHOH FIMBRIATUS, n. sp. (Plate V. B.)

The third segment of the pleon with the lower hinder angles produced into an acute
upturned point, the two preceding segments having these angles acute, but only
slightly produced. ‘The dorsal denticles apparently present on some or all of these
segments were not clearly made out. The sixth segment of the pleon fringed behind
with a close-set row of spinules.

Upper antenne. First joint stout, longer and much broader than the second, which
is nearly twice as long as the third; the flagellum longer than the peduncle, the first
joint as long as the first of the peduncle, and as long as the six following joints united,
armed with a brush of long filaments; the secondary flagellum three-jointed, the small
third joint not quite reaching the end of the second joint of the principal flagellum.

Lower antenne. The first joint broader than long, overlapping the little acute gland-
cone of the second joint, the third joint a little longer than broad, the fourth longer
than the three preceding united, rather shorter than the fifth, each of these two carrying
tufts of setules; the flagellum very slender, longer than the peduncle, consisting
probably of ten or twelve joints, of which eight were present.

Upper lip. The central part of the distal margin forms an almost semicircular lobe,
which in the dissection was folded back, but whether this may be its natural position
or only accidental, could not be determined.

Mandibles. Cutting-edge having two teeth and a smooth border on the left mandible,
on the right having only the two teeth; the secondary plate small with four teeth,
which are blunt on the left, and delicately sharp on the right mandible; the molar
tubercle broad and strong; the first joint of the palp longer than broad, the second
joint elongate, slightly curved, with four groups of sete, the third joint broken,
probably very short.

First maxille. The palp has on the apex five spines and one spine on the outer
margin below the apex.

Second maaille. The inner plate appears to be broader than the outer, both plates
carrying numerous long spines.

Maxillipeds. Inner plates broad, reaching a little beyond the first joint of the palp,
the inner margin produced into a small distal tooth, the distal part of the plate
bordered with nine plumose sete; the outer plates not quite reaching the end of the

NEW BRITISH AMPHIPODA. 39

second joint of the palp, fringed with eight or nine spine-teeth; the second joint of
the palp the longest, not very strongly spined.

First gnathopods. The side-plates, like the somewhat larger second pair, oblong,
directed forwards. ‘The first joint of the limb a little widened distally, carrying a few
slender spines; the second joint having some slender spines at the apex of the hind
margin; the third joint more or less oblong, with five spines on the lower part of the
hind margin; the wrist not twice as long as the hand, furry on both margins, the hind
margin being also fringed with spines; the hand broadest near the base, the front
regularly convex, with spines at the apex, the hind margin convexly angled, carrying a
few partially pectinate spines, the surface of the hand furry; the finger about half the
length of the hand, broadest at the base, with spines at the centre, where at the inner
margin it abruptly narrows.

Second gnathopods. The first joint longer than in the first pair, with a few feathered
spines or sete on the hind margin ; the wrist and hand longer than in the first pair;
the wrist twice as long as the hand, slender, furry, and carrying a few spines; the hand
also slender, furry, with an oblique row of spines on the distal part; the finger much
curved, about a third the length of the hand.

First perwopods. Side-plates longer than the preceding pair, distally widened.
Branchial vesicles about equal in length and breadth to the side-plates, but with
curved margins. The limb slender, the first joint reaching a little below the side-
plate, the third joint with convex front margin, shorter than the fourth, which is
straight and armed on the hind margin with a few spinules; the fifth joint rather
longer than the third, shorter than the fourth, with spinules at four points of the hind
margin; the finger curved, about half the length of the preceding joint.

Second perwopods. The side-plates with convex front and lower margin, the hinder
excavate. Branchial vesicles and limb as in the preceding pair, but the fourth joint
more elongate.

Third perwopods. The side-plates longer than deep, only slightly bilobed. The first
joint narrowly oval, slightly serrate, the third wider but shorter than the fourth, the
fourth than the fifth, each of the three having small spines at three or four points of
each margin; the finger slender, nearly straight, more than half the length of the
fifth joint.

Fourth perwopods. The side-plates with the length and depth equal. The lmb
similar to the preceding pair, but with the third, fourth, and fifth joints much longer ;
and the finger about half the length of the preceding joint.

Fifth pereopods. The side-plates very shallow, longer than deep. The broadly ovat
first joint longer and much wider than in the two preceding pairs, the third joint larger
than in the preceding pair and more produced downwards behind, the remainder of the
limb similar to that of the fourth pair.

Pleopods. The two coupling-spines long and slender, serrate on each side, with two,

F2

36 REY. T. R. R. STEBBING AND MR. D. ROBERTSON ON

three, or four teeth near the apex ; the cleft spines three in number ; the joints of the
inner ramus nine, of the outer ten.

Uropods. In the first pair the peduncle is scarcely so long as the longer ramus, which
is straight, almost smooth, except for the group of spines on the truncate apex; the
smaller ramus is similar to the larger, but not quite half as long; in the second pair
the peduncle is longer than the short ramus, and a little more than half the length of
the longer one, which is acutely lanceolate, carrying six spinules on one border and
three on the other; the smaller ramus is less than half as long or as broad as its
companion, but a little longer than the similar ramus of the first pair; in the third
pair the peduncle is shorter than the rami, more than twice as long as broad, the rami
acute, the inner rather the longer, with six or seven plumose sete on the inner margin,
the outer ramus having five spines and one plumose seta on the inner margin, nearly a
third of the length of the ramus forming a second joint.

Telson elongate, cleft for more than half its length, reaching considerably beyond the
peduncles of the third uropods, the acute apices having each a setule in a notch on the
inner side; there are also three or four setules arranged near each lateral margin.

Length. This small and delicate species was dissected before the measurement of the
body had been taken.

Locality. Clyde.

Remarks. The specific name alludes to the distinguishing fringe of spinules upon
the sixth segment of the pleon.

3. PODOCEROPSIS PALMATUS, n. sp. (Plate VI. a.)

Lateral angles of the head rather acutely produced between the upper and lower
antenne.

yes nearly round, situated on the lateral lobes of the head, in the preserved
specimen showing a light rim round a black centre.

Upper antenne. The first joint twice as long as broad, channelled ; second joint once
and a half as long as the first, much narrower, slightly curved, with sete at four points
of the lower margin; third joint a little longer than the first, with sete at three points ;
flagellum of five or more rather elongate joints carrying filaments; the secondary
flagellum very slender, two-jointed, shorter than the first joint of the principal flagellum,
the second joint minute, tipped with setules.

Lower antenne. The third joint as long as the first and second united, distally a
little widened. The remainder of the antenne missing.

Mandibles. Cutting-edge divided into five unequal teeth; the secondary plate on the
left mandible has four teeth, that on the right about six denticles; the spine-row on
the left consists of five spines, denticulate between the widest part and the apex; on the
right mandible there are only four spines; the molar tubercle has an almost circular
denticulate crown with a seta at the hind corner; the palp is much longer than the

NEW BRITISH AMPHIPODA. 37

trunk, set far forward, the first joint rather longer than broad, the second about three
times as long as broad, with several setee along the front, and one or two on the hind
margin; the rather shorter third joint has four or five sete or spines near the hind
margin, and several on the truncate apex and distal part of the front margin.

first maxille. The long second joint of the palp has four spine-teeth on the
truncate apex.

Mazxillipeds. Ynner plates not reaching the middle of the outer, and these not
reaching the middle of the second joint of the palp, the spine-teeth on the inner
margin passing into curved spines on the apical border, the series about nine in
number; the long second joint of the palp fringed, but not closely, with sete; the
third joint scarcely so long as the first, the fourth much shorter; the blunt apex tipped
with spines.

First gnathopods. Side-plates distally widened. First joint slightly curved, reaching
beyond the side-plates, the third joint almost oblong, with spines on the convex front
and along the apical margin; the wrist fully as long as the hand, not very much shorter
than the first joint, with slender spines on the surface, round the distal border, and in
two groups on the hind margin; the convex front border of the hand carrying four
groups of very slender spines, of which the much shorter hind margin has two groups,
the deeply excavate palm being bordered with spinules and a group of spines; the
finger slender, curved, reaching beyond the extremity of the palm and its palmar
spine, and having some minute setules and microscopic furring along its inner margin,
which has also a very small tooth far from the apex.

Second gnathopods. The side-plates larger and of more uniform width than the first
pair. ‘The first joint reaching beyond the side-plates, not longer aud very much
narrower than the hand, the third joint rather longer than in the first pair, but with
fewer spines; the wrist triangular, cup-shaped, about half the length of the hand,
which is much wider, broadly oblong, with several groups of slender spines along
the hind margin, and one or two on the smoothly convex palm, which is left partly
exposed by the slender curved finger, as this bends almost abruptly from the hinge on
to the inner surface of the hand, its acute tip reaching beyond the middle of that
surface, the concave margin having setules at three or four points. The general
resemblance of this limb to the corresponding one in Melita palmata (Montagu) may
be noted.

First perwopods. Side-plates rather larger than the preceding pair. Branchial
vesicles small. The first joint reaching below the side-plate, both margins convex,
the hinder carrying two or three very slender spines; the third joint about once and
a half as long as broad, slightly armed at two points of each margin; the fourth joint
narrower and shorter than the third, with very convex front margin; the fifth joint
subequal in length to the third, with spinules at three points of the straight hind
margin; the finger curved, not much shorter than the fifth joint.

38 REV. T. R. R. STEBBING AND MR. D. ROBERTSON ON

Second perwopods closely resembling the first.

Third perwopods. Side-plates with the front lobe nearly as deep as the preceding
pair, the hinder lobe much shallower. Branchial vesicles small. The first joint some-
what broadly pear-shaped, almost entirely unarmed; the third joint longer than the
fourth, shorter than the fifth, the two latter having straight parallel margins, with
rather long spines at the hinder apex; the finger curved, not half the length of the
fifth joint.

Fourth perwopods. Side-plates small, bilobed. The first joint a little longer but
narrower than in the preceding pair; the rest of the limb similar to the preceding pair,
but with the joints longer and more strongly armed.

Fifth perwopods similar to the preceding pair, but on a larger scale, and the first joint
fringed with spinules on both margins.

Pleopods. The two coupling-spines short and small, each with two pairs of reverted
teeth; cleft spines two in number, joints of the rami about seven.

Uropods. Peduncles of the first pair rather longer than the inner ramus, carrying
spines of various sizes at three or four points of the upper margin and a long one at
the apex of the lower; each ramus has spines at two points of the upper margin and a
large group at the rather blunt apex; the outer ramus is a good deal shorter than the
inner; the peduncles of the second pair have a length intermediate between the lengths
of the two rami, which resemble those of the first pair, but are a little smaller; the
peduncles of the third pair are nearly as long as those of the second, longer and much
stouter than the rami, which are equal, slender, acute, with spines at two points of the
upper margin.

Telson. The breadth at the base equal to the length, the apical margin slightly
concave, equal to more than half the greatest breadth of the telson; on either side a
couple of spinules are planted on a raised ridge that runs obliquely incurved from each
corner of the apical margin.

Length scarcely a sixth of an inch.

Locality. Cambrae, in the Clyde. A single specimen.

Remarks. ‘The specific name has been chosen because of the likeness presented by the
second gnathopods to those of Montagu’s species, J/elita palmata. The secondary
flagellum of the upper antenne is a little less rudimentary than in most species of
Podoceropsis.

4, PODOCERUS CUMBRENSIS, n. sp. (Plate VI. B.)

Rostrum small and blunt, lateral lcbes of the head produced into a blunt point just
in front of the eye. Hinder angle of the third pleon-segment bluntly produced.

Eyes round, composed of fifty or sixty ocelli.

Upper antenne. ‘The first joint rather longer than broad, the lower margin carrying
spinules at three points; second joint considerably longer than the first, with slender

NEW BRITISH AMPHIPODA. 39

slightly feathered spines at five points of the lower margin; the third joint shorter
than the second, with spines at five or six points; the flagellum of three joints,
together not so long as the first two of the peduncle; the secondary flagellum consisting
of a single joint less than a half or sometimes a third of the first joint of the principal
flagellum.

Lower antenne. Gland-cone small, the third joint with a lobe at the distal margin on
either side, the fourth joint not longer than the fifth, distally widened, the fifth joint
rather longer than the three-jointed flagellum. In general appearance the lower
antenne from the third joint onwards greatly resemble the upper, but they are stouter
and a little longer.

Mandibles. The cutting-edge has four teeth; the secondary plate on the left mandible
has also four teeth, but on the right an irregularly serrate edge; the spine-row consists
of three serrate spines; the molar tubercle is strong, with a little denticulate plate in a
recess of the forward margin; the first joint of the palp is short, distally widened, the
second joint is broad, with nine plumose spines on the convex margin; the third joint
is also broad, but shorter than the second, with a transverse row of four spines on the
surface, and about fourteen on the apex and adjoining border.

Lower lip. The outer lobes rather broad, lightly furred.

First and second maaille, so far as observed, differing little from those of Podocerus
falcatus (Montagu).

Maxillipeds. The inner plates having two spine-teeth on the apical margin and one
on the inner, and four curved spines about the apical margin ; the outer plates reaching
beyond the middle of the second joint of the palp, carrying three spine-teeth on the
inner and four on the distal margin; the third joint of the palp a little more than half
the length of the second; the fourth joint short and blunt, tipped with spines longer
than itself.

First gnathopods. The side-plates very small, almost concealed under the following
pair, the lower margin a little indented. ‘The limb small, the first joint curved, having
along seta or slender spine below the middle of the convex hind margin; the wrist
triangular, scarcely longer than the oblong third joint, each having some spines about
the distal margin and a few on the inner surface; the hand oval, about twice as long
as the wrist, with several slender spines along the hind margin and about the inner
surface; the palm ill-defined, except by the set of three palmar spines, among which
the tip of the curved finger closes, reaching a little beyond them.

Second gnathopods. The side-plates very much longer than the first pair, with convex
front margin directed forwards. The branchial vesicles remarkably small, In the
male the first joint channelled, shorter than the hand; the second joint channelled,
distally widened; the third joint scarcely longer than the second, with a couple of
setules at the apex; the wrist absorbed into the hand, which is very large, when full-
grown two and a half times as long as broad, the front margin convex, the hind margin

40 REV. T. R. R. STEBBING AND MR. D. ROBERTSON ON

having near the base a projecting tooth, which attains to a length equalling the breadth
of the hand, near to the distal extremity of which this margin forms a much smaller
tooth; the varied relations of size between these two teeth are illustrated in the figures
qn. 2, gn. 2.4, and gn. 2B; a very long, much curved finger arches over the small cavity
formed between the hinge and the distal tooth and the large cavity between the two
teeth ; in some cases the inner margin of the finger has a small prominence approaching
the distal tooth of the hand; there are several slender sete about the hinder margin
and teeth of the hand, and setules along the inner edge of the finger. In the female
the marsupial plates are longer and broader than the first joint and are fringed with
sete. ‘The first joint of the limb is longer than the hand, which is here distinct from
the wrist, the whole limb being very similar to that of the first pair, but a little
larger.

First pereopods. The side-plates much broader than the preceding pair, with the
front margin convex, the lower and the hind margin straight. The branchial vesicles
very small. The first joint with convex front and hind margins, the latter having three
sete planted near the middle; the third joint armed at two points on each margin,
much widened distally ; the fourth joint only a little longer than the second, armed at
two or three points of the hind margin and at the apex of the convex front; the fifth
joint as long as the third, armed at three points of each margin; the finger curved,
rather stout, more than half the length of the fifth joint.

Second perwopods. The side-plates large, almost square, with convex front margin,
and the hinder a little excavated. The limb as in the preceding pair.

Third pereopods. The side-plates small. ‘The first joint pear-shaped, the convex
hind margin very slightly indented, the remainder of the limb nearly as in the two
preceding pairs, but rather more slightly constructed.

Fourth pereopods resembling the third, but with the hind margin of the first joint
more strongly indented, and all the joints more elongate.

Fifth perwopods. The first joint broadest at the centre, the hind margin very convex,
the front only slightly ; the limb in general resembling that of the preceding pair, but
with the joints more elongate.

Pleopods. The two coupling-spines very short, with two or three teeth ; the peduncles
have also here and there a plumose seta; the cleft spines are two or three in number,
where there are three the third being much longer than the first; the joints of the inner
ramus from five to seven, of the outer from six to eight.

Uropods. Peduncles of the first pair longer than the rami, of which the inner has
small spines at four points, the outer at three, the outer being rather the shorter; in
the second pair the peduncle is longer than the outer, but rather shorter than the
inner ramus; the peduncles of the third pair are stout, longer than those of the second
pair, with a spine at two points of the upper margin ; the shorter outer ramus is fringed
above on the distal half with a graduated series of about nine minute denticles, the

NEW BRITISH AMPHIPODA. 4]

larger ones near the apex; the rather longer straight inner ramus has a little spine at
the extremity.

Telson triangular, as broad at the base as the length, the upper angles and the apex
rounded, each margin carrying a setule near the apex; the telson reaches about halfway
along the peduncles of the third uropods.

Length an eighth of an inch.

Locality. Clyde. Obtained in some numbers at a depth of 20 fathoms, off Fairlie
Perch, in February 1889.

Remarks. 'The specific name is derived from Cumbrae, an island in the Clyde. This
species makes some approach to Podocerus minutus, Sars, but is clearly distinguished
from it by the three-jointed flagella in both pairs of antenne, by the small concealed
first pair of side-plates, the hand of the second guathopods in the female not excavate,
and the large tooth of that hand in the male being simple instead of double-ended

EXPLANATION OF THE PLATES.

PLATE V.

A. Sophrosyne robertsoni, n. sp.

The full figure is given in lateral view, with a line above it indicating the natural size.

a.s. Upper antenna. mx. 2. Second maxilla.
a.i. Lower antenna. map. Maxillipeds.
m. Mandible. o.t. Triturating organ, from the stomach.
mx. 1, First maxilla, the palp imper- gu. 1. First gnathopod.
fect. gn. 2. Second gnathopod.

prp. 2, 3, 4, 5. Second, third, fourth, and fifth pereeopods respectively.
sp. of plp. Coupling-spines and a cleft spine of a pleopod.
ur. 1, 2, 3. First, second, and third uropods respectively, the telson being shown in
combination with the second and third uropods.

B. Syrrhoé fimbriatus, n. sp.

a.s. Upper antenna. ai. Lower antenna.
m.m. The right mandible, with the palp imperfect, and a portion of the left mandible.
gn. 1. First gnathopod. gn. 2. Second gnathopod.

prp. 1, 2, 3, 4, 5. First, second, third, fourth, and fifth peraeopods respectively.
plp. 1. First pleopod.
Pi.s. 3. Lower lateral portion of third segment of the pleon.
ur. 1, 2, 3. First, second, and third uropods respectively.
t. Telson.
VOL. XIII.—PART 1. No. 6.—January, 1891. G

49 ON NEW BRITISH AMPHIPODA.

PLATE VI.

A. Podoceropsis palmatus, n. sp.
The full figure is given in lateral view, with a line above it indicating the natural size.

a.s. Upper antenna.

m.m. Right and left mandibles at the right and left corners of the Plate
respectively.

mip. Maxilliped.

gn. 1. First gnathopod. gn. 2. Second gnathopod.
prp. 1, 3, 4, 5. First, third, fourth, and fifth pereeopods respectively.
ur. 1, 2, 3. First, second, and third uropods respectively.
tT. Telson.

B. Podocerus cumbrensis, n. sp.
The full figure is given in lateral view, with a line above it indicating the natural size.

oc. One of the eyes adjoining the lower angle of the head.
a.s. Upper antenna. a... Lower antenna.
m.m. The left mandible and the anterior portion of the right mandible.
1.2.8. Lower lip. ‘lhis and the other parts marked B were figured
from another specimen, not from that drawn at the top of
the Plate.
me. 1B. First maxilla. mx. 2B. Second maxilla.
map. Maxillipeds, seen from the outer surface, but with outer plate
and palp removed from the half to the right so as to expose
the inner plate.
gn. 1, gn. 1,2. First gnathopod of male and female respectively.
gn. 2, gn. 24, gn. 2B. Second gnathopod of the male, three different specimens.
gn. 2,2. Second gnathopod of female.
prp. 1, 2, 3, 4, 5. First, second, third, fourth, and fifth pereeopods respectively.
ur. 1, 2, 3. First, second, and third uropods respectively.
T, T.B. Telson.

Trams Loo€ oe Vol MM FEV

Sp.of plp.

N f
ae 0 plp.

Del TRR.Stebbinp J.T.Rennie Reid Lith Edin?

SOPHROSYNE ROBERTSONI!I, N. SP. SYRRHOE RIMBIR TALUS) IN: (Sie:

Trams. Lo0€. Hoc. Vek AM SEV

J.T Rennie Rerd Lith Edin?

Del. TRRStebbing.

PO IDOCG HIRO Sis) VME OS IN 1Sie PODOCERUS CUMBRENSIS, N. SP.

III. On the Morphology of a Reptilian Bird, Opisthocomus cristatus.
By W. K. Parxer, FBS.

Received January 4th, 1890, read February 4th, 1890.

[Puates VII.—X.]

Conrents.

Page

I. Introductory Remarks on the Present Existence of Birds clesely related
bowReptilesy it nipacscpcvamcacnenet Ants veecoxckas sucnsvere reieiousncnede cohen apeueaamucuctebonttieie 44
II. The Early Stages of Opisthocomus cristatus.........2 +. cece cece eens 48
III. The Skull of Opisthocomus eristatus in Embryos and Adult ............ 49
IV. The Vertebral Chain of Opisthocomus cristatus 1.1.1... 0. sees eeeeees 59
VY. The Sternum and Shoulder-girdle of Opisthocomus cristatus ............ 64
WAE DhesWineslofsOpisihocomus) cnzstatus seismic cl sleieisiciels eieieieieleeicle 69
VII. The Hip-girdle of Opisthocomus cristatus 1... 0... cece cece ee cece eee 74
VIII. The Hind Limb of Opisthocomus cristatus .........-...5..--0---0ces 77

IX. Recapitulation and Summary.

a. The Ornithological Position of Opisthocomus .........-00eeeeeeee 80

6. The Light cast upon the Ontogeny of Birds by the Morphology of
ORSVMCUTES 90500000000060000000000000020000000000000000 81
OS MUSH OF A AEITOND ooaboooacoooconbddceco0b DoD DUdCOOODOHaSOAC 83
XIE PDescriptionvoistheek aves peer eetolelrrockhdsk tiki teiorenarr teeters 84

EARLY last year Mr. Sclater received from Mr. John J. Quelch, C.M.Z.S., of the
Museum, Georgetown, British Guiana, a series of embryos of the Hoatzin (Opistho-
conus cristatus). Some of these, after due examination by Mr. F. E. Beddard1, were
sent to me by Mr. Sclater. My study of the adult has been from the two skeletons in
the Museum of the Royal College of Surgeons, kindly lent me by the President,
Sir W. Savory, and the Curator, Prof. Charles Stewart. I take this opportunity of
thanking all these friends for their kindness.

Hitherto my knowledge of the structure of the skeleton of this bird has been
derived from Prof. Huxley's masterly description, given partly in his paper “On the
Classification of Birds,” and more completely in his paper ‘‘ On the Classification and
Distribution of the Alectoromorphe and Heteromorphe” (Proc. Zool. Soc. 1867,
pp. 415-472; and ibid. 1868, pp. 294-319).

* See Mr. Beddard’s paper, ‘ Ibis,’ 1889, p. 283.

VOL. XII.—PART U. No. 1.—April, 1891. HW

44 PROF. W. K. PARKER ON THE

I had also made my own observations on the skeleton of an old male bird of this
species in the Hunterian Museum. Judging from the figures given by Prof. Huxley in
his second paper, I am of opinion that the specimens there figured were from skeletons
of female birds.

What struck me at first in the Hunterian specimen was that it is much more like the
skeleton of a strong Curassow (Craz) than those from which Prof. Huxley’s illus-
trations were taken; these latter have a very Musophagine appearance ; they suggest
an evident relation to the Plaintain-Eaters. Thus, while I am glad to refer to those
figures, and the excellent descriptions given of them, and the accompanying remarks
on the affinities of this bird, my own notions of the structure of the adult, here given,
will be from observations on the Hunterian specimen.

I. Introductory Remarks on the Present Existence of Birds closely
related to Reptiles.

Two or three facts must be noticed at the outset of these remarks: namely, first, that
the known extinct forms are very few in proportion to the multitudes of those that are
still alive ; secondly, that the Tertiary forms are closely related to existing types, and
throw but little light on their origin; and, thirdly, that the very few most precious
relics of the Secondary Rocks startle us at once by being tooth-bearers—not like our
familiar forms with their horny beaks. Leaving out of consideration the mysterious
and apparently quite isolated Archwopteryx, the types described by Dr. Marsh in his
magnificent work on the “Odontornithes, or Extinct Toothed Birds of N. America,”
1880, teach some very remarkable facts. If there is one modification of the skeleton
of a bird which, more than any other, is peculiar and typical, it is the mode in which the
presacral vertebree, in the majority of cases, are articulated together; I refer to the
cylindroidal or heteroccelous condition of the centra. I lately showed (Proc. Roy. Soc.
1888, pp. 465-482) that a much greater number of modern or existing birds than was
hitherto supposed retain the old Reptilian or Opisthoccelous condition in several of
their presacral vertebre, namely, in the dorsal region; and that this condition is seen
in arboreal Altrices, as well as in aquatic and semiaquatic Preecoces. If that is not a
reptilian character I know of no one that can be described as such; it is equal in value,
in my opinion, to anything that can be found in the skull, the shoulder-girdle, or in any
other part of the skeleton.

For practical ornithological purposes the existing birds may be fenced off into
two groups, namely, the Ratitee and the Carinate. We thus get some two dozen
archaic forms, mostly from the Southern Hemisphere, and twelve thousand modern
forms with the Old World at their feet. Now all the existing Ratite have their
presacral vertebre cylindroidal ; whilst some of the most highly specialized and large-

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 45

brained forms, for instance the Parrot family, have opisthoccelous dorsals. Moreover,
classifying the Toothed Birds of Marsh by their sternum, the type that is one of the
Carinatee has a still lower kind of vertebral articulation than the opisthoccelous ; Ichthy-
ornis has nearly all its presacrals amphiccelous; whilst Hesperornis, which is one of
the Ratite, has its presacrals cylindroidal. Nor is this all; if Hesperornis has any
living descendants, or even representatives, these are the existing Loons (Colymbide)
and Grebes (Podicipedide).

But the Carinate [chthyornis appears to me to be an ancient toothed Sea-Gull, one
of the Laridee ; and yet the existing Gulls are far more intelligent birds, birds of a
higher order than the Loons and the Grebes. Now the Gulls have not only opistho-
ceelous dorsals, but the last presacral joint is nearly amphiccelous, as in their Cretaceous
quasi-ancestral relative the [chthyornis. With regard to that character by which the
Struthious birds are most definitely marked off from the ordinary flying birds, namely
the flat breast-bone, correlated with almost useless wings, quite useless as organs of
flight, it is not too much to say that this is a modern and an acquired character.

The fact that the vertebral chain in these low birds is longer than what is the average
in the Carinatee is a noticeable and important fact; but they all fail in the caudal
region; their free caudals are always fewer than in the Carinate. Amongst the
Carinate, however, there are some birds, namely the Swans, which have a longer
vertebral chain than the large Struthious birds.

This abortive development of the tail is also manifestly an acquired character, like
the starving of the wings and sternal keel. But this is part of the same specialization of
these types in which all the strength goes to the hinder limbs, making them so perfectly
adapted for terrestrial life—in the case of the Ostrich especially, which is a bird as
exquisitely fitted for swift running as its desert companion the Wild Ass.

As for what is most archaic and quasi-reptilian in the Ostrich tribe, that is to be
looked for in the skull and its contents. ‘Those Carinate that come nearest to these
birds, the Tinamous, have also an extremely small brain and low intelligence ; and in
some things, namely in the retention of cranial sutures and in the development of one
or even two rows of superorbital dermal bones, they are more reptilian than the Ratite
themselves. But these birds, and the one treated of in this paper, Opisthocomus, have
retained good functional wings, and in the former, the 'Tinamous, a very large sternal
keel, although only the most Gallinaceous species has the sense to fly *.

The Hoatzin (Opisthocomus) rather flits than flies ; its flying power is about equal to
that of the heavier Fowls (Turkey, Peacock, &c.); but it is an arboreal type, like its
nearest relatives the “‘ Peristeropodous ” Cracide ; the Tinamous are the sub-struthious
relations of the Grouse, which are “ Alectoropodous ”—have a high hallux, and are
mainly terrestrial birds.

It comes to this, namely, that birds, like men, must be classified by their brain power

See Frederic A. Lucas, Proc. U.S. Nat. Mus. 1886, pp. 157, 158.
H 2

46 PROF. W. K. PARKER ON THE

and intelligence: a Rook has a lesser body than an average Tinamou, but it has three
times its bulk of brain; the Raven and the Ostrich stand at the extremities of the bird
series. Opisthocomus is not the only Neotropical type that belongs to the region round
about the Ostrich territory; I have just mentioned the Tinamous, and there are several
other rare, unclassifiable, and manifestly archaic types in the rich Avifauna of the
American Tropics. It is in the Neotropical Region that we find the most archaic forms
of every family. It is there also that we meet with the low harsh-voiced Passerines,
aberrant in various ways; Cuculine forms that are so torpid that they become a mass
of fat (Steaternis) ; and true Cuckoos (Geococcya, &c.) that walk the ground firm, like
Fowls, and have a pelvis that is strongly Ornithoscelidan. In this region, also, there
are birds related to Geese that have the face of a Hen (Palamedeide) ; and a genus
of the Crane family (Psophia) with a Pea-fowl’s head and the bony brows of a
Tinamou. ‘These and various others characterize the rich and unique Avifauna of this
region. Nor are these all the rare and isolated types to be found there; we have,
also, Hurypyga, Dicholophus, Attagis, Thinocorus, Chionis, Phaéthon, and Tachypetes. Of
course the Eastern Region south of “ Wallace’s Line” yields many important and rare
forms, especially among the Ratite; but for archaic Carinate it is far inferior to
the Western Tropics.

The type now under consideration, being the only one of its family, and considered
by Prof. Huxley to represent, not a Family merely, but a Suborder or a bundle of
Families (the Heteromorphe), must of necessity be archaic, for all its near relations
have been weeded out in the past. This is as self-evident on one hand, as, on the other,
it isa sure induction that the wise Raven is a modern type; for he has not only
acquired all the highest accomplishments of which a bird is capable, but, as the head
of a long list of Families, he has an ornithological following of more than six thousand
species, or half the number of existing birds.

Anticipating somewhat the descriptions now to follow, I may remark that, besides its
isolation asa type, Opisthocomus is aberrant as a Carinate bird in the Struthious character
of its palate ; its temporary basipterygoids are Tinamine ; its scapula is Batrachian ; its
three clavicles are more Lacertian than those of any other bird; whilst its sternal keel
is permanently rudimentary. Its wing also has the largest claws in it of any known
kind, with a rudiment of a third claw and two rudiments of a fourth digit; and to
crown all, for a time, it has, before hatching, eight distinct carpals; and the inter-
medium of the tarsus is one of the longest and best developed ever yet seen by me in
any bird. Of course, the existence in these days of a bird like this is not merely a
“foreign wonder” to the ornithologist ; its great importance lies in the light it sheds
on the uprise of the feathered types during time. A comparison of the ankle-joint
of the bird, in its early state, with what is extinct in the Ornithoscelidan Reptiles was
a great stroke in this enquiry; but it is only in the hips and hind limbs that those
Reptiles resemble immature birds. In the length of the neck, and in the shortness of

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. AT

the tail, birds are like Plesiosaurs; but in the structure of their skull, and I believe,
also, that I may state in the development in their fore limbs of intercalary digits, they
resemble the Ichthyosaurs.

Like the Chelonia, Tertiary and existing birds have lost all-traces of teeth and have
horny jaws ; and like the Crocodile’s embryo, and the adult Hatteria (Sphenodon), the
hyoid arch and the columella auris become continuous. The superficial parts of the
shoulder-girdle of a bird, its parostoses, are like those of an Ichthyosaurus, an
ordinary Lacertian, and of the Monotrematous Mammalia; but, as a rule, these are
fused together. Moreover, these parts do not remain in their primitive distinctness; as
a rule they graft themselves upon proximal remnants of the antero-inferior fork of the
shoulder-plate (precoracoid). I have lately shown (Proc. Roy. Soc. 1888, pp. 397—
402) that the skull of a bird is rich with the remains of truly Amphibian structures ;
indeed, it is far more Amphibian in its very foundations than any existing Reptile; the
parasphenoid of a Frog and of a Bird closely correspond. Then there are the
remnants of the ethmo-palatines and certain superficial structures that, as a rule,
have been got rid of even in Reptiles. The one or even two rows of supra-orbital
bones are remarkable ; but still more so is the fact that the long Jugo-maxillary chain
of ganoid bones seen in Lepidosteus is represented in some birds (Emu, Owl, Heron,
Cormorant) by a chain of four bones behind the premaxillary, namely, the maxillary,
postmaxillary, jugal, and quadrato-jugal. In the palate, also, besides the endoskeletal
postpalatines of the Passerine birds and the medio-palatine of the Woodpeckers (pre-
formed in cartilage), we have in the latter birds especially, and also in others, a literal
crop of vomers, such as we see also in the Marsupials among the Mammalia. Even
these parosteal remnants are too valuable to be lost sight of or left as unrelated; they
at once remind the Morphologist of what is seen in the generalized pavement of bones
under the rostrum of a Sturgeon.

It is quite certain that not the least patch of bone-cells is ever differentiated and
isolated accidentally ; as a matter of observation in birds, the smallest of such patches
is very uniform in its appearance in various genera and families. One more fact in
evidence of what I am anxious to express—namely, that birds did not appear during
time, as a sort of feathered sport in a truly reptilian type, but from some low, genera-
lized Amphibian or Dipnoan, is to be seen in a character now to be described. The
earliest embryo I have examined of this type has a dilated suprascapula, separated
from the scapula by an arched line of smaller cells, and overlapping the scapula
proper at its edges. This upper element is, for a short time, quite as distinct as in
the Frogs and Toads ; lower down, in the Rays, among the Elasmobranchs, the supra-
scapula is perfectly segmented off from the scapula (see ‘ Shoulder-girdle and Sternum,’
1868, pls. i. and v.). I have one more preliminary remark to make which bears
directly upon the relation of Birds and Reptiles, and that is in the clear evidence we
have of secular shortening of certain parts. In some birds, for instance the Humming-

48 PROF. W. K. PARKER ON THE

birds and the long-billed Shore-birds (Limicol), the bill rapidly elongates and takes
on special curves, up or down, after hatching; these are comparatively recent
specializations. But in some other kinds, e.g. the Guillemot (Uria trotle), the
bill tends to become as long as it is in the toothed birds of the Cretaceous Epoch.
By the middle of incubation, the parosteal tracts first dominate, and then largely take
the place of the endoskeletal elements, clasping them and stopping their growth. For
a time, the cartilaginous rods struggle to grow to the ancestral length, but in this effort
they become bent and defcrmed; this is all put right by the time of hatching; they
cease to grow, overmastered by the enveloping splint-bones. But this twisting, and
then arrest, of the cartilaginous rods appears also where there are no splint-bones, but
merely the nerves, muscles, and ligaments to which these parts are related. This
curious quasi-deformity is seen, not only in Uria trotle, but I shall soon show and
describe it here, in Opisthocomus ; it appears in the pubic rods and in the columella
auris. Of all existing birds, the African Ostrich comes nearest the Iguanodon and its
kindred in the size of the rotated pubes ?.

There has evidently been, first, a secular rotation and elongation of the pubis and
ischium, and then a re-shortening of these parts in the higher kinds of birds. A
similar phenomenon is seen in the length of the sacrum, and correlatively of the extent
of the ilium, behind and before. The African Ostrich and the Swan have each twice,
or nearly twice, as many sacral vertebre as the lesser birds of the higher kinds, both
Passerine and Cuculine; and these lesser forms constitute about half the known
existing birds. ven in the Rook (Corvus frugilegus), one of the chief of the Coraco-
morphe, and in that Order a large bird, there are only eleven sacrals in the adult;
whilst there are twenty-one, and sometimes twenty-two, in the common Swan and in
the African Ostrich.

The young Rook, after it is fledged, has a longer sacrum than the adult; there is,
for a time, a twelfth vertebra in that part of the spine; the ancestral Crows had,
probably, a longer sacrum than the existing forms. At present, with a large number
of unclassifiable facts in hand, it is safer for us to confess our ignorance as to how Birds
and Mammals arose, than to invent crude hypotheses that will only be mocked at by
those who enter into our labours in the time to come. Of one thing I feel certain,
even now, and that is that no feathered or hairy form ever arose from a true gill-less
Amniotic Reptile. A very large proportion of the Reptiles that have arisen during
the geological ages have died out; these all came short, as the existing reptiles now
come short, of the high excellences of the hot-blooded types, feathered or hairy.

Il. The Early Stages of Opisthocomus cristatus.

There were three different stages in the four specimens sent to me; these, being
measured, gave the following lengths from the end of the beak to the end of the tail:—

* See Dollo, on Jguanodon bernissartensis, Blgr. Bull. Mus. Roy. Hist. Nat. Belg. t. ii. pl. v.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS, 49

Stage A. 23 inches long; about half ripe.
Stage B. 33 inches long; about two-thirds ripe.
Stage C. 44 inches long; about three-fourths ripe.

The largest embryo was evidently not quite ripe for hatching, and the smallest was
at any rate half ripe ; I come to this conclusion by comparison of these embryos with
those of the Common Fowl. I am in a position now to compare the developing
skeleton of this rare type with that of its familiar relative, for the descriptions and
figures of the skull, shoulder-girdle, and wings of the latter are already published
(see Phil. Trans. 1869, pls. 81-87, Proc. Roy. Soc. 1868, and Phil. Trans. 1888, B,
pl. 62); and the rest of the skeleton of the chick has been worked out by me and will
soon be published.

II. The Skull of Opisthocomus cristatus in the Embryos and Adult.

The chondrocranium in the first stage is at its fullest development, and is beginning
to undergo ossification in certain parts (Pls. VII., VIII.) ; it is now in the best state for
comparison with the cartilaginous, or osseo-cartilaginous, skull of the Ichthyopsida, and
with the early condition of the skull in Reptiles and Mammals. The solid fore part of
the premaxillaries (px.), where the right and left bony tracts are already fused together,
is as long as the vestibular region of the nasal labyrinth (Pl. VII.). In this short-
headed bird the brain-cavity extends only along the hinder two fifths of the whole
chondrocranium ; but its axial is only part of its real length, as it is tilted upwards
very much, whilst the auditory capsules are tilted downwards and backwards so as to be
almost horizontal (see 3rd stage, Pl. VIII. fig. 2). The whole structure, when
deprived of the investing bones, is a short and rather shallow basin behind, with a long
high wall in front. The foremost part of the face in front of the wall, and one-fourth
of its length, is the free, rounded, somewhat flat, prenasal rostrum of the inter-
trabecula 1.

The nasal labyrinth is like that of the Common Fowl, but the inferior turbinals
(Pl. VIII. fig. 4, 7.t0.) are simpler; they make scarcely more than one turn of a coil.
In the normal ornithic skull we have repeated the high type of cranium seen in so
many osseous fishes; whilst in Mammals the low type of cranium, seen in the Skate
and the Frog, is once more adopted. This is the more important to notice because of
the fact that of all the Ichthyopsida the high-skulled Teleostei are the most instructive
types with which we can compare the Carinate, with their marvellously elastic and
mobile jaws and palate. In these high skulls the posterior sphenoid, only, is fully

1 In the Apteryx on the one hand, and in the Ibis on the other, the notochordal region of the skullis extremely
short as compared with the prochordal; in Opisthocomus the latter, in this stage, is more than half as long as
the notochordal, even allowing for the upward tilting of the notochord between the moieties of the post-
pituitary or clinoid walls.

00 PROF. W. K. PARKER ON THE

developed in front of the large, hollow occipital arch ; the alisphenoids (Pls. VIL.,
VIII. fig. 2, ai.s.) are large and simply postorbital in position; whilst the basi-
sphenoid, dominated by the auditory organs, and underfloored by the huge para-
sphenoid, is one of the largest and most complex structures to be seen in the whole
skull. The anterior sphenoid (p.s.) is the hinder half of the interorbital partition; it
is partly segmented from the meso-ethmoid, even in the earliest of my stages, by an
oval fenestra (Pl. VII., ¢.0.f.). The upper part, which overhangs the large optic
passage (II.), as a rule, has scarcely any development of cartilaginous lips representing
the orbito-sphenoids (0.s.). The only bird in which these are at all well developed, even
in the early embryo, is the African Ostrich (Phil. Trans. 1866, pl. vii. figs. 1-8, 0.s.).
In Opisthocomus they are well developed for a bird (Pl. VIL. 0.s., and Pl. VIIL. fig. 2,
0.8.), and come nearest the Ostriches of any I have yet seen (see in the chick of the
Common Fowl, Phil. Trans. 1869, pl. Ixxxiii. figs. 2 & 4). Inthe 3rd stage the peculiarity
of the ethmoidal region of a bird is well shown; the pars perpendicularis (p.e.) has
already a large reniform osseous centre (PJ. VIII. fig. 2, p.e.) which will ultimately reach
twice as far back to ossify the cartilaginous crista galli (er.g.), and also, below the
interorbital fenestra, will grow some distance backwards to meet the feeble presphenoid
(p.s.) and the enormous basisphenoid (6.s.). This chondrocranium, so different from
that of a Mammal, on the one hand, and from a Batrachian, on the other, already, in
the 3rd stage, shows the beginning of that character which separates the skull of the
Carinatz, not only from that of other Vertebrata, but also from that of the Ratite
themselves. This is the secondary segmentation of the vertical trabecular wall, so that
the septum nasi in front is quite, or nearly, separated trom the perpendicular ethmoid
behind. In this modification, Opisthocomus agrees with the Carinate generally ; and
in the 3rd stage (Pl. VIII. fig. 2) the separation has taken place to an extent equal to
what I have found it in an adult Tinamou (see Phil. Trans. 1866, pl. xv. fig. 8, ¢.fic.).

In the earlier stage the whole prochordal tract is continuous and only loses its vertical
crest where that part is not wanted, namely, in front of the nasal labyrinth; thus the
intertrabecula runs on as a free bar in front of its crested part. But in the Srd stage
(Pl. VIII. fig. 2, 7.tr.), under the notch just mentioned, the fore part of the inter-
trabecula is already separated from that which thickens the dividing wall at its base ; it
is even now undergoing degeneration, and will disappear entirely after a time. This
is only one of the many prenatal transformations to be seen in the embryo bird, all

1 T have spoken before of the remarkable isomorphism existing between the culminating Fishes (Teleostei) and
the culminating forms of the Sauropsida, seen in the extreme mobility of the jaws and palate. This mechanism
is obtained in the former by the hyostylic condition of the pier of the mandibular arch, the palato-
quadrate being swung on the hyo-mandibular and symplectic. But there is no segmentation of the cranial
axis in front; that part is extremely short; and the large dominating premaxillaries ride over it, and in
most cases throw the mavxillaries back, as the edentulous “ossa mystacea.” I may state that in such a
bird as the Cock of the Woods (Zetrao wrogallus) those bones are extremely like their counterparts in the
Teleostei.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 51

showing how high even the lower kinds of Carinate are, as compared with the cold-
blooded Sauropsida !.

In these high skulls there is a partial secondary segmentation of the ethmoid from
the anterior sphenoid by a pyriform interorbital fenestra (7.0); above this the
ethmoid ends as a blunt cartilaginous crista galli, under which the olfactory crura (1.)
run forwards to the simply infolded upper and middle turbinal regions. Below this
groove we see the narrow orbito-sphenoidal alz (0.s.). In all these stages the anterior
sphenoid is wholly unossified : in Stage 1 the only part of the posterior sphenoid that
has any bony matter is the rim of the pituitary hole; the trabecule do not form a
floor to this part; and that bony substance is borrowed from the parasphenoid, a
mere parostosis. ‘The large leafy postorbital alisphenoids are still thin upgrowths of
cartilage (Pl. VII., a/.s.); but in Stage 3 these are largely ossified (Pl. VIII. fig. 2,
al.s.), and although turning in behind the eyeballs, do also form the foremost third
of the lateral wall of the skull, the rest being made by the superficial low-lying
parietals (Pl. VIII. fig. 2, p.). Under these two bones, and wedged in between them
and the large occipital arch, we see the large, long auditory capsules, which are so
much tilted back as to have their top nearly as low as their base. These cartila-
ginous capsules were fused very early with the parachordals, and with their lateral
upgrowths, alisphenoidal and exoccipital tracts.

A considerable osseous centre is already seen in Stage 3 (Pl. VIII. jig. 2) between
the meatus internus (VII., vil.) and the elegant anterior semicircular canal (a.s.c.); this
is the prootic (pr.o.). Below, also, over the edge of the exoccipital (¢.0.), a lanceolate
bone is seen hardening the postero-inferior edge of the capsule; this is the opisthotic
(op.); the epiotic is rarely seen in birds, and then only as a small tract of bone. Even
in Stage 1 the double supraoccipital (s.0.) has become a single tract of bone; the
exoccipitals are growing well at the sides of the arch, leaving, however, the large
tympanic wings (Pl. VIL. ¢.¢0.) in a soft state; the basioccipital is still largely hidden
in the cartilage (Pl. VIII. fig. 1, 6.0.) ; it is forming, however, round the sheath of the
notochord (ne.) ; the occipital condyle (oc.c.) is slightly bilobate. In Stage 3 (Pl. VIII.
fig. 2, nc.) the notochord has retreated into the basioccipital region, although at first
it reached, at least, the top of the post-pituitary wall. Under that wall, in this advanced
stage, much absorption of the basal cartilage has taken place, to form the middle part
of the cavities in front of the cavum tympani—the pre-tympanic recesses of the basi-
sphenoid. ‘These parts will be described with the parasphenoids; but before leaving
the chondrocranium I must refer to the foramina of the cranial nerves.

The first nerve (Pl. VIII. fig. 2, 1.) has already been mentioned ; it is a single crus, and
does not need a lamina cribrosa. The optic nerves (11.) have a large common passage

It seems to me that the primary Dipnoan Bird-stock had a face as long as in the Skate; not short, as in
the existing Dipnoi and Amphibia.
VOL. XIII.—PART 11. No. 2.—April, 1891. I

52 PROF. W. K. PARKER ON THE

in a semicircular notch on the back of the interorbital wall, a little in front of and
above the deep pituitary space (py.). The lesser nerves, 3rd, 4th, and 6th, pass out to
their destinations through the membranous interspace that is now seen between the
foramen opticum and the f. ovale; the latter is very large for the transmission of the
5th nerve (Pl. VIII. fig. 2,v.). It lies over the meatus internus for the 7th and 8th
nerves (Vil, vul.). The passages for the vagus, glossopharyngeal, and hypoglossal
nerves (Pl. VIII. fig. 1, x., x1.) are figured in the least and largest of these embryos.
The glossopharyngeal and vagus pass out under the opisthotic bone, and the hypo-
glossal pierces the exoccipital. In the Sauropsida there is a small passage in front of
it which makes it seem to be the “posterior condyloid foramen;” in Mammals a
notable vascular passage is found behind that for the 12th nerve. That nerve in the
bird (see Pl. VIII. fig. 2, x1.) is shown as escaping opposite the occipital condyle, with
its notochordal dimple (nc.), and having directly in front of it a small vascular passage.
The large internal carotids find their way into the skull over the parasphenoid and pass
through the pituitary hole; they ultimately have a bony tube formed accurately round
them. The rest of the cranium proper is formed of membrane and membrane-bones,
or parostoses.

But besides the inner skull, or chondrocranium, there are the cartilaginous visceral
arches that are formed in segments in that tract of the head of the embryo which
corresponds with the splanchnopleura in the postcephalic region. These cartilages
begin to solidify before the head-cavities are closed ; they are indeed the first three
branchial -arches; for, as is well known, in the Elasmobranchs the quadrato-mandi-
bular and hyoid arches both carry gills !.

Whatever the prochordal tracts of the chondrocranium may turn out to be, whether
visceral or cranial, the post-oral arches are determinable’.

In this type, as in most of the Sauropsida, the palato-pterygoid arcade in front of
the quadrate pier is merely developed as membrane-bone, although secondary tracts of

’ Tf this fact had been attended to, the strange misconceptions which have arisen as to the nature of these
parts would have been avoided. In the study of the Morphology of the Skull, before all things, it is necessary
that the fundamental embryological development of this part should be mastered. Anything more hopeless
than the confusion produced by want of this knowledge in some Memoirs on this subject cannot be conceived.
The most remarkable instance of this is to be seen in Dr. Gadow’s paper ‘‘On the Modifications of the First
and Second Visceral Arches ” (Phil. Trans. 1888, pls. 71-74, pp. 451-485). If our modern Morphology can
do no more for us than this, we had better return to “‘ Transcendentalism ” and blindly follow Oxern.

* Prof. Huxley (see his paper on Petromyzon, Journ. Anat. & Phys. vol. x. pp. 412-429) classifies the
trabecule with the visceral arches; my own view at present is that they do belong to the same original
eranio-facial basket-work, such as is seen in the Myxinoids (Phil. Trans. 1883, pls. viiiixxvi.); but that the
whole of that continuous skeletal tract is, I feel certain, a prochordal skeleton formed in time before any
vertebral rudiments appeared, and even before there were any skeletal elements in the occiput. This basket-
work was developed for the support of the enlarging brain and increasingly complex oral apparatus long before
any paired vertebral rudiments grew for the support of the notochord.

o

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 55

cartilage crop up wherever they are needed in the mobile upper face. ‘The quadrate
or upper segment of the mandibular arch (an epibranchial element) is, in this bird,
all we have above the gape. In the early stage (Pls. VII., VIII. fig. 1, g.) this is a
large segment of cartilage, the body of which has a bony shaft. The true apex or free
pedicle (orbital process) is flat and rounded; the secondary apex or otic process is
rounded, and fits in between the squamosal and the anterior margin of the auditory
capsule, close in front of the anterior ampulla. As in birds generally, but not as, as a
rule, in the Ratite, nor in the Tinamide, the lower articular facet is bilobate; the
outer lobe is somewhat in front of the inner; inside the latter there is a knob for the
pterygoid, and outside the former there is a cup for end of the quadrato-jugal. The
endoskeletal part of the lower segment, or primary mandible, is in a very instructive
condition in these chicks. In the youngest embryo (Pl. VII., mk.) Meckel’s cartilage
is exposed on the inner side up to its fore end; but in the largest (Pl. VIII. fig. 2)
it is enclosed by the dentary (d.) and has what the other has not yet, namely, an
ectosteal plate, the outer ossification of the articulare (ar.); this is a thin, deep, lanceo-
late tract of ossified perichondrium ; the endosteal part of the articulare has not yet
appeared. As bearing upon the order in time of the appearance of these bony
deposits, I may remark that in the half-grown Green Turtle (Chelone viridis) the
endosteal centre is not present; in old Turtles it is large. The Anura have only a
dentary and a large ectosteal articulare as a rule; but in the skull of one of the most
teebly ossitied types, Bombinator igneus, I find, contrary to rule, the endosteal part of
the articulare. The thick malleal part of the primary mandible in Opisthocomus is
generalized ; it is more like that of a Plover than that of a Fowl, for the posterior
angular process (p.ag.) is a very small hook; the internal angular process (¢.aq.) is
large, thick, and normal. The development of the posterior angular process is greatest
in that gigantic Fowl, the Cock of the Woods (Yetrao urogallus), but it is large in all
the true Fowls; in the Anatide, which in some things are marvellously like the Galli-
nacee, and also in the Flamingo and anserine Ibis, it is large. ‘Thus in this part of
its structure Opisthocomus is seen to have fallen short of the Fowl type; it is a pre-
gallinaceous bird, like the 'Tinamou '.

The hyoid arch (Pls. VII., VIII. fig. 3) is quite normal and similar to that of the
Common Chick; the distal part of the second arch is merely the lower half of a
ceratobranchial rod, with no terminal or hypobranchial segment. The median
element (0.hy.) is formed in the usual way by the fusion and partial separation from
the ends of the ceratohyals (c.hy.). The rest of the basal tract behind is a single rod

1 The first visceral arch in the Amphibia and Sauropsida generally is chondrified at first as two tracts,
an epibranchial and ceratobranchial. In the Salmon (Phil. Trans. 1873, pls. 1-3) all the visceral arches are
developed as continuous bands of cartilage, and are segmented afterwards. The same thing takes place in the
mandibular arch in Marsupials (e. g. Macropus major); thus the incus and malleus are continuous at first and

become separate after birth.
12

o4 PROF. W. K. PARKER ON THE

(b.h.br.), and really belongs to the third postoral arch, and to arches that have been
suppressed behind that. That third arch is developed as the “cornu major”; it is
merely divided into a ceratobranchial and epibranchial (¢.br., ¢.br.), the lower pieces
ossifying. The upper part of the second arch repeats the old Amphibian specializa-
tion; it is now the skeleton of the middle ear—the stapes or columella. Morpho-
logically this part (Pl. VIII. figs. 5, 6) is a pharyngobranchial, an epibranchial, and
the beginning of the ceratobranchial region ; this is followed by a membranous tract
of considerable and, during growth, of relatively increasing extent. Below, the cerato-
hyal=ceratobranchial just described, breaks out again. In these arrested and
specialized pharyngohyal and epihyal cartilages we have the archaic character seen
in Hatteria, and in the Crocodilia for a time, namely, the fusion of these two cartilages,
the intermediate and secondary element, the “ interhyal,’ so well known in Ganoids
and Teleosteans, binding together the two tracts. The interhyal (=infrastapedial)
is present in the ordinary Lacertilia, but it is free below; it does not catch the top of
their long epihyal; it does not exist in the Amphibia.

All birds are therefore Hatterian in this respect ; but the embryos of Opisthocomus
show this better than any birds I have yet worked out. This specialized hyomandibular
is at first of the full relative size ; its growth, however, scon becomes arrested : it has
the usual dilated opercular plate; a short thick shaft, the mediostapedial (m.st.); a
tongue-shaped extrastapedial (e.st.); and a forked suprastapedial (s.st.), finished above
by a ligamentous tract. The infrastapedial (¢.st.), (=interhyal of Fishes and Mam-
mals) is developed directly from the columella and ends in a bulbous form ; it is half
the length of the extrastapedial, and one-third of its width. Articulated to this 1s a
tract of cartilage as long as the whole columella proper ; this is sharp above, and then,
after articulating with the infrastapedial, which it exceeds in size, it turns suddenly
backwards, and then makes a second sudden turn forwards, and ends in a large tongue-
shaped lobe, which lies on the edge of the basitemporal bone. This cartilage is the
epihyal, with a continuous but enlarged rudiment of the top of the ceratohyal !.

1 Let this structure be compared with what is seen in Hatteria, and also for a time in the Crocodile (see
T, Z. 8. vol. xi. pls. 68 & 69). The ectocranium of Opisthocomus conforms very closely to that of Carinate
birds generally, but it comes nearest to that of the Gallinaceous tribes. It is a “ holorhinal” skull with a
strongly curved rostrum and highly ossified endocranium. The cranial and rostral parts are of nearly equal
length. The hinge of the rostrum on the cranium is, however, much more perfect than in the Common Fowl;
and this agrees with the normal Cracide, which are very Cuculine in this respect. The splints of the rostrum
or upper face are well formed already (Pls. VII. & VIII.), but in the first stage the premaxillaries (px.) do
not cover the prenasal rod (pn.) in front. The nasals (m.) have a round notch behind the alinasal wall;
this skull is therefore holorhinal. The frontals and parietals (f., p.) are quite normal; the large squamosal
(sq.) lacks the special jugal spur so well seen in Gallinaceous birds generally. In this respect this bird is
generalized ; so it is also as to the lachrymal (J.), which is very small and attached to the back of the nasal.
In the palate (Pl. VIII. fig. 1) the parostoses are slender and feeble, much like those of the Gallinaceve
generally. This is especially seen in the feeble state and hidden position of the maxillaries (ma), which, as in

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 5d

The pterygoids ( pq.) are also simple, with scarcely any epipterygoid process, and with
the articular plate for the basipterygoid spur far back as in the Tinamou and Ostrich.
That this bird is not a special Gallinaceous form is also seen in the fact that the front
part of the pterygoid does not remain on that bone and become specialized as a
particular peg as in all true Fowls, but becomes segmented off, as in Tinamous and
most Carinate, to form a mesopterygoid, which, however, as in the Carinatz generally,
soon fuses with the palatine. These bones meet right and left, and hide the rostrum
of the sphenoid as in most birds. As we pass from the most special kinds of modern
Fowl to the older types the vomer is seen to become larger, as in the Talegalla ; but
it is always azygous. The vomer (v.) in this bird is large ; in the lst stage it reaches
halfway from the ascending plates of the palatines to the end of the intertrabecula
(Pl. VIII. fig. 1); the hinder pointed end just wedges in between these two plates;
its fore end is thick and bifid. The enlarged split fore end in all three stages is such
as to suggest that this was formed at first, for a day or so, of two club-shaped centres
of bone, and not of a single thread, as in the Gallinaceous types generally. This is
another generalized character, for in most birds that have a large or a wide vomer it is
double at first; but it is single in those in which it is a mere vertical plate or a narrow
median needle ; when double, the fusion takes place very early asa rule. Behind the
palate, but on a higher plane, we see the triple bony tract that forms the parosteal
support of the endocranium. This tract, the parasphenoid, was, as the early stage of
the Fowl shows (Phil. Trans. 1869, pl. 80. fig. 2, 0.¢.), composed of a pair of centres
under the skull-bowl—these are the basitemporals; and of a median bar, the rostral
part, which supports the thick trabecular base of the interorbital septum. ‘These
parts are so generalized that they come in character exactly between those of the
Carinate and of the Ratite. The rostrum is nearly Struthious in respect of the
backward position of the basipterygoids, and the temporary cartilaginous core of each
process was evidently a direct outgrowth from the basis cranii, and not a mere
articular tablet of cartilage developed afterwards, as in the Common Chick (Phil.
Trans. 1869, pl. 83. figs. 1, 13, 14). The basitemporals (0.¢) are larger than in the
Struthionide, and smaller by far than in Fowls and Geese; they are, indeed, very
similar to what we find in birds generally; they are generalized, and not specially
Gallinaceous.

osseous fishes, are mere ossa mystacea. Indeed, the palatine part or maxillo-palatine process (ma.p.) is less
developed than even in typical Fowls and Hemipods, scarcely more than in the Picide, quite unlike what is
seen in the Musophagide, or even in the large Cracide, for these latter have secondary Desmognathism. This
bird is as Schizognathous as a Lizard; it is “‘ Saurognathous.” ‘The cheek-splints, jugal and quadrato-jugal
(j-5 qj-), are rather feeble, but normal; the submedial bars, the palatines (pa.), and the pterygoids (jq.) are of
the simplest kind; the former have no postero-external angle, scarcely any groove in the hind part, but have
an extensive ascending plate.

56 PROF. W. K. PARKER ON THE

‘The superficial or somatopleural elements of the mandible are very stout, but quite
normal; the largest is the dentary (Pl. VII. & Pl. VII. fig. 2, d.); it is closing in upon
Meckel’s cartilage in the 3rd stage, and that rod is beginning to shrink; the splenial
(Pl. VIII. fig. 2, sp., deflected in the figure) is long and thin; the coronoid, as in the
Fowl tribe and some others, is absent ; and the supra-angulare (s.ag.) and the angulare
(ag.) are normal, and are at present quite distinct from, and superficial to, the
articulare (ar.). The skull of the adult is very solid and strong *, and does not suggest
a Musophagine relationship so much as the younger and feebler specimens; both in
the skull and general skeleton it will be necessary for me to give some account of the
parts in the adult, notwithstanding the excellent descriptions already given by Prof.
Huxley, for I wish to make my own monograph complete; moreover my observations
have been made from a somewhat different standpoint. The skull of Opisthocomus is
remarkably short, stout, and broad; the bill, when the mouth is closed, is almost
conical; it is much unlike that of the Peafowl, which is more arched and is very
lightly built. The likeness to that of the Musophagide (e. g. Corythaix buffoni) is
much less in the old male than in the feebler specimens; but the skull of those
Ethiopian Cuculines resembles that of the Cracide, and they seem to have something
Gallinaceous in their nature; they might be called Cracine Cuckoos. The true
Cuckoos of the Neotropical Region (Saurothera, Geococcyx, and Piaya) are the proper
isomorphs, not relatives of the Cracide. In Opisthocomus the rostrum is, measured
in a straight line, two-thirds the length of the cranium proper; the lower edge is 24
millim. in extent, and the culmen, measured along its curve, 30 millim. This part is
hinged on to the cranium at a considerable angle, so that if a line be drawn from the base
of the quadrate to the point of the rostrum, the fore end of the jugal bar would be
8 millim. above that line. This general deflection of the rostrum and the arched form
of the mandibles, whose lower edge rises 5 millim. above a basal line, gives a stunted
and strong appearance to the face of this bird; it is short-faced even for a Fowl. The
likeness of this skull to that of a Touraco (Corythaia) will be illustrated by the
following measurements :—

Width of Least frontal Width across

Hinge. Width. Postorbitals.

millim. millim. millim.
Omsthocomus. .. .. 19 19 26
Corythaiz . . econ” 12°5 23

In the large Crax globicera the skull is twice as long as in the two birds just com-
pared together; its hinge is 20 millim. across, but it is all ankylosed except the

* In the Hunterian Museum there are two skeletons of this bird ; these will be referred to here as A & B.
The mounted specimen (A) is evidently that of an old male bird; the other is an injured specimen, and is not
mounted. Prof. Huxley’s figures (P.Z.8. 1868, pp. 310, 311, figs. 13-16) show feebler birds, and were
probabiy younger specimens or females.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 57

narrow nasal processes of the premaxillaries ; in the other two birds the hinge is perfect.
In Crax globicera the narrowest frontal width is 33 millim., so that relatively, although
broad, it is a much narrower skull than that of Corythaix or Opisthocomus. Yet, with
the exception of Opisthocomus, the skull of this bird is the most Musophagine of all
the Gallinacee. These three types of skull are all holorhinal, but the ossification of
the upper face in the strong skull of Crar is much less than in the light skull of
Corythaix, or in that of Opisthocomus, which is intermediate in this respect '.

In Crax globicera the external nostrils in the macerated skull are large, obliquely
oval spaces, 22 millim. by 12 millim. in size; for the alinasal cartilages that so
largely fill this space are lost in such preparations. In Corythaix this space is largely
closed by the ossification of those cartilages; this osseous change takes place in a large
number of the Coccygomorphe, in very few Coracomorphe (for instance Gymnorhina),
and in no normal member of the Alectoromorphe. In Opisthocomus this ossification
is as complete as in Corythaix; this is very remarkable, and helps to stamp this bird as
aberrant. The actual nasal opening in this bird is obliquely reniform, with the
“hilus” below. In Corythaix it is horizontal, and the hilus of the kidney-shaped
hole is above, the opening being modified by the protrusion of an ogssified valvular
process from the inside. In Opisthocomus the hilus is caused by a process of the
alinasal lamine, where it rests on the descending crus of the nasal. In this bird the
septum nasi is ossified, but its vertical extent is small, and it is fenestrate; it is like
that of Corythaix, but feebler. In Craw globicera, with an unusually solid skull, even
for a Curassow, the lower part only of the septum nasi is ossified, and, as in Accipi-
trine birds and Owls, is ankylosed to the swollen maxillo-palatine; so that, contrary to
rule, this Fowl is desmognathous in a secondary manner (see Huxley, P. Z. S. 1867,
p. 483, fig. 15).

The use of such a taxonomic character as Desmognathism or Schizognathism is
very extensive in some groups and very limited in others; and there is no sharp line of
distinction between the two. The most Lacertian palate for openness is that of the
Woodpecker; the most modified by intense ossification is that of the Toucan; yet
these two types, each specialized to the uttermost, have a postcephalic skeleton, not
indeed identical, but extremely similar. Corythaix also, and its relatives, have an
intensely ossified fore palate; but this abnormal Curassow——Opisthocomus—which
looks at first sight as if it might be a member of the Musophagide, is as schizognathous
as the Woodpecker. The structure of the rest of the palatal part of the face of this
aberrant bird will show how far it is removed, not merely from the Musophagide, but
even from the Cracide. But the Cracide in the New, and the Megapodide in the Old
Tropics, are the most archaic forms of the Fowl tribe ; the latter are very reptilian in

1 My meaning as to light, in contrast with coarsely strong skulls, will be seen at once by anyone who will
compare the skull of a Toucan with that of the Cock of the Woods (Tetrao wrogallus) or that of a domestic

Goose.

58 PROF. W. K. PARKER ON THE

their habits, and in both families the foot is flat. ‘They are Peristeropodous, not high-
heeled as the other true Gallinacew, which are Alectoropodous (see Huxley, P. Z. S.
1868, pp. 294-319). The palatal region of Opisthocomus as a whole would have been a
very difficult study if the Tinamou had not come in for our enlightenment. It is only
by tracing the relation of the latter type as well as of Opisthocomus to the Struthious
forms that we shall find the real clue.

When we come to the organs of flight it will appear evident that both these low
Neotropical Families show good proof that they are much less modified from the ideal
archaic bird than the Ratite, which have manifestly undergone a large amount of
degradation on the one hand and specialization on the other.

In the general palatal view (Huxley, op. cit. p. 311, fig. 16)* the skull of the adult
is more Struthious than in the embryo (Pl. VIII. fig. 1), for the long ascending
processes of the palatines meet at no part, and behind the sharp vomerine wedge the
rostrum of the parasphenoid is exposed. The pterygoids behave in this type as they
do in the Tinamou, in both of which the mesopterygoid is segmented off and unites
with the palatine, as in the majority of birds; in the Ratite there is no segmentation,
and no special modification of the fore part of the pterygoid. In Fowls proper, as in
the Goose tribe, that part becomes a neat peg which rolls in a groove on the upper
and hinder face of the palatine. The slight development of the hinder part of the pala-
tines, and the extreme feebleness of the fore part, are essentially general or Struthious ;
so also is the imperfect development of the maxillo-palatines, and the length and
breadth of the vomer. That bone is not so Struthious as it is in the Tinamou, but is
intermediate in size between the vomer of that bird and that of the Fowls proper, in
which it is, as a rule, a slender azygous style. Its breadth and forked form suggest a
primary division of the bone in Opisthocomus, although, as I have shown, my earliest
embryos have it already in one piece. In my earliest embryo of Struthio camelus
(Phil. Trans. 1866, pl. vii. fig. 4, v.) it was, although very large for a bird, in one piece.
The proper pterygoid segment of the adult is remarkable for its dilatation in front ; it
wants the neatness seen in the higher kinds of birds, and the dilatation, which is great
and reptilian in the Ratite, is nearly equal in this bird to what is seen in the Penguins.
The quadrate is not like that of a typical Fowl, or of a Tinamou, or of a struthious
bird ; in these latter, as a rule, and in the Tinamou, the otic process is oblong and
undivided ; in the Fowl there is a round main condyle and a small secondary head on
its inside. But in the Cracide and Megapodide the head is divided into two condyles,
the outer larger than the inner in Crax, but the two are subequal in TYalegalla; in
Opisthocomus they are subequal as in the latter bird ”.

‘ In this figure the basitemporal is shown as abortively developed, or it was injured on one side; it is
perfectly symmetrical in the Hunterian specimen (A), in which both of the Eustachian tubes are well floored.

* In working out the Apteryx, my son, Prof. T. Jeffery Parker, finds that its quadrate has its otic process
undivided, as in the Carinate generally. This is a remarkable fact, as the whole upper face of that bird is
perhaps more immobile than in any other type in the Class.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 59

The occipital condyles are not so bilobate as in the Cracide. The occipital plane is
rounded and complete, not fenestrate as in many water-birds; in this it agrees with
the Fowl tribe. A comparison of Prof. Huxley’s figures with those given by me of an
old Fowl (Phil. Trans. 1869, pl. Ixxvii. figs. 4-6) will show how much this bird comes
short of the typical Fowl. This is seen also in the postorbital region, where the post-
orbital process of the frontal, the zygomatic process of the squamosal, and the sphenotic
process of the alisphenoid are all aborted in Opisthocomus, but, together, form a bridge
over the temporal muscle in the Common Fowl. The Fowl, like the Goose, has the
posterior angular process of the mandible very long; in Opisthocomus it is almost
suppressed ; it is best seen in the embryo (Pls. VII. & VIII. fig. 2, p.ag.). ‘The
figure of the hyoid arch in the 3rd stage (PI. VIII. fig. 3) shows how near this part
comes to that of a typical Fowl.

IV. The Vertebral Chain of Opisthocomus cristatus.

In the Ist stage the vertebre are all formed but not ossified, and the number of the
caudal series can be determined accurately ; this part has generally six in the adult,
and the last of these is composed of four (Pl. VIIL., ed.v.) very rudimentary segments ;
therefore nine may be given as the number in this region. I have met with no species
of bird in which the individual variability in the number of the vertebra is so great as
in this, as though the tendency to produce races and species had gone no further than
to give rise to unuseful individual modifications. I suspect that this bird bears the
same relation to the early Tertiary types as the Tapir does to the Mammals of that
period—the Palwotheriwm and its congeners. Where the number of vertebra is
greatest, for instance in the Swans, I have found the number of presacrals vary in
different species, but not in different individuals; and, as a rule, the Carinate vary
more in asymmetry of the vertebre that are developed than in the number in the
chain. A very large proportion of the Passerine Order have 14 cervicals, or exactly
twice as many as Mammals. ‘The larger the type the greater the number of vertebree ;
and the largest of the sifting birds—the Swans—have as many as the largest Ratite,
notwithstanding their extremely high position among the aquatic birds. Opisthocomwus,
like the Gallinacez generally, takes a middle place in this respect, halfway between
the Swan and the Humming-bird. In giving a vertebral formula I shall, in this ease,
break up the avian sacrum into four secondary regions; namely, the dorso-sacral,
lumbo-sacral, sacral proper, and uro-sacral. The dorsals proper are those that have
perfect rib-cinctures, are in front of the pre-ilia, and as a rule are distinct from the
first that is overlapped by those plates. The following table will show the variability
of this chain of bones in this bird.

VOL. XIII.—PART 11. No. 3.—April, 1891. K

60 PROF. W. K. PARKER ON THE

C.? D DS LS. 5 U.S. Cd
listhisbagey | Metiaaticccls 18 4 2 3 3 6 6+3
Drivel FE) gaocodacedoo 18 4 1 3 3 6 6+3
Sitel GIG) gooocdedco00 18 4 1 3 3 6 643
Hunterian adult........ 19 3 2 3 4 6 643
Hixley/svadulit errr lstare 19 3 2 3 4 6 5+4

So that in the Ist stage the total number in the chain is 45; in the 2nd and 3rd
stages 44; in the Hunterian and Huxley’s adult specimens 46. But this does not
exhaust the variability of the axial skeleton with its inferior arches arrested or
developed. In the 1st stage (Pl. VII.) the last two cervicals have considerable styloid
ribs ; it has four complete rib-cinctures ; and the second dorso-sacral has ribs with a
half-developed sternal piece. Only those ribs that are complete below have the
appendage (c¢.a.), an oblongo-oval cartilage, not diverging backwards from the rib, but
parallel with it. The eight more or less developed ribs are ossifying rapidly; these
are the only axial parts that are not entirely cartilaginous at present. Behind the
developed ribs of the second dorso-sacral there is, in this first stage, a pair of small
rudiments (Pl. IX. fig. 4, s.7°.) which are losing their individuality already, and will
only appear as part of the transverse processes later on.

There are two pairs of similar riblets in the fore part of the uro-sacral series (Pl. IX.
fig. 4, s.r.) ; behind these, again, there are only upper transverse processes (diapophyses) ;
these parts are very uniform, the two pairs of uro-sacral riblets being constant in this
species.

In the 2nd stage the last two cervicals have elongated or styloid ribs, and in two
specimens at this stage the second dorso-sacral had only small rudiments of ribs; but
there were five perfect thoracic cinctures. It is seen at once that there is a vertebra
wanting in the dorso-sacral region in the 3rd stage as compared with the Ist (Pl. IX.
figs. 4 & 6): the 2nd agrees with the 3rd stage in this respect; in both specimens of
the 2nd stage there is a small rib right and left on the first lumbo-sacral vertebra, and
also in the 3rd stage (Pl. IX. fig. 6, s.7°.); this shows how arbitrary is our classifica-
tion of the regions of the chain. The cervical vertebra in the Ist (Pl. VII.) are largest
and strongest in front; behind the atlas (at.) four of these have more or less develop-
ment of the upper and lower spines; there is some downward development also of the
atlas. The cartilage, quite umossified, is very solid, and, examined in horizontal
sections, it is seen that the notochord (Pl. VIII. fig. 7, ne.) is submoniliform, so that

* C. Cervical; D. Dorsal; D.S. Dorso-sacral ; L.S. Lumbo-sacral; S. Sacral; U.S. Uro-sacral; Cd. Caudal.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 61

there is a tendency to produce three times as many vertebre as are needed. There
are two constrictions in each centrum in all the presacral vertebra, besides that which
afterwards is formed at each intervertebral space. Originally the constriction was, as
in fishes, in the middle of each rudimentary centrum (see my paper on the Vertebral
Chain in Birds, Proc. Roy. Soc. March 8, 1888, pp. 465-482). This bird agrees with
the Fowls and the majority of the Carinate in having no rudimentary rib on the atlas
and axis (Pl. VIL., at.,aa.): the rest of the cervicals down to the last two have a
rudiment, right and left: the distinctness of these riblets as cartilage is very temporary :
they complete the canal for the vertebral artery so far asit extends. These riblets have
but a small styloid process in this bird, and in the hinder part of the neck this is lost ;
but the ribs break out suddenly in the seventeenth vertebra (Pl. VII., ¢.r.”). ‘Those
hinder cervicals begin to have an upper spine, and these spines are like those of the
dorsals from the fifteenth to the eighteenth. There is scarcely any development of the
infero-lateral edge of the cervicals in this bird, tending to protect the carotid artery
below.

The centra of all this series, and of all the dorsals also, are very broad, and all these
vertebre are cylindroidal. The interarticular ligament is normal, being perforated in
the middle, the notochord passing through as the ‘“‘ suspensory ligament.” Ossification
is advancing fast in the 3rd stage, but the centres are not all present; when complete
there is a pair for the neural arch, one for the centrum, and a pair for the riblets. In
the axis there are two axial centres, for the centrum of the atlas is fused with it; the
so-called centrum of the atlas is an intercentrum belonging to the junction of the
first vertebra to the occipital condyle. Behind this another intercentrum appears
belonging to the junction of the atlas with the axis, so that the atlas has three azygous
osseous centres. The dorsal vertebrae, with their large quadrate upper spines, show
scarcely any outgrowth below, but are unusually broad at that part; their transverse
processes are large; the elevated cup for the capitulum of the rib is normal. Seen
from below (Pl. IX. fig. 4) the sacral series is spindle-shaped ; in the middle the
centra are developed to a great width, for these contain the dilated, hollow sacral part
of the myelon. ‘There is a synovial cavity, with an interarticular ligament, between the
first of the series and the last dorsal; for the rest, the cartilaginous centra are fused
together, fibrous tissue only appearing between these parts and the hinder half of the
uro-sacral region. The transverse lines marking the junction of the centra are curved
backwards; the notochord (nc.) is obscurely moniliform, and there is only one
constriction in each centrum. In Stage | (Pl. IX. fig. 5) the spines are confluent, and
die out on the last lumbo-sacral, to reappear in the middle of the uro-sacral series.

The ribs, developed on the first and second, the dorso-sacrals, appear as small
remnants on the first lumbo-sacral and the first and second uro-sacrals. ‘The diapo-
physes are present throughout the series; they form strong buttresses to the pre-ilia,
become very small in the last sacral proper, and then gradually approximate to the

K2

62 PROF. W. K. PARKER ON THE

condition seen in the free caudal series. The first of that series (Pl. IX. fig. 4, cd.v.’,
ic.) sends an intercentrum under the fourteenth or last sacral; the rest up to the fore-
part of the uropygial tract have increasingly large intercentra traced from before
backwards; in the Common Fowl these parts are suppressed. Above (Pl. VII. &
Pl. IX. fig. 5) the neural arches are seen to be narrow, and the spines low and blunt,
in these free vertebre. The short uropygial series of four imperfect segments formed
on the end of the notochord is pinched in between the third and fourth, and the last .
segment expands a little. ‘The ossification of the sacral series, as seen in the 3rd stage,
is by a pair of centres for each arch and one for each centrum (Pl. IX. fig. 6); there
is in many young birds the appearance of a double bony centre in this part; it is
really single, the osseous matter which is first formed round the notochord growing
out as a right and left lobe into the surrounding cartilage. The ossification of the
caudal series is like that of the uro-sacrals in front, but becomes simpler behind; the
intercentra are separately ossified. The vertebral chain of the adult may be profitably
compared with that of some other types; in the following formule the avian sacrum
is taken as one region :—

CoRYTHAIX BUFFONT.

C. 14, the two last with free ribs; D. 5, all with arches perfect; S. 13, first with
half-sized free ribs, altogether, four pairs of pre-iliac buttresses, and the rudiments of
a fifth, the ninth or first uro-sacral has riblets; Cd. 7+3: Total 42.

OPISTHOCOMUS CRISTATUS.

C. 19, last two with free ribs, sixteenth and seventeenth with no capitular develop-
ment of fused riblets; D. 3; S. 15, of which five have pre-iliac buttresses, the first
two of these with free ribs, the second of which have sternal pieces incomplete below,
the tenth and eleventh or first and second uro-sacrals have riblets; Cd. 5+4:
Total 46.

CHAUNA CHAVARIA.

C. 19, last three with free ribs; D, 5, ribs devoid of appendages; S. 17, with eight
pairs of pre-iliac buttresses, the first three of these with developed ribs and sternal
pieces, the twelfth and thirteenth and probably the fourteenth had distinct riblets ;
Cd. 645: Total 52.

CRAX GLOBICERA.

C. 16, two last with free ribs; D. 4; S. 16, the first six with pre-iliac buttresses, of
these the first carries perfect ribs, then two follow with no lateral processes, the next
two have strong diapophyses and feeble riblets ; three sacrals follow with these parts

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 63

aborted, and of the eight uro-sacrals the first two have riblets and diapophyses;
Cd. 6+3: Total 45.

This remark may be made, namely that the same bird differs very much in youth
and age; small riblets are apt to lose their distinctness or to become starved ; while
the diapophyses themselves often shrink, so that a strong buttress may become a small
prickle, which often quite disappears on one side. The length of the pre-iliac region
of the sacrum in Chauna is a Cygnine character; indeed it has one more vertebra in
the dorso-lumbo-sacral series than the Swan (Cygnus olor). Opisthocomus agrees with
that Neotropical generalized Chenomorph—Chauna—in the length of its neck, a length
which attains its highest condition in the Swan, which has 25 cervical vertebra ; in
the other regions Opisthocomus comes near to the Gallinacez generally. The reniform
occipital condyle in this bird fits into a notched atlantal cup as in the Fowls. The
neural spines begin on the second vertebra, die out on the eighth, reappear on the
thirteenth and fourteenth, and become complete on the seventeenth.

The axis and the two next have a small, thick inferior spine, which is nearly obsolete
on the fifth. ‘There is a small inferior spine on the fifteenth and sixteenth, and there
is a free median spine with a pair of lateral ridges on the last three cervicals. The
dorsals are peculiar ; they are cylindroidal in their articulation, like the cervicals, and
are flat-bottomed. There are no carotid bridges below in the cervical region. ‘The
first two or rib-bearing sacrals (dorso-sacrals) are somewhat bicarinate ; the rest have
merely the usual convex form. Returning to the cervical region there is a structure
which is very remarkable, and yet not rare in the Carinate. The seventh to the
twelfth inclusive have an oblique bony lamina outside the canal for the vertebral
artery, a kind of flying buttress. In Crax globicera this is seen on the sixth to the
ninth inclusive ; this structure has its greatest development in certain of the Coccygo-
morphe. ‘The cervical riblets are small, but have a large base between the diapophyses
and the parapophyses ; they only form small additions to the former processes in the
sixteenth and seventeenth ; altogether the neck is very gallinaceous except as to its
length. The sacral region is intensely ossified; the spine never quite dies out even in
the true sacrals. Gradually the intervertebral spaces appear behind; but the main
part of the pelvis above is largely plastered over with periosteal bone. ‘The caudal
vertebra are not pneumatic, the rest of the spine is; the transverse and upper processes
of the caudals are thick and of moderate size; the centra have the usual development.
The intercentra are developed all along: that of the first is fused with the last sacral
centrum ; the second is a small grain of bone; the rest form thick inferior spines to
the vertebre ; this is seen also on the uropygial bone, which is formed of four segments
in the embryo (Pl. IX. fig. 5); it is thick below as in Craw; it is somewhat hooked
downwards at the tip. The ribs in the old bird are strong bars ; those on the
eighteenth cervical are 24 millim. long, and have no uncinate processes; those on the

64 PROF. W. K. PARKER ON THE

nineteenth are 55 millim. long, and those on the first dorsal 57 millims. These two
pairs of ribs have evident uncinate processes, whilst those on the second and third dorsals
are rudimentary, only widening the ribs somewhat. In the first stage the free cervicals
have no appendages (Pl. VII.), but these are seen on the four dorsals and the first
dorso-sacral. The ribs gradually narrow backwards; each rib also narrows downwards
and becomes the width of the sternal piece; these lower segments are short and strong.
Even the costal appendages show that this is a generalized bird, both in their form
and in their variability.

V. The Sternum and Shoulder-girdle of Opisthocomus.

From a very early period of my study of the Anatomy of Birds, I have been in the
habit of comparing these intensely-specialized forms with the Imago Insect; the
Reptiles proper being looked at as a sort of active Pupa, and the lower forms of Fishes
as representatives of the larval stage of those noble Invertebrates.

Nothing shows this relation better than the shoulder-girdle; for in this part the
intense life of these hot-blooded creatures has transformed the old elements, creating
them anew, as it were, ‘into something rich and strange,” making of those simple
limb-roots the proximal part of the highest type of an organ of flight. ‘The tooth-
bearing extinct forms (Archwopteryx, Hesperornis, Ichthyornis) throw a much fainter
light on this great change than this isolated archaic Fowl. The lowest Mammals, the
oviparous Monotremes, still retain the simple or unfused condition of the splint-bones
of the shoulder-girdle. ‘The existing Reptiles, especially Lizards, show these splints in
their free state, and also a large generalized shoulder-plate belonging to the endo-
skeleton, although its most superficial part; for it is formed between the ribs and the
skin. Both the ribs and shoulder-plates are formed in the outer layer of the body of
the embryo, namely in the somatopleure’.

The sternum in the Ist stage (Pl. LX. figs. 1, 2) is already perfectly formed in
hyaline cartilage ; even in the 3rd stage no ossification has appeared. The breadth of
this short sternum is five-sixths of its length in the fore half; it narrows in behind the
last sternal rib, and then widens out again so as to be 1 millim. wider than in front.
The form now attained in this half-ripe chick of Opisthocomus is fairly comparable to
that which is seen in the sternum of the Common Chick on the sixth or seventh day
(Lindsay, pl. xlv. figs. 1-3), and is relatively shorter than in the adult (Huxley, op. cit.
p- 306, fig. 8). Otherwise, except in its histological condition and its size, the sternum
undergoes but little further change. The rostral process (7.s¢.) is small, rounded, and
superior; it grows forwards between the upper lips of the coracoid grooves.

* The copious illustrations generously allowed me by the Council of the Ray Society in my memoir on the
Shoulder-girdle and the Sternum, published in 1868, come in very useful now, to illustrate this, my newer

work. I am also greatly indebted to Miss Beatrice Lindsay for her excellent paper “On the Avian Sternum ”
(P.Z.8, June 16, 1885, pp. 684-716, pls. xlii—xly.).

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 65

In the Common Fow] this part is very large and deep and is fenestrate, so that the
coracoid grooves meet in the middle; on the sixth day of incubation (Lindsay, op. cit,
pl. xlv. fig. 3) the top of the sternum projects in front but little beyond the bottom
where the keel is beginning to form. On the seventh day my figures (soon to appear
in the Trans. Linn. Soc.) show an imperforate rudiment of the rostrum and a definite
keel, projecting forwards in front, not far behind the rudiment of the rostrum. In this
half-ripe Ist stage of Opisthocomus the fore edge of the feeble keel is exactly at the
middle of the sternum, and does not reach the end; it is behind the middle if the
rostrum is taken into account (Pl. VII., Pl. IX. fig. 1, s¢.é.).

This bird, both in this stage and in the adult (see Huxley, op. cit. p. 306, figs. 8, 9),
is absolutely unique; the Turkey (Meleagris gallopavo) comes nearest it in this respect
as far as I have seen.

Thus in Opisthocomus the space between the Bird and the Lizard is partly bridged
over, the endoskeletal sternum receiving considerable support from the exoskeletal
interclavicle, as I shall soon show. Nevertheless, the extension backwards of the
sternum, with its spreading metasternal outgrowths and its intermuscular keel, makes
even this sternum something very different from that of a Lizard ; it becomes still more
unlike in old age (compare Pl. IX. fig. 1 with Huxley’s figure, and with the sternum
of various Lacertilia in my memoir, pls. ix., x., xi., xiil.). I see a manifestly archaic
character in the very forward setting on of the lst sternal rib (compare my figures with
Miss Lindsay’s, op. cit. pl. xlv. fig. 2); there is scarcely any precostal process (pe.p.).
The articulation of the five pairs of sternal ribs (s¢.7.) is already complete ; the 3rd is
on the most projecting part of the costal margin; these segments are unossified at
present and are quite normal. The parts which are developed correlatively with that
of the keel for special avian muscular attachment are such as are seen ina large number
of existing Carinate birds. ‘These metasternal processes (m.s¢.) are, with the keel,
‘‘ additions of later phylogenetic date” than the costal margins (Lindsay, op. cit. p. 710,
fig. 4). In this embryo, and in the adult figured by Prof. Huxley, the outer and inter-
mediate metasternal (or xiphoid) processes are only partially divided by a small, oval
fenestra. But in the mounted Hunterian specimen (A) this part is notched by a gap
5 millim. deep and 5 millim. wide; the inner notch is nearly of the same size, but is
more angular. These modifications of the metasternum in individuals is interesting in
this type; so, also, is the want of symmetry in the two moieties of the sternum, as is
seen in the median process of this first stage (Pl. IX. fig. 1, m.st.). I have no stage
showing the mode of ossification of the sternum; but I strongly suspect that it takes
place by distinct ectostoses, as in the Ratite, the Hemipods, and the Fowls?.

‘ In my work on the Shoulder-girdle and Sternum (pls. xvi., xvii.) I have detected an error in my description
and figures of the sternum of Turnia (p. 184, pl. xvi. figs. 13, 14), The so-called coracosteon was formed by
accident; the preparation is still by me, and I see now that what appeared to be a sutwre is merely a

fracture.

66 PROF. W. K. PARKER ON THE

The instance of the tooth-bearing Jchthyornis, and also that of the existing
Tinamous, show that the keelless sternum is a comparatively recent secondary modifi-
cation of this part of the skeleton. The dying-out of the keel has manifestly taken
place as part of the same withering of unused structures as that which is seen in the
wing of absolutely terrestrial birds. These two groups of Pre-Ratite—the Opistho-
comide and the Tinamide—probably co-existed with strong, if short-winged, ancestors
of the existing struthious birds. It is very remarkable that in the first of these the keel
is at its lowest development, apparently in a primary condition; whilst in the Tinamous
the sternum is one of the longest and most carinate of any kind of bird. The shoulder-
girdle of Opisthocomus is still more remarkable than its sternum. In the Vertebrata
generally, this part is of great morphological interest, for in it, as in the skull, the
cartilaginous tracts, ossified or unossified, are supplemented by parostoses or superficial
membrane-bones. It is an awkward necessity of this branch of science that its termino-
logy is dominated by the terms of Human Anatomy, which, when applied to parts of
lower and simpler forms, are often either incorrect or absurd. Thus the term “ coracoid ”
for the lower part of the shoulder-plate is absurd—the only coracoid which is like a
Crow’s beak is that of Man and a few Mammals; whilst the term “clavicle” is
incorrect as applied to a lateral parostosis of the shoulder-girdle in most of the cold-
blooded types, for they have a simple membrane-bone, whereas the clavicle in Man
and his nearest relatives is a compound structure, in which the parostosis 1s soon
blended with eudoskeletal elements. This is due to the hot condition of the blood in
Mammals, and the same thing appears in most birds—that is, in those that are furthest
removed from the Amphibia and Reptilia. Parts of the endoskeleton that are con-
tinuous or unsegmented in the cold-blooded types are variously segmented and
abortively developed in the nobler Vertebrata; and thus we meet with vestiges or
remnants of archaic structures that are used up in many ways in the metamorphosis of
the skeletal elements. In the existing Ratite, and in this ancient Carinate type, the
parts of the shoulder-girdle are in a very primitive condition; in Opisthocomus the
transformation that takes place is mainly due to arrested growth and to the blending
of parts originally separate. At first sight the structure of these parts in the adult
does not seem to be different from that of ordinary Carinate birds; the scapula forms
a single and complete bone, and so does the coracoid, and they are bent upon each
other at an angle less than a right angle, as in flying birds generally.

In the Ratitz the axes of these two bones are coincident ; that, however, is a relapse
into a degraded condition ; that this was not the case in their ancestors I feel quite
certain. In this typically bent and typically narrow shoulder-plate there is, in the
half-ripe embryo, a character as unsuspected as it is instructive: the supra-scapula is
segmented from the scapula, as the latter is from the coracoid. Even in Man the
“posterior edge ” of the scapula is feebly ossified, for a long while at least, thus making
a suprascapula ; but in all known adult birds, extinct or existing, the scapula is ossified

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 67

throughout by one bony ectostosis. Down below, among cartilaginous and semi-
cartilaginous fishes, ¢.g. the Skate and the Sturgeon, the suprascapula is a distinct
segment of cartilage (op. cit. pl. i.). In the Amphibia Urodela this part is not seg-
mented off, but it remains unossified (ibd. pls. iii., iv.); in the Anura, however (did.
pls. v.—vil.), the suprascapula is semi-segmented from the scapula and has its own
ectostosis and endostosis. In the Lacertilia (77d. pls. viii.—xi.) the suprascapula has
its own endosteal tract, but no outer plate of bone; moreover, the cartilage between
the two regions is not so much altered as in the Anura, and therefore the supra-
scapula cannot be bent upon the scapula to the same extent as in Frogs and Toads.
Here, again, the attempt to reconstruct the ancestral bird or birds has to be done in
the light of Amphibian Morphology rather than in that of the Reptilia. The supra-
scapular segment in my Ist stage (PI. TX. fig. 1, s.sc.) is most distinct on the right
side; it is one third the length of the scapula (sc.), is rounded above, uncinate, and
dilated at its base, which enclosed the narrow top of the scapula; that top is as yet
unossified as the suprascapula itself.

The main scapular segment has the usual gently-bent shape, and is dilated below to
form the upper part of the glenoid cavity and the very short acromial process (ac.p.).
The coracoid (c7.) is much wider than the scapula, but it is only three-fifths the length
of the whole bar; its head, which is swollen and large, and its dilated and uncinate
base, or epicoracoid region, is as yet unossified; the rounded and narrow shaft is half
the length of the whole bar. The shaft-bone ultimately ossifies the whole of the
coracoid in this as in all other birds; they have, in their adult state, no semi-carti-
laginous tract below, such as is seen in Anura and Lacertilia, nor any separate epi-
coracoid bone such as we find in the Monotremes *.

Seen from above (Pl. IX. fig. 2) there is a large uncinate flap of cartilage which
grows inwards and hooks downwards upon the upper third of the furcular ramus or
clavicle ; this is a continuous precoracoid, and this part, in a very diminished condition,
forms the so-called “ clavicular process of the coracoid;” it is the precoracoid of
Sabatier (Lindsay, op. cit. p. 715). ‘This part is always present, in the embryo at least,
in the Ratitee (op. cit. pl. xvii. p.cr.) ?.

In the African Ostrich it is as well developed as in the Anura and Chelonia (op. cit.
pls. v.—vii., xii.—xvli.). One thing more has to be noticed: this large, continuous, pre-

* In her otherwise very accurate memoir, Miss Lindsay has made a curious mistake with an unlooked-for
misstatement of my views. She gives a figure (pl. xliy. fig. 1) of the coracoid of Diomedea exulans, and calls
the rough top or head of the bone the “ coracoid epiphysis” (cor.ep.), and states that this is “ Parker’s pre-
coracoid.” It is, at most, an apophysis; birds have very few epiphyses; generally, the only one is in the
‘¢ procnemial process ” of the tibia. Parker’s precoracoid of birds (see ‘ Shoulder-girdle and Sternum,’ pl. xiii.
p.cr.) is a special segment or remnant of the fore margin of the great continuous and fenestrated shoulder-
plate of a cold-blooded type, a part enucleated in birds from the general precoracoidal bar.

* My son, Prof. T. J. Parker, finds it well developed in some embryos of the Apteryx.

VOL. XU1.—Part ul. No. 4.—April, 1891. L

68 PROF. W. K. PARKER ON THE

coracoidal flap lies over the clavicle ; the often massive segment, which is manifestly part
of the original bar of an archaic type, lies below or behind the clavicle (see in Phala-
crocorax and Sula, op. cit. pl. xiii. figs. 8-10, p.cr.), and forms, after separate ossification
and fusion with the clavicle, the flat-topped shoulder of the furcular ramus in those
types. This addition to the clavicle articulates with the head of the coracoid, under-
propping it, and this is the part which in birds I have constantly called the pre-
coracoid 1,

But there is in some cases another nucleus of cartilage at the top of the clavicle;
this I have called the “ meso-scapular segment” (op. cit. pl. xv. figs. 12-15, m.se.s.),
for it is manifestly a segment from the meso-scapular or acromial region of the shoulder-
plate. It gives rise to the enlargement of the upper and posterior lobe of the wide-
topped furcular ramus of all true Passerine birds and of some of the Coccygomorphe,
for instance Rhamphastos, Picus, Alcedo, &c. I mention these specializations in the
most specialized kinds of birds to show how low is the ornithic level of Opisthocomus ;
its furcula will reveal this in a still clearer manner. The furcula of Opisthocomus is
composed of three parosteal bones, two clavicles and an interclavicle; and notwith-
standing the figures and descriptions given by me long ago, in my large memoir, of a
distinct interclavicle in Birds, answering to the long bone of Lizards, Prof. Huxley (op.
cit. p. 307) has been careful not to use the term, but calls the stem of the Y-shaped
merrythought the “ hypocleidium.” A younger biological sceptic, Miss Lindsay, also
doubts my interpretation. She says, speaking of the chick of the 8th day:—‘‘ The
ribs at any rate have established their generic characters at this date, which renders
it probable that the broadening ossification of the median region of the clavicles,
described by Gotte as established during the 8th day, is an outgrowth of the Avian
furcula rather than a pre-Avian interclavicle—a view which is expressed by giving to it
(as has been done throughout this paper) the name of median furcular apophysis rather
than of interclavicle” (op. cit. p. 702).

Opisthocomus has, of all birds, the most perfect Y-shaped furcula (Pl. IX. figs. 1-3,
cli.cl.); its forks and stem are of equal length.

The forks or rami are somewhat sigmoid, are narrow, and bent outwards at the top ;
they suddenly flatten out and then gently become narrower, the lower part being only
half as wide as the upper. The two clavicles not only close in the semi-oval space,
they also run down the stem for half its length, lying close together above, and then
becoming slightly separate. Into this space there fits a fine needle of bone, the
interclavicle ; it is two-thirds the length of the stem ; thus the hypocleidium is composed
of three elements. This interclavicle binds the furcula to the lower face of the sternum,
as in the Lizard (op. c7é. pls. vili.—xi.); it can be seen through the transparent cartilage
in this first stage. In Lizards it often gets in between the right and left halves of the

* See op. cit. p. 150. There I remark that “ the line of segmentation between the coracoid and the great
precoracoid segment has become a large oval, gliding, synovial joint.”

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 69

sternum, and appears on the upper surface (see in Lemanctus, op. cit. pl. ix. fig. 9, 7.cl.).
A delicate ligament, half the length of the bony style, connects it with the arrested
sternal keel. In the adult bird these three elements are fused into the single furcula,
and this bone becomes ankylosed above with the coracoids, and below with the sternum,
both rostrum and body; the scapula only remains distinct.

This arboreal bird has evidently only developed its organs of flight for flitting; the
great amount of ankylosis in its shoulder-girdle suggests this. In my notes on these
parts in the Hunterian specimen (old male) I have remarked that “the ankylosis has
not obliterated the lines of junction of the large clavicular process of the coracoid with
the clavicle, nor of the head of the latter with the acromial process of the scapula, for
that bone remains free whilst the others are all melted together—clavicles, interclavicles,
coracoids, and sternum.”

VI. The Wings of Opisthocomus cristatus.

In the Ist stage (Pl. X. figs. 1-3) the wing is as much developed as in my 5th stage
of the Common Fowl (Phil. Trans. 1888, B, pl. 62. fig. 7)’.

It is necessary for me to refer to those figures, for in them is shown what I find to
be constant in all the birds I have studied at that stage, namely that there are at first
only two proximal carpals. In the paper referred to I have called these the “ inter-
medio-radiale ” and the “centralo-ulnare ;”’ in the adult bird the ordinary nomenclature
and ‘“‘ulnare.” It is necessary to remember that

?

for these carpals is simply “ radiale’
in the strangely adaptive abortion of many elements of the bird’s fore limb, some parts
are primarily, and some secondarily developed, the latter by segmentation of a
cartilaginous mass; and this is often very temporary, the parts becoming fused
together again, even, in some cases, before the time of hatching.

The wing of Opisthocomus, although very much like that of a Fowl in its adult
state, yet differs much from it in its development. The long cartilages in the first
stage lave already a considerable ectosteal sheath (Pls. VII., X. figs. 1-3). As in the
true Fowls, the humerus and cubitus differ but little in length; they are quite normal,
and need not be described in this stage. The manus, however, shows many parts that
are lost and leave no sign in the adult; all these unlooked-for segments are, never-
theless, quite normal parts of a typical cheiropterygium. The two nuclei of cartilage
seen in my early stages of the chick attached to the radius and ulna are here seen as
five more or less distinct tracts (Pl. X. figs. 1, 2). The radiale (re.) is now a pedate
wedge of cartilage lying on the fore margin of the wrist, and most to the flexor or
inner side. The intermedium (¢.) is a four-sided short wedge, somewhat lesser than the

‘1 In figs. 1 & 2, 1st and 2nd stages in that paper, the distal carpal (d.c.') is drawn too near the first meta-
carpal (m.c.’). In fig. 3 the true position is given, namely inside the head of the great metacarpal. The first
distal carpal is always developed in birds as a free nucleus of cartilage in that position; it is a feeble element,
displaced to the flexor side of the limb.

L2

70 PROF. W. K. PARKER ON THE

radiale: it lies on the outer or extensor side of the wrist. On the ulnar side of
the wrist, instead of the thick, two-lobed U-shaped carpal so familiar to us in all adult
birds, there is, on the flexor side, a pear-shaped carpal (we.), the ulnare proper, which
sends its narrow end forwards to articulate with the third metacarpal and its distal
carpal (me.’, d.c.°), and which overlaps at its broad end the bulbous distal end of the
ulna (w.). The shorter thicker crus of the bilobate ulnare of the adult is here repre-
sented by a subcrescentic mass which, behind, is attached to the ulna and free ulnare,
and in front by ligament to the second and third distal carpals (d.c.?, d.c.°) at their
junction, and also to the postero-distal angle of the intermedium (7.). A transverse
chink on the inner face of the mass partially subdivides this centrale into an anterior
and a posterior segment (c., ¢.!). These are all the proximal carpals I find in this
bird; the distal carpals are three in number, and are constant throughout the
Carinate, corresponding to the three constantly developed digits. The second and
third distal carpals (d.c.’, d.c.*) are formed from one mass in the early embryo, but the
hinder and lesser nucleus is soon detached, and then coalesces again with the larger
piece (op. cit. pl. vi. figs. 1-4, d.c.?, d.c.*). The line of segmentation of these two
carpals, at a stage a little earlier than this, is indicated by the direction of the cells of
the cartilage, which, in the two regions, curve from each other back to back, thus
leaving a fine clear tract between the groups of cells. This is soon obliterated, so
rapid is the metamorphosis of the parts in these hot-blooded, fast-growing types. The
intense growth of the second carpal, metacarpal, and digit, which form the main part
of the framework, for the insertion of the primary quills, has dominated the whole
manus, and has thrown the elements of the limb out of gear. The third distal carpal
lies along the ulnar or hinder side of the second, and is as much lessened as the meta-
carpal (mc.") that belongs to it; the two together form a remarkable grooved trochlea
on which the proximal carpals roll. The latter are more tightly strapped to the
radius and ulna, and here, as in the hind limb, the main movement is between the
proximal and distal row of segments. Were there fusion of the proximal carpals with
the radius, and did the ulna end in a point, dissimilarity would be complete. Of what
service the “alula” is to the bird in flight it is difficult to say; it cannot be of much,
as it gives but a slight increase of size to the wing, and that only near the carpal bend.
Certain it is that the larger the manus the lesser is the pollex, as we see in the
“¢ Macrochires,” the Swifts, and Humming-birds. But here, in this Reptilian bird, the
pollex is relatively nearly as large as in Man (PI. X. figs. 1-7); that is, it is so in the
embryo, not in the adult (figs. 8, 9). The condition of the first distal carpal is very
remarkable; it is a small limpet-shaped cartilage, attached to the head of the first
metacarpal on its flexor face (Pl. X. figs. 2, 3, 7, d.c.1). I can give no other inter-
pretation to Dr. R. W. Shufeldt’s “ pentostium ” }.

* See his “ Osteology of the North-American Tetraonidze,” Geological Survey of the Territories, Washington
Oct. 14, 1882, pl. vil. fig. 59, 8.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. (i

In this bird this nucleus of cartilage is further from the metacarpal of the
pollex (m.c.1) than in most birds; I find it nearest to that segment, to which
I am satisfied it belongs, in the chick of Turnix rostrata’. In that type the
ventrally displaced distal carpal is phalangiform, and is of the same breadth
as the first metacarpal, but only two thirds as long; it is bulbous at its radial
end, and is immovably articulated with its own metacarpal, which is hollowed to
receive it. My doubts about the nature of this nucleus are still further dissipated
when I consider the manner in which my own pollex is ventrally displaced adaptively.
This little nucleus (Pl. X. figs. 2, 3, 7, d.c.') is, 1 doubt not, the counterpart of the
human “trapezium.” The first metacarpal (Pl. X. figs. 1, 2, mc.’) is nearly as wide as
the second, but is only one fourth its length; it has already, in its cartilaginous
condition, become fused with the second for a short space proximally, in a manner
similar to the fusion of the third distal carpal with the same large dominating segment.
The “ trochanter” projecting from the short, confluent metacarpal, shows some signs of
a marginal addition (see Fowl’s Wing, pls. 62-54, me.”); this is the part from which
the main spur grows in certain birds. The proximal phalanx of the pollex (dq.") is
three-fourths the thickness, and two-thirds the length, of the huge metacarpal joint
behind it (me.”); for half its length in the middle it is ossified ; its metacarpal does not
ossify until near the time of hatching. The distal or ungual phalanx is relatively
larger than the proximal; it is very little less than that of the second digit (dg.’); that
measured with the claw on is 2°5 millim. long ; this, of the pollex, is 2 millim. ; the tip
of this distal joint is beginning to ossify.

If the manus in this stage be measured from the top of the second distal carpal, it
will be found that the pollex is more than half the length of the huge index (dq.’,
dg.’); the one is 6°5 millim. long, the other 12°5 millim. On the other side of the
Class, in the Macrochires, in the adult state, the abortion of the pollex, with its small
quills forming the alula, is very remarkable, as the following instances will show :—

Length of the Manus Length of the Manus
along the Pollea. along the Indew.
Cypselus affinis. . . 9 millim. 34 millim.
Chetura caudacuta . 15 BS 68 eu
opacanpellan wee anes) Aye. 21 B
JGGOTONG GUYS. 5 3 BO 34g ONmecs

Thus, instead of the first finger being more than half the length of the second, this
latter is, in these cases, taking the average, four and a half times as long as the aborted

1 From Formosa, the gift to me, many years since, of the late Consul Swinhoe. The skull of one of these
valuable chicks has already been described by me in these Transactions (vol. ix. pl. liy.); the sternum and
shoulder-girdle were also described in my work on those parts, and figures and descriptions of the remainder
of the skeleton will soon be published.

72 PROF. W. K. PARKER ON THE

first digit; the disproportion is greater in the Humming-birds than in the Swifts. I
must now notice a remarkable structure. The manner in which the ectosteal sheath
of the second metacarpal passes along the ulnar side of the first (Pl. X. figs. 1-7,
mc.2, mc.!) is repeated on a smaller scale in the pollex, while at the radial margin of
the top of the proximal phalanx there is a rounded flap of cartilage which runs
forwards in a sharp wedge ending in ligamentous fibres at the middle of the joint.
Besides this, in all these embryonic stages there is a wedge-shaped mass of inter-
articular cartilage (é.a.c.), which, solid and definite as it is, only remains for a time; in
the adult (figs. 8, 9, dg.) it is reduced to a mere film. That these structures belong to
a veritable “ prepollex” I have no doubt; they are not the only cartilaginous tracts
on the radial side of the manus to be seen in Birds, as I shall show in another paper.
The index (dg.’) is greatly overgrown ; but for that the pollex would be seen to be quite
normally developed as compared with that of an average Reptile. The phalanges have
undergone the same degree of ossification as those of the pollex, and moreover its
metacarpal (me.") is ossified in an equal degree ; its distal phalanx has also the tip
ossified even in the Ist stage, as in the pollex. The metacarpus is twice as long as
the proximal phalanx, and it is one sixth longer than the second and third. That
third phalanx is generally lost, often suppressed, in Birds; here it has its most
remarkable development, and is but little less than that of the second digit in the foot
(Pl. VIL). This digit is not simple; it has been complicated by intercalary or
secondary structures; which, however, are not peculiar to this bird, but are very
common in-the Carinate. These are, first, a tract of cartilage on the extensor face of
the wrist, which creeps down between the head of the second and third metacarpals
(Pl. X. figs. 1, 6, me.”); this breaks out at the top of the interosseous space into a
bead-like swelling, which in Gallinaceous birds (op. cit. pls. 62-64, mc.”), in Passerines,
and in some of their Cuculine allies, grows into a large flat plate, which bridges
over the proximal end of that space. Below, on the ulnar edge of the proximal
phalanx of this great digit (dq.), there is a lanceolate tract of newer cartilage that
receives its bony matter from the phalanx, and this becomes a mere flange to that joint,
broadening it for the insertion of the “ primaries ;” this part has its own osseous centre
in the Raptores. I look upon this structure as a sort of bifurcation, like that which is
so often seen in the roughly finished cheiropterygium of the Ichthyosaurs.

The starved third digit (dg.’) has its slender metacarpal (me.*) separated from the
second by a large lanceolate interosseous space; proximally it begins much lower
down than tlie second ; it is continued a little further below; it is largely ossified. The
phalanges of the third digit (dg.°) are in a curiously-aborted condition ; there is an
attempt at segmentation into what should be, according to the Reptilian norma, four
phalanges. In the Ist stage (Pl. X. fig. 1) there is one cartilaginous tract two-thirds
the length and half the breadth of the proximal phalanx of the second digit; the end
of this tract is raised and hooked, and is, indeed, a semi-segmented ungual phalanx.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS, 73

In the 2nd and 3rd stages (figs. 5-7) the main part has an ectosteal sheath, so that it
looks as if it were forming three segments. The ungual phalanx has now become a
mere cap of cartilage on the end of the main segment (fig. 5, dg.*). In my paper on
the Fowl’s Wing (pl. 64. figs. 12, 13, pl. 65. figs. 1-3) I have shown that in Struthio
and Rhea there is a very small wnguwis on the third as well as on the first and second
digits. In the same paper (pls. 62-64.) I have shown, also, that in many cases there is
the rudiment of a fourth digit; this is cautiously lettered (dg.*) as an addendum to
the third metacarpal (me.”). It is a notable part in some old birds, for instance in
Rhamphastos toco.and Dicholophus cristatus. ere, as in many other birds, I do not
find a fourth metacarpal, but there is a rudimentary phalangeal cartilage in all these
three stages (figs. 1, 5, 6, 7, dg.*); it is most clearly seen in the 2nd stage (fig. 5). It
is a pyriform lobe of cartilage, with its narrow end forwards, and is adherent to the
ulnar edge of the abortive third finger; it is about one third its size. Most of the
metamorphosis of a bird occurs in the encysted stage, whilst it is still in the egg, and
this takes place in a marvellously rapid manner. But during the first year, as I long ago
showed in my paper on the Fowl’s Skull (Phil. Trans. 1869, pls. 81-87. pp. 755-897),
and afterwards to the end of life, in some degree slow obliterative change still
takes place through the gradual increase of bony substance. We shall look in vain in
the wing-bones of the adult Opisthocomus for much of what I have described in the
embryo. A comparison of this wing with that of a few other types will show that its
manus is unusually long. Notwithstanding the great distance of this bird from the
Macrochires, it has like them, but not to the same degree, a long distal segment to its
wing-skeleton ; in this respect it comes nearer to the Pteroclide and Turnicide than
to the Cracide. This wing agrees with that of the Cracide and Megapodide (Peris-
teropodes), and also with the Pteroclide, Turnicide, and 'Tinamide, in having the bridge
over the proximal part of the interosseous space of the manus (Pl. X. fig. 8, me.”’); in
this, like them, it differs from the Alectoropodous Fowls, the Phasianide, Tetraonide,
&c. The length of the three main divisions of the wing-skeleton in six types is as
follows (all these are taken from adults, except Turnix rostrata, which was a chick of a
week or two old) :—

Humerus. Cubitus. Manus.
Opisthocomus cristatus. . 70 millim. 72:5 millim. 69 millim.
CrariGlovicend an 40) es 150 &s 115 ¥,
TOUR GOSH. 5 6 5 MR CAN ose 18 ss
Chauna chavaria. . . . 182 BS 204 Fe 155 -
Corythaiz buffoni . . . 48 3 42 a 37 se
Talegalla lathami . . . 84 3 86 _ 71 5

So that this type agrees with Crax, Chauna, and Talegalla in having the cubitus
larger than the humerus; its manus is but little less than its cubitus, differing in this

74 PROF. W. K. PARKER ON THE

respect from all the others except Turniz, in which the distal is nearly one third longer
than the middle segment. The humerus of Opisthocomus is almost as well developed
as ina Pigeon; the thick uncinate snag below the head and in front of the large
pneumatic foramen is very large, and there is a strong crest for the pectoralis major *.

The radius, ulna, and manus are all strong, normally gallinaceous, and indeed very
similar to those of Crax globicera ; the quill-marks on the ulna are less than in Corythaig,
but larger than in Crax. The two proximal carpals are now normally ornithic; all signs
of their division are lost (Pl. X. figs. 8, 9, i.7., ¢.w.). The distal carpals are completely
fused with each other and with the three metacarpals (figs. 8, 9, d.c.", d.c.”, d.c.°; mc.’,
mc.?, me.*). ‘The trochanter on the short first metacarpal is low down, but strong ; the
first distal carpal (fig. 9, d.c.') is a rounded knob of bone looking obliquely downwards
and forwards toward the first metacarpal. The proximal and distal intercalary parts
behind the index (mc.”, dg.”) are completely fused with that member. There is a
distinct semi-oval ungual phalanx to both the pollex and index (dg.’, dg.”), and in spite
of growth and ossification the distal joint of the third digit (dg.*) is still visible, and
also the still more instructive remnant, the aborted phalanx of the fourth finger (dyg.”).
Mr. Perrin, in his figures of the wing, shows the unguis both on the index and pollex
(ibid. pl. Ixiv. figs. 1, 2).

VII. The Hip-girdle of Opisthocomus cristatus.

In the Ist stage (Pls. VIL. & IX. figs. 4 & 5) the moieties of the hip-girdle are
scarcely more than half ossified; they have already acquired their permanent form and
position, and are passing from the Ornithoscelidan, through the Struthious, to the
normal Carinate condition. Thus the axes of the pre-ilium and pubis are nearly
coincident (pr.z., pb.), and the ischium is still separate from the post-ilium (p¢.7.). The
post-ilium is only three fourths the length of the pre-ilium ; this has only a short tract
of cartilage in front; the other part is largely unossified at present. The two moieties
are now a great width apart—one third more than their own breadth; this is due to
the large relative size of the sacral chain of vertebre. The pre-ilia narrow in forwards
sinuously ; the post-ilia at the hinder part of the acetabulum form the projecting facet,
right and left, for the articular surface of the trochanter major. ‘They then narrow in
suddenly, and again widen out to form the projecting eaves of the pelvis; they then
become narrow and die out, and the end of the post-ilium has a sub-uncinate form ; it is
slightly curved inwards. Under the projecting eave the post-ilium grows downwards
in its last two-thirds to meet the ischium; its fore third is deficient, and thus forms
the sacro-ischiatic space (s.i,f.). The three elements of the hip fail to join completely
at the acetabulum, which is always, in birds, open within. The ischium (ésc.) reaches

1 That muscle is really very massive (see Mr. Perrin’s valuable paper, Trans. Zool. Soc. vol. ix. pl. Lxiii.
fig. 3).

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 75

a little further backwards than the post-ilium; behind it is notched above and
lobate below. The pubis (p.) is half the breadth of the ischium, but it is one third
longer ; it thickens behind and then narrows to a rounded point. At the last fourth
of its length it is still, as in front, unossified; and this part is twisted upon itself in
the same manner as I find it in the Guillemot (Uria troile). In that bird the prenasal
rostrum and Meckel’s cartilages are similarly twisted, as also is the epicerato-hyal
band in this bird. If I had found these twisted cartilages in the embryo of a tame or
domestic bird 1 should, of course, have put them to the account of Teratology. I
cannot do this either in the Guillemot or the Hoatzin ; in these birds, which are in a
state of nature, they have manifestly an ontogenetic meaning. In the fossil Toothed
Birds (Marsh, ‘ Odontornithes,’ pls. i., xx., xxi., xxii.) the bill is relatively much longer
than in average modern birds related to them: Hesperornis is a Colymbine Grebe ;
Ichthyornis is an ancient Gull; the length of their jaws bears out this view. In the
hyoid arch I have shown that the bird agrees with Hatteria (Huxley, P. Z.S. 1869,
p- 397, fig. 4); it merely loses part of that arch by absorption, but not without an
attempt on its part to grow larger. The rest of the organization forbids this, and it
becomes a twisted tape. All these things have but one explanation.

The forms that among extinct Reptiles come nearest to Birds in their hip-girdle—
namely, the Ornithoscelida—have an exceedingly long pubis and ischium (Dollo,
‘Bulletin du Musée Royal d'Histoire Naturelle de Belgique,’ tome i. 1882, planche ix.
& tome ii. planche v.). The ancestral birds must have come much nearer these
Reptiles than modern types; but these latter all pass in their hip-girdle through an
Ornithoscelidan stage after they have lost the general Reptilian condition of this part.
‘The gently modified Ornithoscelidan pelvis of the African Ostrich has its very accurate
counterpart in the embryo of the Swan (Cygnus), in which bird, for teleological reasons,
the postacetabular region of the pelvis is extremely long and the pubes very large, as
in the Ostrich, but they are not, as in that bird, fused together below. The hip-girdle of
the adult Opisthocomus has to be studied with the sacrum as part of the pelvis.
Prof. Huxley (op. cit. p. 308) says :—<'The pelvis (figs. 10, 11) is more like that of
Coturnix than that of Corythaix ; but though it resembles both, it differs from both in
the absence of any ilio-pectinal process, and in the circumstance that the ilio-sacral
fossee are completely roofed over by bone. The obturator foramen, as in many Galli-
naceous birds, is not bounded by bone behind ; in Corythaia it is:” [ bounded, but not
perfectly enclosed, by ankylosis of the ischium and pubis, any more than in Gallinaceous
birds and Opisthocomus. Moreover, Corythaix has the roofing of the post-ilium over the
ilio-sacral fossee, although not to the same extent as in Opisthocomus}. The pelvis of
Opisthocomus is very generalized, and differs very much from that of the Gallinaceze
proper (Peristeropodes and Alectoropodes) ; they have the postacetabular region much
broader, and the breadth is seen in its most remarkable degree in Cupidonia cupido
(Shufeldt, op. cit. “ Tetraonide,” pl. xii.). In that bird, and in Talegalla lathami, the

VOL. XIII.—PART I. No. 5.—April, 1891. M

76 PROF. W. K. PARKER ON THE

prepubic spur is very small; it is somewhat larger in Crax, and still larger in Penelope
(Huxley, op. cit. p. 298, figs. 8, 4). In the Alectoropodes it is, as a rule, well deve-
loped, and also in Tinamous (Tr. Z. 8. vol. v. pl. xl. fig. 3; Marsh, op. cit. p. 73,
fig. 20). ‘The greatest development of this part in the Carinate is in certain Coccygo-
morphe—for instance, Geococcyx (Marsh, op. cit. p. 73, fig. 19, and Shufeldt, Journ.
Anat. Phys. vol. xx. pl. 8. figs. 9 & 11); and I find it in a similar state in that
American Cuckoo, Saurothera vieilloti. This bird agrees with the Rallide, both
extinct and existing, in respect of these post-iliac eaves (see Owen on “ Aptornis
defossor,” ‘Tr. Z. §., vol. viii. pl. 15, and Shufeldt on “ Porzana carolina,” Journ. of
Comp. Med. & Surg. July 1888, art. 17, p. 89, figs. 4 & 5). The existing Rallide
have the prepubic spur very distinct, although rather smaller than in the high-heeled
Fowls. ‘There is a family of birds, however,—the Turnicide—in which the spur is
suppressed, and in which, also, the post-ilium forms an eave over the sacro-ischiatic
fenestra, exactly as in Opisthocomus—for instance, Hemipodius varius (Tr. Z. 8. vol. v.
pl. 35. figs. 5 & 8). The same structure, also, is to be seen in Yurnix rostrata.
Another peculiarity of the hip-girdle, in which these two types also agree, is the very
generalized condition of the ischium in form and in its relation to the pubis. Long
ago (Tr. Z. S. vol. v. p. 172), speaking of the congeners of the Hemipods, I remarked :—
‘These allies are very numerous; and it is hard to say which of them should be placed
nearest this, one of the most ‘ mixed’ forms in the whole range of Ornithology.” This
family, the Turnicide, is not so near extinction as that one-membered family, the
Opisthocomide.

I will conclude this account of the pelvis of the Hoatzin by giving, for comparison’s
sake, the length of the pre-ilia and post-alia in six types :—

Pre-ilium. Post-ilium.
Conythain bufiont (0s) 0.) 4) 24emillim, 21 millim.
Opisthocomus cristatus: . 1. © 9 =. 88.) 5, Qi
Hemipodius varius 2 5 = =) - 21 i fare
Durnia nostrata, avs 60) ese OLN. Tae:
Chae MOWGG 8 2 6 6 6 6 6 9 TAs Oma
ONAN GUD 6 co 5 46 6 6 « WS oy 54,

A careful study of the muscular system of the hind limbs, and its meaning and
relation to the habits of these birds, would show cause for the various relative lengths
of the two regions of the enormously extended ilium; in the true Gallinaceous birds
there is great oscillation, so to speak, in these lengths, as the five figures given by Prof.
Huxley show (op. cit. pp. 298-301, figs. 3-7). These figures, and those of the Hoatzin
(op. cit. p. 308, figs. 10 & 11), show that there is one remarkable difference between the
pelvis of this bird and that of the Alectoromorphe. In the latter the pre-ilia are so
united with the sacral spine by ankylosis as to form a right and left gallery, open at

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. U0

both ends; the same thing is seen in Corythaix; in Opisthocomus there is a clear space
between these parts, right and left; this is also seen in the Turnicide. Even this part
of the skeleton, therefore, like the rest, suggests a low ornithological position for this
bird, and justifies the use of the term archaic as applied to it.

VIII. Lhe Hind Limb of Opisthocomus cristatus.

The leg of Opisthocomus in the Ist stage is strong and rather stout (Pl. VII.); its
long elements have a considerable ectosteal sheath ossifying the solid cartilage. The
parts of this limb differ little from those of an embryo of the Common Fowl at this
stage ; we shall find a few notable differences, showing that we are, here, at the parting
of the ways. ‘The relative length of the main divisions of the limb, at this stage, is
very similar to what is seen in a large number of those Carinate that have moderately
long legs, whether they are Perchers, Walkers, or Runners. Measured from the chief
points of flexure, these parts have now their length as below :—

Femur. Tarso-tibia. Tarso-metatarsus. 3rd toe.
11 millim. 13 millim. 8 millim. 14 millim.

The ends of the large cartilages are still unossified, and they will not be finished by
epiphyses, except in the case of the cnemial process of the tibia: in a few Cuculine
birds the top of the fibula has a small epiphysis, but that is rare in the Carinate; in
the fore limb, in some rare cases, sometimes the radius and sometimes the ulna has an
epiphysis. ‘The cnemial process is not large; the fibula (f0.) has shrunk to half the
length of the tibia (¢.); the tibia is quite distinct from the proximal tarsals (¢.,
joe.). All the true tarsal elements are free from bony deposit at present (Pls. VII. &
X. fig. 10). There is some contention as yet as to the true nature of these segments ;
but, after much labour and thought, my own mind is made up as to their morpho-
logical meaning. When I first asked whether or no the lower part of the tibia was an
epiphysis or “the homologue of the Mammalian astragalus ” (“On Baleniceps rex,”
Tr. Z. S. vol. iv. p. 343), I was still in the dark as to the nature of the ‘ sesamoid os
calcis.” Since then, several anatomists, notably Gegenbaur and Huxley, have worked
at this part of the bird’s skeleton.

One of the most valuable pieces of work on this subject is Dr. Morse’s paper “ On
the Intermedium in Birds” (Ann. Mem. Boston Nat. Hist. 1880), though his views,
which are mine also, have been controverted by Dr. G. Baur’.

In a paper (ready for publication), my son, Prof. T. J. Parker, also holds the same
views as Dr. Baur; his observations have been made on the development of the
Apteryx. Jam perfectly convinced of the truth of Dr. Morse’s views. Nevertheless
I believe that there are only two morphological elements in the proximal tarsal series—
an astragalus, or tibiale, and a calcaneum, or fibulare. The so-called sesamoid, or

' See his paper, ‘ Der Tarsus der Vogel und Dinosaurier,’’ Morph. Jahrb, Bd. viii. pp. 417-456, Taf. 19, 20.

M 2

78 PROF. W. K. PARKER ON THE

“caleaneal ossicle” of the older writers, belongs to a lower stratum; it is a centrale.
The intermedium, the so-called ‘ascending process” of the astragalus, evidently
belongs to a higher series—namely, to those of the leg. I classify it with the tibia
and fibula. In the Chick, after seven days’ incubation, my Ist stage in this research,
in the 2nd stage, after eight days, and in the 3rd stage, after ten days, I have examined
these parts with great care. In all these stages the ascending process of the astragalus
is an oblique flange of hyaline cartilage, which runs from the top of the tibiale towards
its outer side to the top of the fibulare on its inner, and then grows upwards. It
cannot, therefore, in this early condition be proved to be the intermedium. Miss
Johnson’s preparation (Stud. Morph. Lab. Camb. vol. ii. 1886, pl. v. fig. 9) and Dr.
Baur’s sections (op. cit.) do not, in embryos answering to my Ist stage, show the
ascending process. At this date the globular fibulare is much more solid than the
wedge-shaped tibiale; the front surface of the latter, and especially the part which
forms the ascending process, is still composed of indifferent tissue. In a day or two,
however, the chondrification is completed, and then the ascending process partly
overlaps the fibulare and the end of the fibula. In the Chick the process is short, but
in Opisthocomus, as in the Ostrich, it is long. In my Ist stage (PI. X. fig. 10) the
tibiale and the fibulare are confluent ; the outer condyle is formed by the latter. The
rest of the ascending process, which is almost four times as long as the interspace
between the two condyles, has become partly confluent with the antero-superior face
of the fibulare; a notch marks its line of junction. A little above the notch a small
ectosteal sheath has already appeared—-the “os intermedium ;” above this the carti-
lage first enlarges and then becomes a narrow style, which lies in front of the tibia
towards its outer side. The remaining cartilage, at the lower end of the tibia, which
has the same depth as the distal tarsal mass, is still sharply separated from that mass.
Postero-internally there is a crescentic cartilage (Pls. VII. & X. fig. 10, ¢.), the centrale
or scaphoid, which is formed out of the interarticular plate ; this forms the bone which
was supposed to be the os calcis. Below the joint the distal tarsal mass (d.¢.?“) is
cupped right and left to receive the condyles, but sends upwards in the middle an
intercondyloid knob ; its lower surface fits to the flattened tops of the second, third,
and fourth metatarsals, which are not ossified above. A rudiment of a fifth metatarsal
did exist on its outer side; the aborted first metatarsal is placed distally ; it has also
begun to ossify. In the relation of the distal tarsal mass to the three large and one
abortive metatarsal segments, a single cartilage or “ chondrite” may be the connate
representative of several chondrites. In the third stage (Pl. X. fig. 11) the ankle-
joint is considerably altered; the tibia has much less cartilage below; that is still,
however, distinct from the proximal tarsal mass. Over the interspace between the
two condyles there is now a cartilagimous tendon bridge (PI. X. fig. 11, ¢.dr.). The
ascending process is relatively narrower in its lower half; its upper three fourths
is entirely ossified by an ectostosis, like the tibia and fibula, The larger tibiale and

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS., 79

the lesser fibulare have not yet appeared as bony centres; they commence two or three
weeks after the intermedium, and are formed as central endostoses, like the bony matter
in the lower tarsal mass. This anachronism as to the ossifications of these three parts
is very remarkable, and the type of ossification is entirely different; the intermedium
is a shaft-bone, whilst the tibiale and fibulare, as in tarsals and carpals generally, are
ossified like epiphyses.

The toes at the Ist stage (Pl. VII., dy.!4) are already developed; the second is a
little shorter and also stouter than the fourth; the third is very long, but its proximal
and distal phalanges are smaller than those of the first toe or hallux; the size of this
heel is very remarkable. The ossification of the ungual phalanges is at the end and
not in the middle, as in the others. The so-called ascending process of the tibiale
requires an ontological explanation ; its ossification, I am quite satisfied, is an inter-
medium ; and this, I take it, is not a tarsal, but belongs to the leg, as the intermedium
in the wing does to the forearm.

The intermedium, in the generalized fore-paddle of certain extinct Reptiles, is seen to
reach up to the humerus, between the radius and ulna (see Marsh on “ Sauranodon,”
Amer. Journ. of Sci. vol. xix., Feb. 1880, p. 170, fig. 1 ; and D'Arcy W. Thompson, Journ.
of Anat. & Phys. vol. xx. pp. 1-4). Even in Birds, and those of the higher sort, I find,
in some cases, the wedge of the intermedio-radiale, which grows upwards between the
radius and ulna, elongated and segmented off, as in the nestling of Spizella pusilla,
and in another species of Spizella I found the apex, representing the intermedium,
separately ossified from the radiale.

In Opisthocomus, as we have seen, and in some other birds, the intermedium of the
carpus is, for a short time, separate. We cannot derive the Bird from any known
Reptilian or Amphibian type; the true ancestor of the Feathered Fowl did not
possess a cheiropterygium, but its limbs were ichthyopterygia ; and the same may
be said of the existing Amphibia, all known Reptiles, and the Mammalia also 1.
In the adult Opisthocomus the hind limb is stout and strong, like that of a Gallina-
ceous bird. The femur is well arched; the tarso-tibia also is normal; its cnemial
outgrowth is not large; there is the usual tendon bridge below, and a shallow inter-
condyloid space for the short process of the tarso-metatarsus. The fibula is half the
length of the tibia; the centrale forms a notable pseudo-sesamoid. The tarso-metatarsus
has a strong grooved process behind its head, which was pre-formed in cartilage
developing posterior vertical ridges to serve as pulleys for the flexor tendons. There
is only one hole in this mass (Huxley, op. cit. p. 309, fig. 12), as in Corythaix, and
also in Craxz, and in the Gallinacee generally. ‘his hole is finished by a periosteal

1 The marvellous series of extinct Reptiles discovered in stratum under stratum has emboldened some
biologists in the matter of ancestry. Guesswork, however, is not science; and we have no proof that certain
archaic forms begat others that are of a more recent date: we “ have nothing to draw with, and the well is

deep.”

80 PROF. W. K. PARKER ON THE

plaster. ‘There is a small hole between the middle and inner metatarsals near the top.
The lower condyles of the shank agree very closely with those of Crax, but the tarso-
metatarsus differs from that of the Curassow in its flatness, from side to side, and in the
degree of concavity of the anterior outline. The condyle for the fourth digit is
quite Cracine, being somewhat grooved, and having an outer process, which reaches a
considerable distance backwards. In Corythaix this condyle is slightly grooved, but
the inner half is larger, and that bird has a mobile outer toe. There is very little
difference, in Opisthocomus, between the proximal and penultimate phalanges. The
hallux, or first digit, is larger than in the Megapodide and Cracide, and its claw is
the largest of the four, as I have shown in the Ist stage (Pl. VII., dg.1); thus the foot
of this bird is truly Insessorial; it is more of a Percher than its nearest relatives, the
Curassows.

IX. RECAPITULATION AND SUMMARY.

a. The Ornithological Position of Opisthocomus.

The existing Carinate birds are extremely hard to classify; we are embarrassed with
the riches of this great Avifauna; but there is one clue that can be used, namely that
the fewer there are in any family the more archaic that family is, and vice versd. 'The
Ratite have acquired their distinguishing character ; it is evidently not primary. The
Tinamous and the Hoatzin are, in many respects, as archaic as the Ratite; in some
more so. If the term Dromeomorphe is to be retained, let it take in the Ratite and
the Tinamide. Next to that group would be the Alectoromorphe, and the lowest of
these would be Opisthocomus——the Opisthocomide ; the highest the Columbide: that
group would also take in the Pteroclide, which lie under the Pigeons.

Here, also, between the Tinamous and the Quails, would be placed the Turnicide
(not to be called Turnicomorphe—that term is too general). Then, on the other
side of the class, the Cypselomorphe and the Psittacomorphe should be melted into
the common mass with the rest of the Coccygomorphe ; we should thus get an Order
to compare with the Coracomorphe, as variable as that Order is uniform, Ornitho-
logists must understand that terms taken from the anatomical structure of birds are
not ornithological; they are morphological terms, which the Ornithologist uses for
taxonomic purposes. Sometimes they are of very little value; at other times they can
be applied to large assemblages of birds. Thus Prof. Huxley’s term Aigithognathe is
of great value; the only Family outside the great Passerine group -(Coracomorphe)
which has that peculiar modification of an open or schizognathous palate is the Cypse-
lide. Thus the morphological term Aigithognathe can be nearly superimposed upon
the ornithological term Coracomorphe. But the term Desmognathe has no such wide
use, nor the term (one of my own coining) Saurognathe; the Toucans are Desmo-
enathe, the Woodpeckers are Saurognathe ; yet these two groups are closely related.
In the present state of our knowledge Taxonomy is a very tentative science ; it is only

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 81

a sort of rough scaffolding, and not a Temple of Nature; moreover, each ornithological
artisan will use his own classification, and that of no one else; thus there are as many
systems as there are Ornithologists. We see this most instructively in the case of that
most skilful worker, whose classification I have just reviewed. In his Coracomorphe
six thousand species are fagoted together; his Heteromorphe contains only one
species—Opisthocomus cristatus. It is true, however, that this bird differs more from
any other existing type than any, the most diverging, do amongst the whole multitude
of Passerines. Again, by choosing those characters in the members of two distant
Families that agree, and forgetting those in which they differ, it is not difficult to make
a “happy family” out of very discordant members. If we are desiring to use a compre-
hensive term, well and good; but the binder round such a group should be made of an
elastic withy, and not of cast iron. Now in Prof. Huxley’s Cecomorphe (op. cit.
pp. 457, 458), the Gulls, Petrels, and Auks are put together; but the Penguins are
left out, and the term Spheniscomorphe merely means the same as Spheniscide. But
the Great Auk was a very accurate isomorph of a Penguin. The extinct toothed birds,
Hesperornis and Ichthyornis, have to be packed together inside this Cecomorphous
bundle ; for the former is the quasi-parental form of the Grebes and Loons, and the
latter is the Gull of an old dispensation. The problem is rendered more difficult by the
fact that the Grebes, Loons, and Petrels, like the ancient Hesperornis, have the newest
form of avian vertebral articulation. Their cervical and dorsal vertebre are cylindroidal
or heteromorphous; whilst the dorsals in Gulls and the Alcide are opisthoccelous. In
the Gulls, most instructively, there is a partial retention of the amphiccelous condition,
which is seen in the presacrals, generally, of Ichthyornis. Now as to Opisthocomus,
I would drop its group term Heteromorphe, and yet hold on to the spirit of that
term, whilst I discard the letter. This bird, an archaic Fowl, or Gallinaceous type,
differs from the Gallinaceous birds of its own Neotropical Territory, namely the
Cracide, as much as the most Passerine of the Coccygomorphe do from their nearest
allies in the great Passerine group. ‘This bird, like the Tinamous, belongs to the same
general stratum of old forms as the Ratite ; but it has not, nor have the Tinamous, lost
the sternal keel, nor the size and strength of the wings.

Its isolation suggests at once its ancientness; its morphology, which throws some
light on its ontogeny, teaches the same thing. Like the Tinamou, it is an arrested
type; the Ratite are both an arrested and a degraded group. I suspect that the
earliest birds were Carinate.

b. Lhe Light cast upon the Ontogeny of Birds by the Morphology of Opisthocomus.

In the skull of Opisthocomus we have the fundamental form of that of a Carinate bird ;
it has undergone that remarkable change by which the upper face can be moved upon
the frontal region. In this it is above the Struthionide, which, like Reptiles, have no
such hinge. Its basipterygoids, although they die out, are truly Struthious. The

82 PROF. W. K. PARKER ON THE

hyoid arch is not only essentially the same as that of Hatteria, but the descending bar
of cartilage, which is aborted in the middle in all birds, is in it unusually large and
shows signs of secular shortening by its folded or puckered condition. ‘The vertebral
chain has the same development as in Gallinaceous birds generally ; its dorsals are
eylindroidal, but so also are those of the Tinamou, the Ratite, and, as I have just
mentioned, Hesperornis, a bird so long extinct ; hence it is evident that this peculiarly
avian structure is extremely ancient. The appendages of the ribs are not typical;
they, however, are well seen in Hatteria and, to some degree, in the Crocodiles, whilst
they are absent in Chawna—an archaic Chenomorph.

Those parts of the body which, becoming transformed for the purposes of flight,
dominate all the rest of the organism—the sternum, shoulder-girdle, and wings—are,
in this bird, so very instructive as to make it very precious to the Morphologist: in
this it has no peer.

The sternum, in the backward position of its small keel and the feeble development
of the postcostal or metasternal region, is an intermediate structure between that of a
Bird and that of a Lizard. So also is the shoulder-girdle in all its parts and relations ;
besides which the scapula is developed in the same manner as in the Frogs and Toads.
The wings in the embryonic stage not only have the claws of the first and second digits
nearly as large as in the foot, but there is, as in the Ostrich and Rhea, a rudimentary
unguis on the third half-aborted digit, and a phalangeal remnant of a fourth digit ; a
metacarpal remnant is seen in the Fowl and several other birds. ‘The carpus, as in a
few other birds, is, for a time, divided into at least seven segments, and is then essep-
tially like that of an Amphibian or a Chelonian.

For a time the pubis strives to attain to an Ornithoscelidan length ; after the middle
of incubation it acquires its proper relative avian length. But the history was taken
up at its middle—at the middle, actually, of incubation ; it is, however, in the still
earlier stages that the ontogeny of this Class is seen most clearly; those earlier stages
have had to be worked out in less archaic types. ‘The true Archwopteryx is lost in the
almost infinite past; the “so-called Archwopteryx” is very doubtful as a parental
form. It does, however, help us to imagine how the vertebral internodes became
aborted and partly suppressed, so that a feather-shaped tail became fan-shaped, the
rectrices being set in a semicircle instead of growing out of the sides of a long
Reptilian tail, We have, however, nothing intermediate between these two types of
tail even in the Cretaceous birds. Something of the same sort has occurred in the
limbs. If the limbs of Ceratodus may be taken as a pattern of an ichthyopterygium
there has been a secular shortening of some, and suppression of other, axial parts or
internodes: thus we get the shortened radiating fan-shaped limb—the proper cheiro-
pterygium ; that, in turn, had to be transformed into the avian fore limb or wing.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS.

X. List of Abbreviations used in the Letterpress and Plates.

(The Roman Numerals indicate nerves and their foramina.)

ac. Acetabulum. ds.v. Dorso-sacral vertebre.
ac.p. Acromial process. d.v. Dorsal vertebre.
ag. Angulare. é.br. Epibranchial.
al.e. Aliethmoid. e.hy. Epibyal.
al.n. Alinasal. em. External nostril.
al.s. Alisphenoid. e.o. Exoccipital.
al.sp. Aliseptal. e.st. Extrastapedial.
ao. Antorbital. eu. Eustachian opening.
ar. Articulare. f. Frontal and Femur.
a.s.c. Anterior semicircular canal. fo. Fibula.
at. Atlas. foe. Fibulare.
a.t.r. Anterior tympanic recess. fm. Foramen magnum.
au. Auditory capsule. fr. Furcula.
ax. Axis. g. Ganglion.
6.h.br. Basihyobranchial. gl. Glenoid cavity.
b.hy. Basihyal. h. Wuamerus.
b.o. Basioccipital. h.s.c. Horizontal semicircular canal.
b.s. Basisphenoid. i. Intermedium.
b.¢. Basitemporal. i.a.c. Interarticular cartilage.
c. Centrale. 7.ag. Internal angular process.
c.a. Costal appendage. z.c. Internal carotid.
c.br, Ceratobranchial. ic. Intercentrum,
cd.v, Caudal vertebre. i.cl. Interclavicle.
cf.c. Craniofacial cleft. al. Tlium.
c.hy. Ceratohyal. i.o,f. Interorbital fenestra.
el, Clavicle. i.r. Intermedio-radiale.
em, Centrum. isc. Ischium.
cn.p. Cnemial process. i.st. Infrastapedial.
er. Coracoid. i.tb. Inferior turbinal.
c.r. Cervical rib. i.tr. Intertrabecula.
er.g. Crista galli. j. Jugal.
cu. Centralo-ulnare. 1. Lachrymal.
c.v. Cervical vertebrae. lc. Lachrymal canal.
d. Dentary. Is.v. Lumbo-sacral vertebre.
d.c. Distal carpals. mk. Meckel’s cartilage.
dg. Digit. m.st. Metasternum and Mediostapedial.

VOL. Xil1.—Part 11. No. 6.—April, 1891. N

84

nC

PROF. W. K. PARKER ON THE

. Metacarpal.

mt. Metatarsal.

m.ty

me.
M.D.
Nn.

ne.
Np.
n.W.

ob.f.

06.0.
Op.
. Parietal.

. Palatine.

. Posterior angular process.
. Parasphenoid.

. Pubis.

. Precoracoid.

. Precostal process.

. Perpendicular ethmoid.

. Membrana tympani.
Maxillary.
Maxillo-palatine.
Nasal.

Notochord.

Nasal wall.
Obturator fenestra.
Occipital condyle.
Opisthotic.

. Pterygoid.

. Prenasal.

. Prepollex.

. Pre-ilium.

. Prootie.

. Presphenoid.

. Posterior semicircular canal.
. Post-ilium.

. Premaxillary.

. Pituitary space.
. Quadrate.

Nasal process of premaxillary.

). Quadrato-jugal.

. Radius.

. Rostrum of basisphenoid.
. Radiale.

.st. Rostral process of sternum.

. Supra-angulare.

Scapula,
Sacro-ischiatic fenestra.

. Septum nasi.

. Supraoccipital.

. Splenial.

. Squamosal.

r. Sacral rib.

. Suprascapula.

. Suprastapedial.

st. Stapes and Sternum.
. Sternal keel.

. Sternal ribs.

Sacral vertebra.

. Tibia.

. Tibiale.

. Tendon bridge.

. Lympanic wing of exoccipital.
. Tympanic cavity.

. Ulna.

. Ulnare.

. Uropygium.

. Uro-sacral vertebre.
. Vomer.

. Vestibule.

*. Vertebral rib.

XI. DESCRIPTION OF THE PLATES.

PLATE VII.

Opisthocomus cristatus (1st Stage): skeleton, side view, x 44 diam.

MORPHOLOGY OF OPISTHOCOMUS CRISTATUS. 85

PLATE VIII.

Fig. 1. The same (1st Stage): skull, lower view, x 6 diam.
Fig. 2 0 (3rd Stage): skull, section, x 45 diam.
Fig. 3 N A hyoid arch, upper view, X 44 diam.
Fig. 4. Pr (2nd Stage): nasal labyrinth, section, x 20 diam.
Fig. 5 ne se section of quadrate and stapes, outer view, X 15 diam.
Fig. 6 3 . stapes, Inner view, X 224 diam.
Fig. 7 a ib section of cervical vertebree, * 24 diam.

PLATE IX.
Fig. 1. The same (lst Stage): shoulder-girdle and sternum, lower view, X 6 diam.
Big2: ss ss part of shoulder-girdle and sternum, upper view, x Y

diam.

Fig. 3. i 43 furcula, lower view, X 9 diam.
Fig. 4. 5 es pelvis, lower view, x 6 diam.
Fig. 5. Fe ¥ pelvis, upper view, x 6 diam.
Fig. 6 55 (5rd Stage): part of pelvis, lower view, x 74 diam.

PLATE X.
Fig. 1. The same (Ist Stage): manus, outer view, X 9 diam.
Fig. 2 ss om part of manus, inner view, X 9 diam.
Fig. 3 ie re part of Fig. 2, x 27 diam.
Fig. 4. 4 (2nd Stage): part of pollex, outer view, x 12 diam.
Fig. 5 ee ee second to fourth digits (part), xX 12 diam.
Fig. 6 i (Srd Stage): manus, outer view, x 6 diam.
Fig. 7 BS A part of same object, inner view, x 6 diam.
Fig. 8. A (Adult): manus, outer view, x 1) diam.
Fig. 9 Fe 5 manus, inner view, X 13 diam.

Joes, OE ay (1st Stage): ankle-joint, front view, x 12 diam.
Bien ies (3rd Stage): ankle-joint, front view, x 6 diam.

(This memoir was not set up in type until after the Author's decease, which took
place July 3rd, 1890, nor had the Plates been lettered at that date. It has conse-
quently been a difficult task to correct the proofs, and I have to thank Mr. W. N.
Parker and Mr. F. E. Beddard for very material assistance in this matter.—P. L. S.,
Jan. 12th, 1891.]

SQWOODOHLSIdO AO ADOTIOHdGYHOWN

; EL STE LW
dur ueurmayy' say) Fea pe Pp dy M

WEG BIg

suits
haa 0)

OR an a
Wane Wercn Wh

SOeEyN
SAT

Mey
Wey
i

oY A,
van

4

emilodl ee (CLAM

4 x 20 Tepe

ssl

\} 6 x 22%

WKP del.ad nat.
MP. Parker hth.

DAA Masta |
Cope
Oe ATV aA

hte

ty yy
ine
Dania

No
i
Bi

iY

i Gi e [ i Vv air, IW M G " ‘ “) 4
Whe ‘ f Ath : G ‘ ee Aa rie ’ Tay 1 % one 4 i
<b ; ; re : ; ea ah : ; ; it ; et hah aa sotat

4
Medici

e
rine
Dae
aie
usar te

ate

Sans Lol, Soe, Vol MTS De

West, Newman,im}

MORPHOLOGY OF OPISTHOCOMUS.

WK Pdel ad.mat.
MP Parker, lith.

; ur Pha

vu

th 1Y THOCOMUS. ; West, Newman,imp.
Le eetae MORPHOLOGY OF OPISTHO

Soe

Th;
An
i

Be
Tak al os

i,

UP NTA R I
(tes teed many

fest Seep INE
ape:

oh

Fest]

IV. Contributions to our Knowledge of the Antipatharian Corals. By F. JErrRey
Bett, WA. Sec.R.M.S., Corr. Mem. Linn. Soc. N.S.W., F.Z.S., Professor of
Comparative Anatomy and Zoology in King’s College, London.

Received April 11th, read May 6th, 1890.
[Puates XI. & XII.]

I. Observations on a particularly fine example of the “ Black Coral” of the Mediter-
ranean lately acquired by the Trustees of the British Museum.

Tue Trustees of the British Museum have lately purchased from Messrs. Cresswell,
the well-known importers of Sponges, a very remarkable example of one of the best-
known of the products of the Mediterranean. Common enough as pieces of the
“black coral” of the Mediterranean are, and elaborate as were the classical researches
made on it by the eminent French naturalist, H. de Lacaze-Duthiers!, the size and
beauty of the specimen herewith figured are sufficient to justify me in offering the
Society some account of its history and appearance. I have sent photographs? of it
to various naturalists and curators of museums, and I have been favoured by Dr. y.
Marenzeller, of the Museum in Vienna, with the following remarks :—‘‘I have never
seen such a splendid specimen. Our Museum possesses an example from the Adriatic
nearly as high, but not so densely branched. ‘There are also very large trunks without
branches. Your example is indeed what we call ‘ Cabinetstiick ersten Ranges.’ ”

Dr. F. Bernard, of the Muséum d’Histoire Naturelle of Paris, has favoured me with
the following notes on the specimen, which, in the second edition of Lamarck’s
* Animaux sans Vertebres’ (t. ii. p. 491), is said to be “un échantillon gigantesque....

?

dont le tronc égale la grosseur du bras ”— :“ I] a 63 centim. seulement de hauteur ; il
est en effet réduit & ses grosses branches; tous les rameaux plus petits ont été brisés.
Le trone mesure 12 centim. environ de large sur 8 du profondeur.”

Our President, who has lately visited many of the larger museums of the Continent,
assures me that nowhere has he seen a specimen which can vie with that which has
lately been acquired for the nation.

Mr. Cresswell tells me that the specimen which he sold us was dredged by sponge-

1 Ann. Sci. Nat. (Zool.) (5) ii. p. 169.

2 These were taken for me, with great kindness, by my accomplished colleague Mr. Antony Gepp, M.4A.,
of the Department of Botany.

VOL. XIII.—PaRT Il. No. 7.—April, 1891. oO

58 PROF, F. JEFFREY BELL—CONTRIBUTIONS TO

fishers, in some 20 fathoms of water, not far from the Island of Negropont. ‘The size
of the boat and the appliances it contained were doubtless well adapted even for the
magnificent sponges which those fishers obtain, but it was not sufficient to carry a
specimen more than 6 feet high and more than 6 feet wide. ‘To these unfortunate
limitations of strength and space are due the fact that the base of the specimen had
to be cleft in twain, so that part of the brittle base was lost. No other serious acci-
dent befell the specimen, though tact and diplomacy were needed to effect its removal.
It is now in process of being skilfully mounted, and will, no doubt, be for many years
one of the most noticeable features in the Zoological Galleries of the Museum.

Description of the Specimen, with measurements.

As will be immediately seen from the figure (Plate XI.), the great beauty of this
example lies in the density of the reticulation. The base is convex with a sharp edge,
where it is uninjured, and spreads over an area of 350 by 200 millim. ; from it at once
arise two great trunks, the larger and finer of which gives rise to a fan which is almost
2 metres high (6 ft. 6 in.), while it is more than 2 metres wide (6 ft. 8} in.); the
smaller fan is 1-425 metre (4 ft. 8 in.) high, and is 1-280 metre broad (4 ft. 24 in.).
The main trunk of the former is 425 millim. (1 ft. 5 in.) in circumference, and that of
the latter is about 290 millim. (114 in.).

This last trunk rises some 100 millim. and then divides into two branches, one of
which, that on the left, as seen in the figure, is more than twice as wide as the other ;
it again rises some 100 millim. before dividing, and then gives off several branches ;
the one most to the right begins almost at once to enter into intimate reticular con-
nection with the original right trunk; the next does not do so for some distance from
the point of origin, and still further off it becomes connected with its neighbour. It
is only quite at the top of the whole of this smaller fan that this second trunk to the
right becomes connected with the branches given off to the left; but somewhat lower
down there are distinct signs of a fracture and loss of pieces which might have well
effected a more extensive reticular connection.

Of the trunks which I speak of as being given off to the left four belong distinctly
to the left side, and one is almost median in its position; the four laterals almost
immediately become connected with one another, and three of them fuse to form a stem
150 millim. in circumference ; this rises almost parallel to the more median trunk, with
which it enters into close connection. From the middle of the reticulation between
them another trunk arises, which effects unions with the continuations of the stems
both to the right and to the left of it. To the left, and ina plane more remote from
the observer (who may be supposed to be standing in front of the whole colony), a large
trunk rises up, soon swelling into a considerable enlargement. ‘Thus, joining their
neighbours here, bending forward there, or sidewards, now outwards, then inwards,

OUR KNOWLEDGE OF THE ANTIPATHARIAN CORALS. 89

the stems pass on, getting steadily, if slowly, more and more slender; sometimes the
reticulating bands are stout, short, and frequent; sometimes a considerable space is
bridged over by a much more delicate bar.

The fan behind is of greater beauty than the more stunted and more injured branch
that lies in front of it, but it is needless to enter into the details. It divides into three
main branches, one of which occupies the left, and the other two the right halves of
the fan; this mass, being more complete than that in front of it, is much more beautiful,
far more beautiful than any description of mine would lead the student to suppose ;
the imagination may well allow itself to revel in the idea of the vision of beauty that
must have presented itself when this magnificent fan was covered by the living matter
which formed it, when the polyps with their expanded tentacles were drawing on the
nourishment around for the means to sustain and increase it.

On the hinder fan there are to be seen outgrowths from stems of various sizes which
did not take any share in the formation of the network; these may be merely sessile
knobs, or they may be as much as 60 millim. long, but the free end is always converted
into a rounded head.

Saving only the information which the specimen gives us as to the size and beauty
to which the “ black coral” may grow, it adds nothing to our knowledge of the mor-
phology of the species, but this has already been studied. When, however, we reflect
on the delight which an object of such beauty can give us, we may congratulate our-
selves on the good fortune of the Trustees in securing it.

On the Name to be applied to the “ Black Coral” of the Mediterranean.

I must pass from this poor attempt to express to others the feelings that the sight
of this object arouses in me, to the more prosaic and infinitely less agreeable question
as to the generic and specific names which this species should bear.

Prof. Lacaze-Duthiers has already ! pointed out that it has received more than one
“ scientific name ” from having been described by different authors in different degrees
of perfection; on these there is no need to dilate, as the question was, to a certain
extent, settled by the eminent zoologist to whom I have referred, and there may be
said to have been a universal acceptation of the name he applied to it, Gerardia
lamarcki, Haime. Yet more recently we have been again made familiar with the
generic name by the same zoologist’s beautiful investigations into the history of the
crustacean parasite which he has called Laura gerardie°.

Mr. Brook, in his valuable report on the Antipatharia collected during the voyage
of H.MS. ‘Challenger,’ uses the generic name Savaglia.

1 Ann. Sci. Nat. (Zool.) (5) ii. (1864) p. 173.
2 Mémoires de l’Acad. des Sciences de l'Institut de France, xii, (1883) no. 2, p. 4.

90 PROF. F. JEFFREY BELL—CONTRIBUTIONS TO

It has been a matter of some trouble to trace the history of the name Savaglia}.
The first reference which I can find to it is in the “ Dell’ historia naturale di Ferrante
Imperato” 2, published at Naples in 1599; on p. 724 of which we read, “Savaglia. La
Savaglia é pianta nel ramiggiare, e Veffigie tutta simile a Corallo,” &c.; these words are
repeated on p. 632 of the edition of 1672, which was published at Venice. Nearly a
century later (in 1755) there were published the ‘* Opere postume ” of Giuseppe Ginanni,
who takes (p. 17) Savaglia as his type of “ quelle Pianze dell’ Adriatico, che sono di
sostanza legnosa senza foglie.” Kleven years later Pallas writes in his notes to Gorgonia
Antipathes— Quee cum Corallio nigro vulgo confunditur et a multis auctoribus pro
vero descripta fuit, Savalia Maris Mediterranei, truncus magnorum Ventilabrorum,
detritis ramis, politus esse solit” (* Elenchus Zoophytorum,’ pp. 194 & 195).

Up to this time it is clear that the term had no definite generic or specific significance
in the Linnean sense. But some years later (in 1819) A. Bertolini published his
‘Ameenitates Italice,’ and gave an account of Savaglia, for which he makes a definite
zoological position by calling it Gorgonia savaglia. In 1844 Nardo, in the ‘ Atti della
quinta Unione degli Scienziati italiani,’ published (p. 433) a classification of Zoophytes,
wherein the portion which interests us runs thus :—

Fam. IV". ANTIPATIDI.

Sotto famigiia I* Antrpatin1.—Polipi a sei tentacoli.
Gen. Anthipathes, Pall. Gen. Cirripathes, Blainv.
S. f. II* Savanrn1.—Polipi a sedici tentacoli.
Gen. Savalia N.

The new genus is for the Gorgonia savoglia |sic] (Bertolini). Nardo does not write
out the full name or names of the species to be placed in this new genus, but it is clear
that had he done so he would have then written Savaglia savaglia | or Savalia savalia}.
This use of a specific for a generic name has been forbidden by the rules of the
British Association.

Meanwhile this species had become famous by the researches of Lacaze-Duthiers,
who, conferring on it (in 1864) the generic name of Gerardia, retained for it the specific
name of lamarcki given it by Haime (in 1849) when he called it Letopathes lamarcki.

Mr. George Brook, in his recently (1889) published ‘ Challenger’ Report, which
adds so much to our knowledge of the obscure group of Antipatharia, seems to have
suffered from incomplete bibliographical information, for he writes :—* I have not seen
the original, and do not know if Nardo gave the species a specific as well as a generic

1 The word is not in any Italian dictionary that I have been able to consult, and my learned colleagues in the
Printed Book Department have been unable to throw any light on the origin of the word. Count Salvadori
does not know the word, and his learned friend Prof. Lessona is unable to throw any light on its history.

2 Twas not assisted in my search by Nardo’s exquisite misprint of ‘ Ferravite Traparuto;” but I was
greatly aided by the bibliographical knowledge of Mr. Carruthers, F.R.S., on whose patience I made severe

demands.

OUR KNOWLEDGE OF THE ANTIPATHARIAN CORALS. val

name ; there isno mention of one in his recent publication. I have, therefore, retained
the specific name of Haime.”

It is clear that there is no escape from the conclusion that the proper specific name of
this Coral is savalia, and the generic Gerardia, and the synonymy will stand thus :—

Family GERARDIID, Verrill, Trans. Connectic. Acad. i. p. 499.
Savagliide, G. Brook, Chall. Rep. Antipath, 1889, p. 79.
Savaiini (subfamily), Nardo, Atti 5* Union. Scienz. Ital. (1844) p. 433.

Genus GERARDIA,

Gerardia, Lacaze-Duthiers, Ann, Sci. Nat. (Zool.) (5) ii. 1864, p. 175.
Savaglia, Nardo, loc. cit.
Savaglia, id. Atti R. Ist. Veneto, (v.) iii. 1876, p. 674; Brook, loc. cit.
Gorgonia (pars), Antipathes (pars), Leiopathes (pars), auctor. complur.

Species GERARDIA SAVALIA.
Gorgonia savaglia, Bertolini, Ameen. Ital. (1819) p. 219.
Leiopathes lamarcki, Haime, Ann. Sci. Nat. (Zool.) (3) xii. (1849) p. 225; M.-E. & H. Hist.
Nat. Corall. 1. p. 322.
Gerardia lamarcki, Lacaze-Duthiers, loc. cit. (1864).
Savaglia lamarcki, Brook, op. cit. p. 80 (1889).

II. On a remarkable Antipathid from the neighbourhood of Mauritius.

Shortly after the arrival of the beautiful specimen just described, M. de Robillard
forwarded to us a very remarkable Antipathid from the neighbourhood of the island of
Mauritius. As the specimen is dry it is impossible to assign it definitely to any one of
the genera now strictly limited by Mr. Brook; and, like some other Antipathids, it
may, following that naturalist’s proposal, be called [ Antipathes]. As it is proposed to
exhibit this example, which I am fairly confident is at present unique, it is necessary
to give it a name, and to publish such a description and figure of it as shall enable it
to be recognized.

M. de Robillard has forwarded, during the last few years, many fine examples of
Anthozoa to the British Museum, and I am glad to have this opportunity of com-
memorating his services by associating his name with this remarkable growth.

Description of [Antipathes] robillardi. (Plate XII.)
From a small horny base there arise abruptly several trunks ; these soon divide and
vive rise toa number of greatly elongated stems; some of these are, henceforward,
simple; others divide at once two or three times, and others do not divide till they are
some slight height from the base. In the case of one stem only is there any division
at a distance greater than 7:5 centim. from the base. The result of this mode of
growth is an appearance quite different from that of most Antipathids.
VOL. xui.—Ppart ir. No. 8.—April, 1891.

92 CONTRIBUTIONS TO OUR KNOWLEDGE OF ANTIPATHARIAN CORALS.

The stems taper quite gradually, and are fairly flexible near the tip, though rather
brittle at their base. Where a branch is given off from a stem it is nearly always given
off a short distance only from the base, and is ordinarily set at a very wide angle. In
a few cases the stems have, during growth, been diverted from their line of growth, and
an angle or elbow is thus produced, or there is a more or less irregular curve in the
course of the stem. ‘The stems vary in length and thickness, and those that are thicker
and longer appear to be older than those which are thinner and shorter.

Where the sclerenchyma is well preserved it has the appearance of being transversely
striated, as its dark yellow colour is relieved by narrower and lighter bands; it is
quite rough to the touch owing to the shagreen-like spinulation of its horny axis; the
spines on this axis are blunt and very numerous (Plate XII. fig. 3).

There are about 45 of these stems, the longest of which are about 3 feet 3 inches
(that is, almost exactly one metre) long; the shortest are about 15 inches, or rather
less than 40 centim. long.

Hab. Mauritius.

It is to be hoped that the publication of this notice will lead to a fuller knowledge
of this interesting form ; for the present we must be content to know of its existence,
but the attention of collectors should be called to it and every effort made to obtain
examples preserved in spirit ; from such specimens alone can we get the information
which will enable us to assign a satisfactory systematic position to it, and justify us in
speculating as to its origin and affinities.

DESCRIPTION OF THE PLATES.
PLATE XI.
Gerardia savalia; the size of the specimen figured may be estimated by the foot

measure placed at the side of the Plate. The figure is a tracing of a photograph
of the object, so that the relative proportions of the branches may be relied on.

PLATE XII.
[Antipathes| robillardi.
1. View of the skeleton of the whole colony, showing its general form, the rela-
tions of its stems, and the mode of branching; + nat. size.
24, 2. Branches at base ; nat. size.
3. Surface of stem, magnified four times, to show the character of its spinulation.

Sram. Loo. Loe. Ul MMM

GERARDIA SAWALIA .

Ay
pag Thy

ney Ae
eae

as

len SC ae
na ee

Tram. Loot. Loe Vit MUBEAT,

[beac

M&N.Hatihaxt imp

Berjeau & Highley del.et lith .

ROBILLARDI

ANTIPATHES

i f wo i } i} } + ie
ae PE eit ,
i i ive 1 ‘
} i \
\ i : i
is 4 1 coh :
Pte ;
se r De aCe
i Lend ; ; { ay ORME
ae he . \ j toy r
MES 4 ; iy D
t i 1 y * me
Ai) ; ue / if
no; ] iN ‘ I dtp auk, i
\ me i U fi i
at ie H i

ye hk -
-

’
tite

ee

V. Catalogue of the Reptiles and Batrachians of Barbary (Morocco, Algeria, Tunisia),
based chiefly upon the Notes and Collections made in 1880-1884 by M. Fernand
Lataste. By G. A. BouLeNncER.

Received October 2nd, 1890, read November 18th, 1890.

[Puatres XIII.—XVIII.]

Introduction.

ON his appointment to the posts of Professor of Zoology and Assistant-Director of the
Museum in Santiago, Chili, in 1889, M. Lataste felt reluctantly compelled to abandon
his projected Catalogue of the Reptiles and Batrachians of Barbary. He handed over
to me the whole of his notes, with the request that I should bring out the work. The
constant correspondence with my distinguished friend ever since he took up the study
of that fauna, as well as my acquaintance with his collection, had prepared me for such
a work; and knowing I could rely on his kind help in revising my MS., I thought it
would not be presumptuous on my part to undertake the Catalogue which I have now
the honour of offering for publication to the Society.

Accounts of his visits to Algeria and Tunisia have been given by M. Lataste in the
introduction to his Catalogues of the Mammatia', to which I must refer for the detailed
itinerary.

The first journey was undertaken in 1880, and lasted from the middle of February
to the end of July. The environs of Algiers, Setif, El Guerah, Batna, Lambesa, and
Biskra having been explored, an excursion was made in the south of the High
Plateaux and far into the Sahara, which lasted from the middle of March to the end
of May. The route followed was from Biskra to Wargla through Tuggurt, and back
to Biskra through the Mzab, Laghouat, and Bou-Saada. From June Ist the following
places were visited in succession :—Batna, Constantine and environs, Bona and environs,
Philippeville, Setif, Chabet-el-Acra, Bougie and environs, Dellys, and Tizi-Ouzou and
environs.

The second visit, from April lst to June 20th, was chiefly devoted to the exploration
of Kabylia and the region of the Hodna, after a stay of two weeks at Oran and two
weeks at Algiers. The itinerary was from Algiers to Aumale, Beni-Mansour, Tilla-
Rana, Fort National, Bougie, and thence south to Msila, Wed Magra, Barika, and
Batna, and back to Algiers through Constantine, Setif, and Palestro.

1 “Catalogue des Mammiféres Apélagiques sauvages de Barbarie,” Actes Soc. Linn. Bordeaux, xxxix.
1885.—Exploration Scientifique de la Tunisie. Catalogue critique des Mammifcres Apclagiques sauvages.
Paris, 1887.

voL. X11.—Part m1. No..1.—October, 1891. Q

94 MR. G. A. BOULENGER ON THE REPTILES

Asa member of the Official Scientific Mission, M. Lataste visited Tunisia in 1884,
arriving at Tunis on the Ist April. The greater portion of his stay was spent on the
Gulf of Cabes and the Island of Djerba (April 15th to May 25th). He then marched
from Cabes, through Tozzer, 'Taferma, Gafsa, Feriana, and Tamesmida to Tebessa in
Algeria (July Ist), whence he proceeded, by coach and rail, to Bona.

On the occasion of the meeting of the ‘ Association Francaise” at Oran, in April
1888, M. Lataste paid a second visit to that place, again exploring its environs, and
making an excursion to Ain Sefra, on the frontiers of the Sahara and Morocco.

Besides the materials brought together by M. Lataste, the rich collection of the
British Museum has afforded me much information, especially respecting Morocco.
And I have to tender my best thanks to my friends Dr. O. Boettger, of Frankfort-on-
the-Main, and Dr. F. Miiller, of Basle, for loan or gift of additional material.

As to the scope of the present Catalogue. My object has been to give only so much
description and synonymy (together with a reference to a good figure, when such exists)
as 1s necessary to ensure correct identification of the species, and I hope the keys I have
given will prove to work satisfactorily for that purpose. In a few cases I have had to
enter into discussions on the value of certain characters, and for some little known
species I have inserted fuller descriptive notes; but the distribution in Barbary of
the various forms has been dealt with as comprehensively as the data available at
present permit.

I have accepted the geographical delimitation of this fauna as traced out by my
friend in his Catalogue of Mammalia, and for the reasons, purely practical, which he
has given in that work. For it goes without saying that the Sahara forms no part of
Barbary in a zo00-geographical sense, whilst the exclusion of the Cyrenaica, which, as
observed by Sir Lambert Playfair, must not be confounded with the Oases of the
Sahara, but is an island detached from the eastern spurs of the Atlas in the ocean
of the desert, is further justified from the fact that its herpetologicel fauna is still too
imperfectly known’.

Natural Divisions of Barbary.

The division of Algeria into three parallel zones, viz., the Tell, the High Plateaux,
and the Sahara, is so familiar as hardly to require a definition. The Tel/ is the region
which borders the Mediterranean, and includes the Lesser Atlas; its fauna is essentially

’ A list of reptiles collected by G. Rhumer around Bengazi in the Cyrenaica, and named by Dr. Reichenow,
was published in 1883 in the Sitzungsb. Ges. naturf. Freunde Berlin (p. 149). The following species are
enumerated :—Chameleon vulgaris, Hemidactylus verruculatus [=turcicus], Agama savignyi, Lacerta muralis,
Acanthodactylus boskianus, A. lineomaculatus, Gongylus ocellatus, Zamenis florulentus, Coronella (Macro-
protodon) brevis [ =cucullata, var.], Naia haie, Bufo variabilis [ =viridis}.

The record of Lacerta muralis is of especial interest ; that of Acanthodactylus lincomaculatus may be due to

confusion with A. pardalis ; and the so-called Agama savignyi is probably A. inermis.

AND BATRACHIANS OF BARBARY. 95

that of the borders of the Mediterranean, and differs but little from that of the Spanish
and Italian peninsulas. The Plateau, consisting mostly of nude steppes with a mean
altitude of 2000-3000 feet, is the zone comprised between the Tell and the Sahara,
and presents in its fauna a mingling of Northern and Southern forms. Geryville,
Laghouat, and Biskra are exactly on the limit of the Plateaux and the Sahara. Towards
Oran the Plateaux extend almost to the sea-coast. The Sahara, south of the Great
Atlas, and extending uninterruptedly from the Atlantic Ocean to the valley of the Nile,
has an altogether special fauna, only a few of the forms of the Tell (Rana esculenta,
Bufo viridis, and Tropidonotus viperinus, for instance) extending into its northern
portion. But this distinction applies only to Algeria; the Plateaux, which to the
west are limited by the north-eastern ramifications of the Great Atlas of Morocco, form
towards the east a wedge which culminates in the east of the Province of Constantine,
so that in Tunisia the Tell passes directly into the Sahara, and we consequently obtain
in Southern Tunisia a mingling of certain forms which in Algeria, separated as they
are by the range of plateaux, are characteristic of either the one or the other district.
The lofty Atlas of Morocco separates the Sahara from the plain of Morocco, which to
the north and north-east passes into the Tell proper and is completely cut off from the
Plateaux ; this plain having a distinct fauna, deserves to be dealt with separately, and
I propose to allude to it as the Moroccan district. Its fauna is still very imperfectly
known, the only places at which herpetological collections have been made being
Casablanca, Mogador, and the route from Mogador to Morocco. A few forms are
known from the valley of Sous, but of the Great Atlas proper we know nothing.
Hooker and Ball, remarking on the scarcity of animal life as a characteristic feature
of the Great Atlas, state that ‘‘of the numerous Reptiles that abound about the skirts
of the mountain-range, few, except Lizards, seem to frequent the interior valleys ; and
the latter are wanting, or at least rare, in the higher region.” Of Kastern Morocco we
likewise know absolutely nothing. But the northern promontory, which bears Tangier,
Ceuta, and Tetuan, has been tolerably well investigated, and forms altogether so distinct
a feature in the fauna of Barbary that it deserves to be recognized as the Twagitanian
district. There we meet with species otherwise peculiar to the Moroccan district,
and with a few endemic forms, but with comparatively few, except the most widely
distributed, that are identical with those of the Algerian ‘Tell; and, what is still more
surprising, no more special affinity is shown to the fauna of Southern Spain than in the
latter district, with the only exception of the occurrence of Molge waltli.

Barbary is therefore divided into five districts, viz.:—1, The Moroccan; 2, the Tan-
gitanian ; 3, the Tell; 4, the Plateaux; 5, the Sahara. In addition to these districts,
the political divisions of Oran, Algiers, Constantine, and Tunisia are adopted to serve as
landmarks in the distribution from West to East of the species shown, as far as present
knowledge permits, in the following Table.

96 MR. G. A. BOULENGER ON THE REPTILES

List of the Reptiles and Batrachians of Barbary, showing their Distribution.
(The names of genera, species, or varieties confined to Barbary are printed in italics.)

A. Moroccan district ; B. Tangitanian district; C. Tell; D. Hauts-Plateaux; E. Sahara; F. Proy. Oran;
G. Proy. Algiers; H. Prov. Constantine; I, Tunisia.

North—South. West— East.
C D E A 1 8 G H I
| Reprires.
* * *% il, Ghesianclo Wea, YO, sooo navccedoneuo0s * 5S * * * *
* 3h 2 May OHM, IZ, socacdooscddo0ca0e a6 Se on * se
* * * 3. Clemmys leprosa, Schw................. * * * * * CS
* * 4, Stenodactylus guttatus, Cuv............. a0 ae * * * *
* 5. ‘Tropiocolotes tripolitanus, Ptrs........... 3.0 9.0 dc O00 dD *
* * * 6. Saurodactylus mauritanicus, D. & B....... * Ro * *
7. Gymnodactylus trachyblepharus, Bttg. ....| *
8. Phyllodactylus europeeus, Gené (Galita Isl.).) .. oh 54 on 30 *
| * * 9. Ptyodactylus lohatus, Goffr............. * * *
| * * * | 10. Hemidactylus turcicus, Z)........:..... oe wh * * * *
sk * x | 11. ‘Tarentola mauritanica, Z............... * * * * Bo *
* oy | meglectan| Strauchymeriietel ir y-teicl errr *
* co | Uy ANeehney mess JAS “Goacecnoo0uosbscc OD 66 20 * * *
| * | 14, tounncucllav alata eral: on ae aus be *
* * 5 a COURT, As IDEN Secbodsoucs000DD * * * *
* * | 16. Uromastix acanthinurus, Bell .......... oa ae * * * *
17. Ophisaurus oellikert, Gthr. ............ *
* * | 18. Varanus griseus, Daud................- 26 a a0 * * *
* * oo) |) IO diem Cherm, Wank ocog oheabobood sc * * * Z *
* * 20. Tregonophis wiegmanni, Kaup ......--.. * * * * * *
| * 21. Lacerta ocellata, var. pater, Lat. ........ ae a * * * *
| 21a , var. tangitana, Blgr. ........ So *
* B60 68 || 225 ———Ssutineli Iiwim ooooomoobU oe od6 ooo a0 * * Ba * *
| 3% 23 perspicillata, D. & B. ............ *
| * * 24, Psammodromus.blanci, Lat. ...........- ae be Aue * * *
25. microdactylus, Bttg. .............. * *
* * * | 26. ell ourushyelsh yer vst vexesrohevenohstevteven rors tenets * * * * * *
* * | 27. Acanthodactylus boskianus, Daud. ...... ** * * *
* * | 28. ScutellatusseAccc mer merie erates ree * * * *
| © * <3 || 2 PaTncalisswAicheaer eee rere: go o-0 % * * **
| * * -. | 30 Wulkernes, VE GP UES Gaooobebocsensoo * * * * *
| * * * | 31. Eremias guttulata, Licht. .............. * * * * *
| * * | 32. Ophiops occidentalis, Blgr. ............ * * *
| * ce || os Wihloume, valine, QO6 caboose accd0s oduots * *
* | 34. Eumeces schneideri, Daud. ...........- é ane * *
* * oo || iy Gig envensis PATS. mein cnet e kone * *
co || Oleh SetCHS uoehney Jee Goonee Gou016 cObC ae a * 0:0 at *
* | 37. Cungmellis, JOM, Gaoosbaodooooabus ar 58 * * * *
* | 38. Chalcides ocellatus, Morsh............... : * *
* * 38a Was ale, (Cris So eoed09 06 is ae * * * *
386. —— 5 aR WAI, IETS. o Ga oc od oN 2.0 *
38¢, —— , var. polylepis, Blgr, ........ *
* 39) lineatusselcuclmMen eet ere * * an a0 *
* * 40. tridactylushwZaur wratterus ova a * * * *
41, MOOMGHOD, WAHWS55 a6 nas60600000000 * *

AND BATRACHIANS OF BARBARY. 97

North—South. West— Kast.
C D E A B. F G H I
* aa os 42. Chalcides mawritanicus, D. & B. .......... hs ate *
x | 43 AG oOas, walneh os ccc c0009 bo0000 KC aC Si ae Bo * * 2
* * x% | 44. Chameleon vulgaris, Daud. ............-. * * * * * *
x % e3 ||| Ab}, Ibinyee eKOMIE, We Ss oceaavousvoccgoo0uo0. ny ns * * *
* 3 ve) p46 Coronellaramulan Biter ances ce seen: a: # *
* eA ericonaliCn, Dm; ooo 50804000000006 * a *
* | 48. Lytorhynchus diadema, D. f B. .......... a0 oc * 50 * %
* ¥ | coe Lamenisnalginuss Jann iu serene cael 50 =e 56 * * *
* * 50 hip POcKepiswe lei em yee eter * * * * ** *
Sea |oile diademaw Schirano * * *
* bet oo || G2b UO Mb NONS MEE, JG “So 5cccuddooooDoC aye Br $0 * *
* * * | 53. ; Woes, ei ooasoouenpoouocdes * * * * * *
* * x | 54. Macroprotodon cucullatus, Geoffr. ........ * * * * * *
* * | 55. Psammophis sibilans, Z., var. punctatus, D.
Gena Fodavenr fei sv shaets) suspcray ora Ayes aah Aeopeeausyenens A ais * * * *
* * * | 56. Coelopeltis lacertina, Wagl. .............. * * * * %
* | 57. (OGM CKD so cdgoocccnesdenec a ie * as *
x | 58. Naia haie, Z., var. annulifera, Pirs. ...... * a6 : ar * *
*% 06 che POOsmaperadlatastitneDoscdimeaeme racist a2 * 6 * *
% * * | 60 Nebe tunately 5: epee eLa core ea eke * ee Ba * *
61 arietansy Verna acnnciiiee ee *
3 | Gey Oana valor, Jas opaocoodoossdoe00a0s bys be * * * *
* * | 63 CormutuseHonsicameeeretenteeiere We ae * * * *
eee (OAs Hichisicarinatasy Scie) aeasciaclc creleterele icicle * *
BarRacHrians.
* * * 1. Rana esculenta, Z., var. ridibunda, Pall.....| * * * * *
* * * DL. Ipiuiyoy aval View oconaooudobadondedono * ss * * * *
* * By, MAUILANICUS SCO Meera here: * * * * * x |
* 4, Vill SATIS Se LZCtiT emer Rare errerstcr nae Ae * * * *
* vie 5. Hyla arborea, Laur., var. meridionalis, Bttg.| * Ba * * * *
* * 6. Discoglossus pictus, Olth .............-.. * * * * x *
* 7. Salamandra maculosa, Laur., var. algira, Badr. * * * *
* 40 Bh Mla jratRaw, Came odie bonoscsenbooadodo * * ee
* * o), Ragenmucllerd slaty yn teraiteneer cries ot MM a *
10. Waltliteiache er sracuinera criemiererrenhs ae a *
Bibliography.

In the work of

1. Saw, J. Travels, or Observations relating to several parts of Barbary and the
Levant. Oxford: 1738, 4to,

we find the first reference to the ‘‘ Oviparous Quadrupeds” of Algeria (pp. 249-250),
which, although not described, are mostly identifiable, viz., as Testudo ibera (‘ The
Land-Tortoise”), Clenmys leprosa (“The Water-Tortoise”), Chameleon vulgaris (“The
Chameleon”), Varanus griseus (“The Warral”), Uromastix acanthinurus (“* The
Dab”), Lacerta ocellata (‘The Common Green Lizard”), Psammodromus algirus and
P. blanci (“The Zermoumeah ”), and Tarentola mauritanica (*'The Niji-daimah ”).

98 MR. G. A. BOULENGER ON THE REPTILES

But the first list according to the system of Linneus is given in the first volume of
2. Porret. Voyage en Barbarie. Paris: 1789, 8vo,
in which a chapter (pp. 283-290) is devoted to the “ Animaux Amphibies ” of Algeria.
In addition to a marine turtle said to be very common in the Mediterranean, and
which therefore could hardly have been the Testudo coriacea of Linneus to which
Poiret refers it, the following species are enumerated :—

1. Testudo greca, L. (=T. ibera, Pall.).

. Lacerta agilis, L. (=L. ocellata, var. pater, Lat.).
. Lacerta algira, La. (=Psammodromus algirus, L.).
. Lacerta chameleon, L. (= Chameleon vulgaris, Daud.).
. Lacerta chalcides, L. (= Chaleides tridactylus, Laur.).
. Lacerta vulgaris, L. (Unidentifiable—L. vulgaris, L.=Molge vulgaris.)
. Lacerta palustris, L. (= Molge poireti, Gerv., or M. hagenmuelleri, Lat.).
. Lacerta salamandra, L. (= Salamandra maculosa, Lauy.).

aI OD oF OO LD

ie)

The next list is supphed by
3. Rozet, M. Voyage dans la Régence d’Alger. Paris. 1833, 8vo,
who adds (vol. i. pp. 230-233) the following species not recorded by Shaw or Poiret :—
. Rana esculenta, L.
. Rana temporaria, L. (= Discoglossus pictus, Otth).
. Bufo, sp. (=Bufo mauritanicus, Schleg.).
Coluber natrix, L., var. (=Tropidonotus viperinus, Latr., var.).
. Coluber monspéssulanus, Herm. (= Celopeltis lacertina, Wagl.).
. Coluber hippocrepis, L. (= Zamenis hippocrepis, 1.).
. Vipera daboia, Lacép. (=? V. lebetina, L.).
. Tiliqua ocellata, Gray (=Chalcides ocellatus, Forsk.).

ota oe oo be

The nomen nudum of Lacerta viridissima is bestowed on L. ocellata, and the “ Niji-
Daimah ” of Shaw receives the name of Platydactylus fascicularis, Cuv.

4. Gmrvais, P. Enumération de quelques espéces de Reptiles provenant de Barbarie.
Ann. Sci. Nat. (2) vi. 1836, pp. 308-313.

Gives alist of 27 species from Morocco and the Province of Algiers, which, together
with a revised nomenclature, adds 15 species, of which as many as five have to be
erased, viz., Testudo marginata, Schoeptf, as undoubtedly due to confusion with old
specimens of 7. dbera, aud Anguis fragilis, L., A. punctatissimus, Bibr., Pseudopus
serpentinus, Merr., and probably Coluber agassizit, Mich., as based on erroneous state-
ment of locality (see p. 114).

The additions are :—

1. Gecko (Hemidactylus) verruculatus, Cuv. (=turcicus, L.).
2. Gymnodactylus mauritanicus, D. & B. (Saurodactylus).—Algiers.

AND BATRACHIANS OF BARBARY. 99

. Lacerta agilis, L. (This I suppose to be meant for L. muralis.)—Algiers.

. Lerista dumerilii, Coct., MS. (= Chalcides mauritanicus, D. & B.).—Algiers.

. Scincus cyprius, Cuv. (= Eumeces algeriensis, Ptrs.).—Algiers.

. Amphisbena cinerea, Vand.—-Tangier.

. Amphisbena elegans, Gerv. (=Trogonophis wiegmanni, Kaup).—Tangier,
Zaffarine Islands.

8. Coluber austriacus, L. (undoubtedly = Macroprotodon cucullatus, Geoftr.).—

Algiers, Tangier.

I mG OF Bm OS

The Tortoises receive the correct names of Testudo ibera and Emys leprosa; the
“Dab” that of Uromastix acanthinurus, Bell; L. ocellata, var. pater, appears as
L. viridis, Celopeltis as Coluber csculapii; the variety of Tropidonotus viperinus,
alluded to by Rozet, receives the name of var. awrolineatus, Gerv.; a toad from Bona
is named Bufo arabicus, Riipp.; and the name Triton poireti is proposed for the
Lacerta palustris of Poiret.

Several of the Reptiles recorded by Gervais from the province of Algiers have ulti-
mately proved not to occur in that part of Algeria.

Gervais’s list was shortly followed by a contribution by Schlegel, published as an
Appendix to the first volume of

5. Waener, M. Reisen in der Regentschaft Algier. Leipzig: 1841, 8vo. Bemerkun-
gen iiber die in der Regentschaft Algier gesammelten Amphibien, von SCHLEGEL.
Pp. 106-139.

The following species are added :—
1. Stenodactylus guttatus, Cuv.

. Lacerta guttulata, Licht. (Hremias).

. Lacerta pardalis, Licht. (Acanthodactylus).

. Vipera echis? and brachyura, Cuv., from Oran (= J. lebetina, L.).

Vipera cerastes, L., from Biskra (= Cerastes cornutus, Forsk.).

. Hyla arborea, L.

S Or Be OF bo

The true Lizards are referred to under the correct names of Lacerta ocellata and
L. muralis, Macroprotodon stands as Coronella levis, Discoglossus is referred to as
Rana picta, and the new name Bufo mauritanicus is bestowed on the Moorish toad.

A second list by
6. Gervais, P. Sur les Animaux vertébrés de l Algérie. Ann. Sc. Nat. (3) x. 1848,

pp. 204, 206,
enumerates 47 species from Algeria, mostly from specimens received by the Paris
Museum, which number, however, is reduced to 40 when we omit synonyms. Of
these 40 species, 7 are recorded for the first time in any list dealing specially with the

100 MR. G. A. BOULENGER ON THE REPTILES

fauna of Barbary, although several had already been noticed by Duméril and Bibron,
‘ Erpétologie Générale’ :—
1. Agama colonorum, Daud. (=. bibroni, A. Dum.).—From the Chotts in
the Province of Oran.
Acanthodactylus lineomaculatus, D. & B. (originally described from -
Morocco).
_ A. scutellatus, Aud.—From the Souf.
A. boskianus, Daud.—From Dbiskra.
. Sphenops capistratus, Fitz. (= Chalcides sepoides, Aud.).—From the Souf.
. Coluber natrix, L. (Tropidonotus).
. Bufo variabilis, Pall. (=B. viridis, Laur.).

bo

“IO OP 09

7. Guicuenot, A. Exploration scientifique de l’Algérie pendant les années 1840-1842.
Reptiles. Paris: 1850, 4to. 30 pp., 4 plates.

This work affords but a meagre list of species, and the descriptions have but little
original value, being to a great extent compiled from the ‘ Erpétologie Générale.’ And
sufficient care was not taken in recording the localities. The following species have
to be added to those which we have mentioned before :—

. Cistudo europea, Schn. (=Emys orlicularis, L.).
. Lacerta perspicillata, D. & B.—From Oran.

. Acanthodactylus vulgaris, D. & B.

. Bufo vulgaris, Laur.

Vipera lebetina, L., is described as Echidna mauritanica, sp. n.; Macroprotodon
cucullatus, Geoftr., as M. mauritanicus, sp.n.; and Molge potreti, Gerv., as Huproctus
rusconit, Gené, and Triton nebulosus, sp. n.

H

bo

H 9

8. Ercuwatp. Naturhistorische Bemerkungen iiber Algiers und den Atlas. Nouv.
Mém. Soc. Nat. Mose. (2) ix. 1851. (Rept. and Batr. pp. 414-444.)
This paper contains but a single addition to our list, viz. :—
1. Agama agilis, Oliv. (=inermis, Reuss).

9. Gervais, P. Sur quelques Ophidiens de Algérie. Mém. Ac. Sc. Montpellier, i.
1857, pp. 511 and 512, pl. v.
Adds 4 species :—
. Heterodon diadema, D. & B. (Lytorhynchus).—From the Sout.
. Psammophis punctatus, D. & B.—From the Western frontier of Alzeria.
. Zamenis florulentus, Schl. (=algirus, Jan).—From Laghouat.
. Celopeltis productus, sp. n.—From between Bou-Alem and the Arbas.

i

em OF BS

10. Ginter, A. On the Reptiles and Fishes collected by the Rev. H. B. Tristram in
Northern Africa. Proc. Zool. Soc. 1859, pp. 469-474, with notes by Tristram
himself, pp. 475, 476.

AND BATRACHIANS OF BARBARY. 101

Enumerates only 12 species, all of which were already on record. Uromastix spinipes,

Daud., stands for U. acanthinurus, Bell, and Zootoca deserti, sp. n., for Acanthodactylus
pardalis, Licht.

The list of * Reptiles of the Sahara” (53 species), which forms Appendix VI. to

11. Tristram, H. B. The Great Sahara: Wanderings South of the Atlas Mountains.
London: 1860, 8vo,

is carelessly compiled, and includes the names of many species ‘not found by the
author, but given on the authority of trustworthy local naturalists.” And it is to be
noticed that the author restricts the term. “Sahara” to the High Plateaux and the
northernmost part of the desert, using the term “‘ Desert” for the greater portion of
what is commonly called the Sahara. This explains how examples of several species
which are known not to extend south of the High Plateaux are labelled in the British
Museum as from Mr. Tristram’s Collection made in the Algerian Sahara. This list
contains Naia hate, which has since been found round Biskra by M. Lataste, and, very
curiously, a Salamandra—? from Tuggurt. However, no information is given as to
the source of these entries.

Such was the state of our knowledge of the Herpetological Fauna of Algeria, when

Dr. Strauch published an excellent and most conscientious descriptive Catalogue,
entitled :—

12. Srraucu, A. Essai d’une Erpétologie de Algérie. Mém. Ac. St. Pétersb. (7) iv.
no. 7, 1862, 86 pp.

This work is based upon the previous literature, the collections of the ‘ Exposition
permanente’ in Algiers, and the author’s own collections made during several months’
stay in Algeria. A great many of the species enumerated on the authority of others
have to be erased, as based on errors of determination or of localities, so that of the 76

species included in the list only 56 (including two marine Turtles) deserve to stand.
The additions are :—

1. Euprepes vittatus, Oliv. (Mabuia).—Mzab.

. Eryx jaculus, L.—Oran.

. Coronella austriaca, Laur. (=*% C. amalie, Boettg.).

. Zamenis cliffordii, Schleg. (=diadema, Schl.).—Sahara.

. Vipera avicenne, Alp. (= Cerastes vipera, L.).—Western Sahara.
. Vipera aspis, L. (=latastii, Boscd).—Algiers.

oO of OO LD

13. Lattemant, C. Erpétologie de l Algérie. Paris: 1867, 8vo. 41 pp.

This opuscule is principally an abridgment of Strauch’s ‘ Essai,” with notes on the
author’s own collections. |

Thus far, with the exception of Gervais’s first list, the centributions to our knowledge
VOL. XIII.—PART 11. No. 2.—October, 1891. R

102 MR. G. A. BOULENGER ON THE REPTILES

of the Reptiles and Batrachians of Barbary had dealt merely with Algeria. The first
paper containing a special account of the herpetological fauna of Morocco was published
in 1874 :—

14. Borrrerr,O. Reptilien von Marocco und von den Canarischen Inseln.—I. Ueber-
sicht der von den Herren Dr. C. von Fritsch und Dr. J. J. Rein im Jahre 1872 in
Marocco gesammelten Reptilien. Abh. Senckenb. Ges. ix. 1874, pp. 121-170,
pl. i. (With Appendix, op. cit. xi. 1877, p. 1, footnote.)

19 Reptiles and 1 Batrachian are therein described, and by adding the species
previously noticed by Gervais and in various general works, the total number of
Moroccan Reptiles is raised to 24, of Batrachians to 3, the following being the
additions :—

. Gymnodactylus trachyblepharus, sp. n.

. Seps mionecton, sp. n. (Chalcides).

Coronella girondica, Daud.

. Vipera arietans, Merr.—(Valley of Sous.)

. Naia haie, L. (fide A. Dum.).

. Pleurodeles waltlii, Mich. ( fide Schreiber) (Molge).

Doniwm © bp

A short paper by

15. Camerano, L. Osservazioni intorno agli Anfibi Anuri del Marocco. Atti Acc.
Torin. xiii. 1878, pp. 542-558,

contains descriptions of Rana esculenta, L., Discoglossus scovazzii, sp. n. (=D. pictus),

Bufo vulgaris, Laur., from Larache, new to Morocco, B. pantherinus, Boie (=B. mauri-

tanicus), and Hyla arborea, L., from specimens collected by the Italian Consul Scovazzi

at Tetuan, Tangier, Larache, Casablanca, Babat, Masagan, Saffi, Mogador.

With M. Lataste’s excursions to Algeria in 1880 and 1881 a fresh era was started.
Large collections were made both north and south of the Atlas; the discrimination of
species was subjected to a severe test upon fresh material; and an investigation into
the literature resulted in the elimination of several forms unduly recorded as from
Algeria, the amalgamation of certain species previously regarded as distinct, the
separation of others previously confounded, the discovery of altogether new forms, and
above all in a much clearer understanding of the distribution of the various forms, as
will be apparent on comparing the present Catalogue with that of Dr. Strauch.

As stated in the introduction, M. Lataste was prevented from putting his documents
into final shape, and his published contributions are merely the following :—

16. Lavasrz, F. Descriptions de Reptiles nouveaux d’Algérie. Le Naturaliste, i. 1880
and 1881.

AND BATRACHIANS OF BARBARY. 103

Zerzoumia blanci, sp. n., p. 299 (1880).

Ptyodactylus oudrit, sp. n., p. 299 (1880).

Lacerta ocellata pater, subsp. n., p. 306 (1880).

Agama tournevillet, sp. n., p. 325 (1880).

Acanthodactylus bedriagat, sp. n. (=A. pardalis, Licht.), p. 357 (1881).
Glossoliga hagenmuelleri, sp. n., p. 371 (1881).

Also recording for the first time the presence of Ophiops and Naia haie in Algeria,
and giving notes on a supposed Simotes? in the collection of the ‘ Exposition
permanente. The latter snake, which is possibly not from Algeria, I have been
unable to identify from M. Lataste’s notes, unless it be a dark variety of Zamenis
gemonensis (viridiflavus).

17, Laraste, F. Liste des Vertébrés recueillis par M. le Dr. André pendant l’expé-
dition des Chotts. Arch. Miss. Sc. (3) vii. 1881, pp. 398-400.
A list of 22 species obtained in Southern Tunisia and the neighbouring part of
Algeria by the Roudaire expedition.

18. Borrrerr, O. Liste der von Herrn Dr. W. Kobelt in der Prov. Oran, Algerien,
gesammelten Kriechthiere. Ber. Senckenb. Ges. 1880-81, pp. 145-147.
Notices 13 species, among which Coronella girondica and Bufo vulgaris from Tlemsen
are of special interest.

Turning his attention once more to the Reptiles of Morocco, Dr. Boettger supplies a
most valuable contribution :—

19. Borerrerr, O. Die Reptilien und Amphibien von Marocco. II. Abh. Senckenb.
Ges. xiii. 1883, pp. 93-146, pl. i.

Based chiefly upon collections brought together by Lieut. Quedenfeldt and Dr. W.
Kobelt. Leaving out 4 species which Boettger introduces on the authority of
other authors and which should evidently be erased, we see that the list of Moroccan
Reptiles has risen to 37 species, of which the following are recorded for the first
time :—Lhinechis amalic, sp. n. (Coronella), Vipera euphratica, Mart., var. mauritanica,
D. & B. (=V. lebetina, L.), Vipera latastii, Bosca, Lacerta muralis, Laur., Algira
nucrodactyla, sp. n. (Psammodromus), Podarces simont, sp. n. (=Eremias guttulata,
Licht.), Pseudopus apus, var. n. ornata (=Ophisaurus koellikeri, Gthr.), and Bufo
viridis, Laur.

20. Borrraer, O. Liste der von Hrn. Dr. W. Kobelt in Algerien und Tunisien
gesammelten Kriechthiere. Appendix to W. Koserxt, Reiseerrinerungen aus
Algerien und Tunis. Frankfurt/M.: 1885, 8vo, pp. 457-478.

Notes on 30 species, all previously recorded from Algeria and Tunisia.
R2

104 MR. G. A. BOULENGER ON THE REPTILES

21. Boutencer, G. A. On the Reptiles and Batrachians obtained in Morocco by
Mr. Henry Vaucher. Ann. & Mag. N. H. (6) iii. 1889, pp. 303-307.
A list of 23 species from the vicinity of Tangier.
New to Morocco :—Lacerta ocellata, var. tangitana, Blgr., Chalcides lineatus, Leuck.,
Salamandra maculosa, Laur.

Cataloque of the Reptiles and Batrachians.

REPTILIA.
The Reptiles with which we have to deal in this Fauna belong to two Orders :—

I. CHELONIA, Tortoises and Turtles, in which the body is encased in a bony shell and the jaws
are destitute of teeth.
II. SQUAMATA, Lizards, Chameleons, and Snakes, with the body covered with scales and the
jaws armed with teeth. They are divided into three Suborders. ;

Order f. CHELONIA.

Apart from the Marine Turtles, three genera of Chelonia, each represented by a
single species, occur in Barbary, which are easily distinguished :—

Mimbstcelub=shapedisaterrestrialtan wiser) enn ee ae LE SLU DOE
Digits webbed ; aquatic.
Plastron connected with the carapace by ligament . ...... =. =. . 2. Emys.
Plastron united with the carapace by suture 3. CLemMys

All three of which. belong to a single family.

Fam. 1. TESTUDINIDA.

1. Trstupo, Linneus, 1766.

Shell very convex. Head covered with horny shields. Alveolar surface of upper
jaw with ridges. Limbs club-shaped. Terrestrial and herbivorous.

1. Testupo 1BERA, Pallas, 1831.
T. greca, Poiret, Rozet.—T. mauritanica, Guichenot.—T. pusilla, Strauch.

Supracaudal marginal shield undivided, never spread out and subhorizontal. Hind
lobe of plastron movable in the adults; suture between the humeral plastral shields
much longer than that between the pectorals ; suture between the anal shields nearly

AND BATRACHIANS OF BARBARY. 105

as long as, or longer than, that between the humerals. Anterior face of fore limbs
with large imbricate scutes, forming four or five longitudinal and five or six transverse
series; a large convex or subconical tubercle on the hind side of the thigh. Carapace
of young yellowish or pale olive, each shield spotted and bordered with black ; the
black spots more irregular and predominating in the adult ; some specimens uniform
brownish or olive ; plastron more or less spotted with black.

Length of shell 25 centim.

Common throughout the Tell, from Morocco to Tunisia ; also found on the Algerian
Plateaux, and even further south in Tunisia. M. Lataste found it at Palestro,
Misserghin (Oran), Aumale, Guyotville (Algiers), Salah-Bey (Constantine), and Bou-
Saada, in Algeria; at Zarzis and the ruins of Utique in Tunisia. It also inhabits
South-western Asia.

For good figures of this tortoise we may refer to Bell’s ‘ Monograph of the Testudi-
nata’ (sub nom. 7. grwca), and to Lortet, Arch. Mus. Lyon, iv. 1887, pl. 1.

Old specimens have been taken for the allied T marginata, Schoepff, s. campanulata,
Strauch (by Gervais and Lallemant), the habitat of which appears to be restricted to
Greece.

2. Emys, Duméril, 1806.
Shell depressed. No shields on the head. Alveolar surface of upper jaw without
ridges. Digits distinct, webbed. Plastron joined to carapace by ligament, and divided
into two movable lobes in the adult. Aquatic and carnivorous.

1. Emys orsicuLaris, Linneeus, 1766.
Cistudo europea, Guichenot.—C. lutaria, Strauch.

Carapace with yellowish dots or radiating lines on a dark ground; head dark brown
or black above, with yellow or pale brown dots, yellow inferiorly, spotted with black.

Length of sheli 19 centim.

Inhabits Southern and Eastern Europe, Western Asia, and Algeria north of the
Atlas. In spite of Guichenot’s statement that it is found in all the rivers of Algeria, a
statement evidently due to a confusion with Clemmys leprosa, this tortoise is very
locally distributed in Algeria, the only specimens examined by M. Lataste having been
obtained by Dr. Hagenmiiller near Bona. Lallemant records it. from Harrach, Lake
Fetzara, and Wed Sebaon.

Figures: Bonaparte, Faun. Ital. (Zmys lutaria), and Lortet, Arch. Mus. Lyon, iv.
1887, pl. vi.

3. CLtemmys, Wagler, 1830.

Shell depressed. No shields on the head. Alveolar surface of upper jaw without
ridges. Digits distinct, webbed. Plastron without hinge, united to carapace by suture.
Aquatic and carnivorous.

106 MR. G. A. BOULENGER ON THE REPTILES

1. CLemmys LepRosa, Schweigger, 1814.
Emys vulgaris, Schlegel.—E. sigriz, Guichenot.

Carapace dark olive in the young, with an oval orange spot or short longitudinal
streak on each shield; uniform olive, or with traces of the orange spots in the adult.
Head olive, sides with orange or yellow streaks or vermiculations, and a round orange
spot between the eye and the ear, and a more or less defined ring of the same colour
round the latter. Neck and limbs with orange or yellow streaks. These bright
markings become very indistinct in old specimens.

Length of shell 20 centim.

Inhabits the South of Spain and Portugal, Barbary, and Senegambia. Common
throughout Morocco, Algeria, and Tunisia as far as the northern border of the Sahara.
M. Lataste’s Algerian specimens are from the Wed Zig, near Oran (where he found it
in extraordinary abundance), L’Arba, Bona, Batna, and Bou-Saada; he found it in
Tunis at Cabes, and his colleague M. Valéry-Mayet at Wed Leben and Gafsa.

Figures: Gray, Cat. Sh. Rept. i. pl. ix., and Proc. Zool. Soc. 1860, pl. xxx., 1869,
pls. xxxvil. & 1.

Order Il. SQUAMATA.

Divided into three Suborders :—

I. LACERTILIA, Lizards.—Nasal bones entering the border of the nasal apertures; pterygoid in
contact with quadrate; mandibular rami united by suture. Pectoral arch or its vestiges
present ; clavicle present whenever the limbs are developed. Tongue flattened.

II. RHIPTOGLOSSA, Chameleons.—Nasal bones not bounding nasal apertures; pterygoid not
reaching quadrate; mandibular rami united by suture. Clavicle absent ; limbs well developed.
Tongue vermiform, projectile. .

lil. OPHIDIA, Snakes.—Nasal bones bounding nasal apertures ; mandibular rami connected by
ligament. No trace of pectoral arch. Tongue flattened and bifid at the end, and sheathed at
the base.

Suborder 1. LACERTILLA.

The Lizards of Barbary belong to seven Families, distinguishable as follows :—

I. Tongue smooth or covered with villiform papille, feebly nicked at the
end, not retractile into a basal sheath.
Head covered with small scales; eyes without movable lids, with vertical

pupil : 0 6 anes AY ge ay ae . l. Geckonide.
Head covered with small sisal’ 3 eyes one EAE lids, with Hee pupil 2. Agamide.
Head with symmetrical shields. . . . j . . 9&8. Anguide.
II. Tongue very long and slender, bifid, sosertall into a Disall eine 4. Varanide.

AND BATRACHIANS OF BARBARY. 107

III. Tongue bifid, covered with rhomboidal, imbricate papilla; head with
symmetrical shields.

Body vermiform ; eyes hidden . 5. Amphisbenide.
Femoral pores ¢ : : 6. Lacertide.
No femoral pores ; scales eet Gaueente 7. Scincide.

Fam. 1. GECKONIDZ.

The 8 genera by which this family is represented in Barbary may be distinguished
by means of the following key :—

I. Digits not dilated.
A. Digits straight, of equal diameter throughout.
1. Digits deuticulated laterally and keeled inferiorly.

Scales small : 1. SrENopAcTYLUS.
Scales large and imbricate . 85 2. TRoprocoLorEs.
2. Digits not denticulated Tater ally, hh smooth inineity inferior 3. SAURODACTYLUS.
B. Digits flattened at the base, compressed at the end 4, GyMNODACTYLUS.
II. Digits dilated at the apex only.
Digital expansion inferiorly with two plates . . . - . » « 5. PHYLLODACTYLUS.
Digital expansion inferiorly with two diverging series of feces . . . - 6, Pryopacryius.
III. Digits dilated at the base or throughout.
All the digits clawed, the claw supported by a free, compressed joint ; sub-
digital lamelle in pairs . . ‘ . . . 4%. HEMrIpacryLus.
Third and fourth digits with a dase ee ; shied Taineus entie . . 8. TaREnro.a.

1. SrenopactyLus, Fitzinger, 1826.

Digits not dilated, furnished with a long claw, and a lateral fringe or denticulation
of pointed scales ; inferiorly with a series of keeled scales. Scales juxtaposed or sub
imbricate. Pupil vertical. No preeanal or femoral pores.

Sand-Geckos, represented by one species in Algeria and Tunisia.

1. SrenoDactTYLus GuTratus, Cuvier, 1829.

S. mauritanicus, Guichenot.

Head very variable in shape; snout rounded or more or less pointed. Body short,
limbs long and slender. Body covered with subequal granules, which may be convex,
smooth, or slightly keeled, or flat and subimbricate on the back; the size of these
granules varies considerably. Nostril pierced in the middle of a more or less distinct
swelling, between three nasals, the first labial, and usually also the rostral; no chin-
shields. Tail covered with small juxtaposed keeled scales. Light buff or brownish
above, with round whitish spots between a brown network, sometimes with ill-defined
brown cross-bands; tail with brown annuli; white inferiorly.

From snout to vent 58 millim., tail 40.

108 MR. G. A. BOULENGER ON THE REPTILES

Material received since the publication of the British Museum Catalogue of Lizards
induces me to adopt M. Lataste’s view (i litt.) that S. wilkinsonii, Gray, which I
ventured to keep distinct from S. guttatus, and which has since been recorded from
Batna by Dr. Strauch, is not specifically distinct.. In fact the variations, both of scaling
and of proportions, are as great in this species, and of the same kind, as in Piyodactylus
lobatus. The form that I regard as the typical S. guttatus has a moderately pointed
snout, the dorsal granules are rather large, convex and coarsely granular, the rostral
shield enters the nostril, and the hind limb reaches barely the axil. ‘Stouter specimens
with shorter snout have been named S. mauritanicus. In Gray’s S. wilkinsonii the
snout is more pointed, the dorsal scales flat and subimbricate, the rostral is excluded
from the nostril, and the hind limb reaches the shoulder. All these differences, how-
ever, break down, as specific characters, on examination of large series of specimens,
and I do not even see my way to distinguishing the three forms as varieties. A speci-
men from Bou-Saada, collected by M. Lataste, has the short head and short limbs of
S. mauritanicus ; the dorsal scales very small, flattish, smooth, but not imbricate; and
the nostril well separated from the rostral.

Stenodactylus guttatus ranges from the Algerian Sahara to Egypt, Arabia, and Syria.
It extends also into the Algerian Plateaux (Batna, Strauch; Bou-Saada, Lataste).
M. Lataste found it at Wed Dermel, Laghouat, and Bou-Saada, and received specimens
from Mraier and Wargla. Gervais records it from the Souf, and Guichenot from the
province of Oran. Dr. André obtained it in the region of the Chotts, at Bir Knafes
and Bled Berrada. In Tunisia, M. Lataste found it at Houmt-es-Souk (Djerba Island)
and at Wed el Ftour (south of Cabes), and his colleagues M. Valéry-Mayet at Sfax,
Kerkenna Island, and Bou-Hedma, and M. Sédillot at Feriana and Gafsa and Kriz.

Good figures of this species are given in the Expédition d’Egypte, Rept. pl. v. fig. 2,
and by Guichenot in Explor. Alg., Rept. pl. i. fig. 1.

2. Troproco.ores, Peters, 1880.

Digits not dilated, furnished with a long claw, and denticulated laterally; inferiorly
with a series of keeled scales. Scales rather large and imbricate. Pupil vertical. No
preanal or femoral pores.

This genus comprises two species :—T’. tripolitanus, from Tunisia and Tripoli, and
T. steudneri (Gymnodactylus steudneri, Peters, 1869, Stenodactylus petersii, Blgr.,
1885), from Egypt and the Sennaar.

1. TROPIOCOLOTES TRIPOLITANUS, Peters, 1880.

Body and limbs rather slender, covered with imbricate, rhomboidal, keeled scales ;
42 to 44 scales round the middle of the body. Nosiril pierced between the rostral,
the first labial, and three nasals; mental followed by a pair of chin-shields. Tail
tapering to a fine point, and considerably ionger than head and body Colour above

AND BATRACHIANS OF BARBARY. 109

sandy, with small brown spots; a brown streak on the side of the head and neck,
passing through the eye; lower parts white.

From snout to vent 66 millim., tail 38.

The type specimens are from Wadi M’bellem, Tripoli. The same Lizard has since
been found in Tunisia; one specimen at Taferma by M. Letourneux, one between
Cabes and Gafsa by M. Sédillot, one at Oum-ali, near Gafsa, and another at Bou
Hedma by M. Valéry- Mayet.

Described in detail and figured by Peters, Mon. Berl. Ac. 1880, p. 306, pl. —. fig. 1.

3, SAURODACTYLUS, Fitzinger, 1843.

Digits not dilated, clawed, not denticulated laterally, inferiorly with a series of
smooth lamelle. Dorsal scales small, subimbricate. Pupil vertical. No femoral or
preeanal pores.

This genus is very closely allied to Zropiocolotes, differing mainly in the smooti
subdigital lamelle, the absence of a distinct lateral digital denticulation, and the smaller
dorsal scales. A single species is known, which was formerly referred to the genus
Gynnodactylus.

1. SavRopactyLUS MaURITANIcUS, Dum. & Bibr., 1836. (Plate XIII. fig. 1.)

Habit lacertiform. Snout subacuminate; ear-opening small, roundish-subtriangular.
Nostril pierced between the rostral, the first labial, and three nasals ; five or six upper
and four or five lower labials ; mental followed bya pair of chin-shields. Scales on the
head small and granular, the granules largest on the snout ; dorsal scales small, smooth,
flat, roundish, subimbricate, increasing in size towards the belly, where they are large,
roundish-hexagonal, and imbricate ; about 70 scales round the middle of the body.
Tail a little longer than head and body, tapering to a fine point; caudal scales cycloid-
imbricate, with the median inferior series transversely enlarged. Grey-brown above,
with small white dark-edged ocelli; a dark brown streak on the side of the head,
passing through the eye; lower parts white; tail yellowish with brown spots, or
brownish with round yellow spots.

From snout to vent 30 millim.

This species is now represented in the British Museum by seven specimens collected
at Mogador by Lieut. Quedenfeldt. Boettger has recorded it from Djebel Hadid, near
Mogador, from between Mogador and Morocco, and from the Plateau of Chiodma.
Duméril and Bibron’s statement that the type specimen came from Algiers requires
confirmation. Two specimens were obtained at Nemours (Prov. Oran) by M. Gazagnaire
in 1888, and are now in M. Lataste’s collection. Strauch saw specimens from the
Algerian Sahara in the Loche collection, and F. Miiller records the species from the
Plateau of Sersou, in the Province of Algiers.

VOL. XIII.—PART Ul. No. 3.—October, 1891. s

110 MR. G. A. BOULENGER ON THE REPTILES

4, GymnopactyLus, Spix, 1825.

Digits not dilated, clawed, not denticulated, cylindrical or depressed at the base,
compressed in the distal portion, which forms an angle with the basal; inferiorly with
a series of smooth lamelle. Dorsal scales juxtaposed. Pupil vertical. Males usually
with femoral or preeanal pores.

This large genus is represented in Barbary by a single species from South-western
Morocco.

1, GYMNODACTYLUS TRACHYBLEPHARUS, Boettger, 1874.

Head much depressed ; snout rounded, a little longer than the distance between the
eye and the ear-opening ; latter transversely oval. Limbs rather slender ; digits elon-
gate, depressed in the basal, compressed in the distal portion. Above uniformly and
finely granular, the eranules largest on the snout; upper eyelid with several pro-
jecting tiiangular scales on its free border; rostral pentagonal, nearly twice as broad as
deep, with median cleft above ; nostril between the rostral, the first labial, and four
nasals; seven upper and six lower labials; mental large, subtriangular; no regular
chin-shields. Ventral scales large, hexagonal, subimbricate. ‘Tail slender, depressed,
covered above with uniform small scales, beneath with a median series of enlarged
transverse plates. Greyish olive above, whitish beneath; tail with rather indistinct
yellowish cross-bands.

From snout to vent 40 millim., tail 57.

A single specimen is known, from Djebel Hadid, near Mogador, which, thanks to
Dr. Boettger’s courtesy, I have been able to examine.

Figured by Boettger, Abh. Senck. Ges. ix. 1874, pl. i. fig. 5.

5. PHYLLODACTYLUS, Gray, 1850.
Digits all clawed, the extremity dilated, with two large plates inferiorly separated by
a longitudinal groove in which the claw is retractile. Pupil vertical. No preanal or
femoral pores.

1. PHYLLODACTYLUS EUROPaUS, Gené, 1839.

Upper surfaces covered with equal small smooth granules; no regular chin-shields,
but very small polygonal scales passing gradually into the minute granules of the
throat. Tail prehensile, covered with equal small squarish scales. Grey-brown above,
marbled with darker and dotted with lighter ; a more or less distinctly marked dark
streak on the side of the head, passing through the eye ; lower parts whitish.

From snout to vent 40 millim., tail 30.

This small Gecko inhabits many of the Islands of the Mediterranean west of Italy,
and was found on Galita by Marquis Doria.

Figured in Bonaparte’s ‘ Fauna Italica.’

AND BATRACHIANS OF BARBARY. ul

6. Pryopactyius, Gray, 1825.
Digits all clawed, the extremity dilated, with two diverging series of lamelle

inferiorly disposed somewhat like a fan; the claw retractile in the anterior notch of
the distal expansion. Pupil vertical. No preeanal or femoral pores.

1. Pryopacryius Lozatus, Geoffroy, 1809. (Plate XIII. fig. 2.)

Head large ; body rather short; limbs long and slender. Upper surface of body
and limbs covered with granules intermixed with small keeled tubercles ; lower surfaces
with flat smooth scales. ‘Tail slender, tapering. Greyish or yellowish brown above,
spotted with darker ; lower parts white. ;

Ranges from Algeria to Egypt, Nubia, Arabia, and Syria.

In Algeria this species is found in the stony Sahara, and in the southern parts
of the Plateaux. M. Lataste collected specimens at Ghardaia, Laghouat, Bou-Saada,
between Bou-Saada and Biskra, and between Biskra and Batna, and received a specimen
from Djenian bon Resk, near Oran, through Dr. Maury. Dr. Strauch also records it
from Batna (‘ Bemerk. tiber die Geckoniden-Samml. St. Petersb. Mus.’ p. 55, 1887).

The following is M. Lataste’s description of Algerian specimens, for which he
proposed in 1880 the name P. oudrii:—“ Smaller and less slender than the Egyptian
Ptyodactylus. When the arm is stretched forwards in the latter, the wrist reaches
the nostril, whilst in the new species it reaches only midway between the eye and the
nostril. When the hind limb is stretched forwards the heel reaches the axil in the
former, but far from it in the latter. Length of the head twice in the length of the
trunk in P. owdrii, twice and a half in the other species. The largest specimen measures :
head 17 millim., from snout to vent 55, tail 57. Scaling generally coarser in P. ouwdri/,
the granules as well as the tubercles larger. ‘The ventral scales, very small and almost
granular in the Egyptian species, are comparatively large, distinctly hexagonal, and
quite flat in the Algerian; this difference particularly conspicuous on the lower belly,
where, as well as under the thigh and leg, the scales reach a considerable size in
P. oudrii, comparable to that of the sublabials and chin-shields. In the Egyptian
species the chin-shields are immediately followed by minute granules, smaller than on
the breast, whilst in P. oudrii the anterior gulars are larger than the following, which
gradually decrease in size. Finally I count 10 distinct sublabials in the Egyptian
species, and only 7 or 8 in the Algerian.”

I am, however, unable to agree with my friend in separating these Algerian specimens
as a distinct species. The difference in the number of lower labials does not hold good.
There is every gradation between P. owdrii and FP. lobatus, and some Syrian specimens
are undistinguishable from the former, except in their larger size. I think the following
notes, taken from all the specimens in the British Museum, are sufficient to show that
P. oudrvi cannot be regarded as more than a variety of P. lobatus.

112 MR. G. A. BOULENGER ON THE REPTILES

1-5. ¢ & her. Bou-Saada (Lataste). Types of P. oudrit.

25 lll IY Tih 1) 1
Walbialse a tanehomo mie

millim. millim. millim. millim, millim.

From snout to vent. . 57 53 AT 44 39
JSAM carer Ma ary. oL(G) 15 14 13 12
IOs Uhh 4 5 6 6 2 21 2ili 19 16
labiogl Wien) 6 vo 6 6 5 AO 29 27 25 22

6&7. 6 &her. Egypt (Wilkinson). Typical P. lobatus.

Form slender, nostril in a very distinct swelling. Scaling finer than in P. oudri,
head less depressed, eye larger, auricular meatus more elongate. Wrist reaching
halfway between the eye and the nostril (6) or the tip of the snout (7); heel to the

12 12

axil (7) or not quite so far (6). Labials 35, 35.

millim. millim.

Exomygsnoutitonvents sane a ae nn 47
Ker VolOR Me Ne oa eens Sikes anit 2.8 1) an we duce 5 AL 10
Hone dimib yi, =) aueee tesa) Rev ee ney cole thes heen 25
a3 Guavoeel bar omree een ELE Atta ceruLiee rdnh Miia caer, <2 NT 33

8. 3. Egypt (Burton).
Similar to the preceding, but dorsal tubercles larger, quite as large as in P. oudrii.

Wrist to the nostril, heel to axil. Labials =

millim.
SrOMESMOUt CORVEI tl enemies seater pe ite aie ner aT
1B Ree yc eR artam aan UREN CME MAN mn Murals Ania Same teed gl: Alb
More lumilypspecase le ee Leen eee ten er ee ene ee ane.
Hind limbs: 24), cayenne eee ue Gee cbc aaa ee

9. 9. Egypt (Christy).
Like 6 & 7, but the shape of the head approaching that of P. oudri, and the nasal
swelling less marked. Wrist halfway between the eye and the nostril; heel to ¢ the

, 5 ; gies Uh
distance between groin and axil. Labials ;;.

millim.
HSKOMMESHOUtSEORVET tamleye ces et cy nee as) ran cee Oe
TTGad WA ye Vise cen beeen vepoosa Saw Lone LOT Ob aaa
Exo ree lin raat ay yaa osastaley ers arose cin ese Ws Recon Syed ic vc rts nae

Eline Sim bs ee ete leat a ck ncaels tRRRUPEROORO

AND BATRACHIANS OF BARBARY. 15

10. 9. Mt. Edfou, Eeypt (Anderson).
Like the preceding. Labials =.

millim.
Hromasnout,tonventuueress: | ears fon enna OS
Ea letitheshet cote Ce eet Aep Thy ice neh eee eae bese Niealenaian LG
ones hin wk pare tab ela) eee ee elec tek et ro)
Elincesliimby ecw an ae atin eeu et ee ao:

eo Ge Sitaoimar:
Like 6 & 7. Wrist not quite to the nostril; heel nearly to the axil. Labials =

millim.
SrOMeSMOUCELOLVCHthian yee eatin ewe ete ui
Head . 20
Horculina bee erect ts, cc enter cake ie te oie A ceger i
ond thunders incon toy as: Sea, Somat cs om ueaee

12. @. Muscat, Arabia (Murray).
Like the preceding. Wrist to halfway between the eye and the nostril; heel to the

: 19
axil. Labials 55.

millim.
IBC GNOME WO WOME 6 o 6 0 6 o.8 © 0 o 2 wl
EGA ie eae WOR eA eS ce ahh Ao Pee icey ooh wa eae
Graph GE sens eae eon) chase, pete is eae ate apenas FLO
Eline larity Peete, AOR ad OM en She werk 1. TSS) Va eee ag Ot]

13. 9. Dead Sea (Tristram).
Shape of the head intermediate between P. lobatus, typ., and P. owdrii; eye very
large ; ear asin P. oudrii. Scaling as in P. lobatus, typ. Wrist to halfway between

the eye and the end of the snout ; heel to axilla. Labials

millim.
From snout to vent 64
Head 19
ores bree Bs) cathe. en os 30
Hind limb 42

14. g. Mt. Carmel (Zristram).
Agrees in every respect, except size, with P. ovdrii; the
little more elongate. Labials 2,

From snout to vent
Heady Nas ie sys
Fore limb.

Hind limb

ear-opening is, however, a

millim.
Age

(
24
37

49

~

b

114 MR. G. A. BOULENGER ON THE REPTILES

15. 9. Jerusalem (Tristram).
Like the preceding, but tubercles smaller. Labials

millim.
Jao NOU HO WEN 6 o 46 o 36 0 o ooo a o OA
Heady i3 weed) cat ee Ok ae rat ee ae
Foresimbey Sine voce also te HR ea eee eter aso
Hind amiby, 2e ein ee ea ee ete areas
16. 2. Jerusalem (Anderson).
Like 14, but the heel reaches the axil. Labials =
millim.
[Dreohinl, SOULE HO) EINE. ol ge vo 6 Gets 09 Ge Se to 6 eT
Lead Steck sseuaeen oan ign eee onl eget ae mel a are
Horeclimbys ye een renee ye meer oe i nts eh eee SI)
md alin bY ae eee ea hots esa mee amie co) L

17. @. Galilee (Anderson). +
Agrees with P. oudrii, but ventral scales a little smaller, and limbs shorter still.
Wrist to the anterior border of the eye, and knee hardly to ? the distance between

; 3 eee
groin and axil. Labials 7.

millim.
[AROMA NOME HO AEM 8 oo oe ee eo ow (OA
1G 271s Maen Maier ame Be Wn adi ih beet ayn! JL)
oreclimibte 23.2243 pa nanetej el uct ch NE an ee a
ERG iam ee Se eee etek ieee se ae See

18. ¢. Between Khan Tubb Tusef and Ain Mellaha, Palestine (Anderson).
Like with the preceding. Labials = :

millim.
JRO TA VOU HO HEME So ss 6 ok ad Je! ee OE
PSE Line one aan se rem ies Kun Due EN at Ban) as Seailionhy ey ACD)
Ore NIMs cee eet ee nee ea erg re net ae)
Hind limb 56

The typical form is figured by Savigny, Descr. de l’Egypte, Suppl. pl. i. fig. 2. We
have figured the Algerian form from one of the type specimens of P. oudri7, on
Plate XIII. fig. 2.

7. Hemipacrytus, Gray, 1825.
Digits free, or more or less webbed, dilated, inferiorly with two rows of lamelle; all
the digits with slender distal clawed joints angularly bent and rising from within the

extremity of the dilated portion. Pupil vertical. Males with preeanal or femoral
pores.

AND BATRACHIANS OF BARBARY. 115

1. Hemipactyius turcicus, Linneens, 1766.
Hemidactylus verruculatus, Gervais, Guichenot.—H. cyanodactylus, Strauch.

Upper surface of body granular, with 14 to 16 longitudinal series of trihedral
tubercles; tail with transverse series of large keeled tubercles above. Males with 4 to
10 preanal pores. Light brown or greyish above, spotted with darker; many of the
tubercles white ; white beneath.

From snout to vent 53 millim.

Found all round the Mediterranean and the coast of the Red Sea, and extending
eastwards to Persia and Sind, but not yet recorded from Morocco, It is not very
common in Algeria. M. Lataste collected specimens near Oran, Algiers, Beni Mansour,
and Bona, and Dr. Kobelt at Biskra. In Tunisia M. Lataste found it at Djebel Rezaz
and Zarzis, M. Valéry-Mayet at Sfax, Kerkenna, and Thala, and Marquis Doria on
Galita.

The best figure of this species is that given in Bory de St. Vincent’s ‘ Expédition de
Morée,’ Rept. pl. xi. fig. 2.

8. TaRENTOLA, Gray, 1825.

Digits strongly dilated, free, with undivided lamellz below, and a flat nail-like scute
on their upper surface near the tip; third and fourth clawed, others clawless. Pupil
vertical. No femoral or preanal pores.

Two species in Barbary.

1. TaRENTOLA MAURITANICA, Linneus, 1766. (Plate XIII. fig. 3.)
Platydactylus fascicularis, Rozet, Gervais. —P. muralis, Guichenot.—P. facetanus, Strauch.

Head large, with swollen temples, covered above with large convex, usually obtusely
keeled scales, 13 to 15 in a transverse line between the eyes; temple with enlarged
tubercles ; anterior border of ear-opening not denticulated ; sides of neck with rosettes
of large conical tubercles surrounded with smaller ones. Dorsal region granular, with
transverse series of large, strongly keeled, subtrihedral, very prominent tubercles, sur-
rounded by smaller ones; sides with rosettes of conical tubercles. Gular scales much
smaller than ventrals ; mental shield large, separating the chin-shields, the inner of
which are as much developed as the labials. A more or less distinct fold along each
side of the belly. Digits strongly dilated at the end. Anterior part of tail with
posteriorly directed spine-like tubercles. Grey or grey-brown above, more or less
distinctly spotted or marbled with blackish and whitish, or with undulous dark trans-
verse bands on the back ; a more or less distinct dark streak on each side of the head,
passing through the eye; tail above with dark cross-bands ; lower parts white.

From snout to vent 70 to 80 millim.

Saharian specimens (var. deserti, Lataste, in litt.) are distinguished by a larger size,

116 MR. G. A. BOULENGER ON THE REPTILES

measuring up to 103 millim. from snout to veut, a somewhat longer and more pointed
head, finer granulations between the tubercles and on the throat, and the very pale,
yellowish-white coloration, without or with very indistinct pale brown spots.

This Gecko, common all round the Mediterranean, is found all over Barbary, north
and south of the Atlas; it frequents houses, old walls, and cliffs.

The typical form is figured in Bonaparte’s ‘Fauna Italica, and the var. deserti on
Plate XIII. fig. 3 of this Memoir.

2, "TARENTOLA NEGLECTA, Strauch, 1887.
TL. angusticeps, Strauch, 1887.

This species, which in its physiognomy reminds somewhat of Hemidactylus turcicus,
was recognized as distinct by Lataste, whose description was, however, anticipated by
that of Strauch.

The head is smaller than in 7. mauritanica, the temples usually less swollen, and the
scales on its upper surface larger, smooth or feebly keeled ; 10 to 12 scales in a trans-
verse line between the eyes ; tubercles on the temple small and feebly prominent. All
the tubercles on the back, the sides of the neck and body, and on the limbs much less
prominent, never conical; and the dorsals are entirely isolated, not surrounded by
smaller ones; the granules between the tubercles larger. Lateral ventral fold absent
or very indistinct. Digits less dilated, with subparallel lateral borders. Caudal
tubercles less developed. Pale brownish or yellowish white above, without or with
small brown spots or interrupted longitudinal lines; head usually with four to six
brown longitudinal lines, most distinct on the snout; tail with more or less distinct
darker cross-bands ; lower parts white.

This isa small species, the largest of many examples examined by M. Lataste measuring
53 millim. from snout to vent; the usual length is 45 millim. from snout to vent,
tail 50.

LT. neglecta was found in abundance by Lataste in the sandy Sahara in Algeria
between El Mala and Arifji, and at Wargla, on old palm-trees, never about houses
or walls like its larger congener. I have also received a speCimen from Kef el Dhor,
between Biskra and Tuggurt. It would appear, however, that the species occurs also
on the Plateaux, for Dr. Strauch’s specimens were purchased as from Batna.

Figured by Strauch, ‘“‘ Bemerk. ib. d. Geckoniden-Sammlung Zool. Mus. St. Petersb.”
(Mém. Ac. St. Pétersb. xxxv. 1887, no. 2), pl. —. figs. 1-4.

Fam. 2, AGAMID.
Two genera in Barbary :—
Tail round or feebly compressed; no femoral pores . . . . .. =. =... I. AGAMA.
2.

‘Tail short, depressed, covered with whorls of large spinose scales; femoral pores . Uromastix.

AND BATRACHIANS OF BARBARY. 117

1. Agama, Daudin, 1802.
Tail round or feebly compressed. No femoral pores; males with callose preanal

scales.
Of this African and South-west Asian genus three species occur in Barbary :—

A. Third toe much shorter than fourth, fifth not extending as far as first ; occipital
scale not enlarged ; ear-opening smaller than eye-opening.
Dorsal scales unequal; head not or but slightly longer than broad; tail cylin-
dricall;?male-wathout gularspouch) 5) 9 2 va0 2 Alee) cee ee ya le enenmis:
Dorsal scales equal; head longer than broad; tail compressed; male with a
larsegoul armpOUC lh ieren fre timce Wolo ten ecullls steugs yeincWipetanscimene warps 2. tournevilli.
B. Third and fourth toes equal, fifth extending beyond first ; occipital considerably
larger than the surrounding scales ; ear-opening larger than the eye-opening. 3. bibronii.

1. AGAMA INERMIS, Reuss, 1834.
A. agilis, Kichwald, Strauch.—A. ruderata, Strauch.—A. mutabilis, Lataste.

Head very short, not or but slightly longer than broad. Nostril directed upwards
and backwards, pierced on the canthus rostralis in the posterior part of a flat nasal.
Upper head-scales convex, smooth or very slightly keeled; occipital not enlarged ;
usually a few scattered small spinose scales on the back of the head; a fringe of small
spinose scales on the upper edge of the ear, which is smaller than the eye-opening.
Male without, or with only an indication of, a gular pouch. Body depressed, covered
above with unequal, rhomboidal, imbricate, keeled, more or less mucronate scales ;
ventral scales smooth or feebly keeled. Limbs moderate; tibia as long as the skull, or
a little shorter ; third finger shorter than fourth, fifth not extending as far as second ;
third toe much shorter than fourth, fifth not extending as far as first. Tail about twice
as long as the distance from gular fold to vent, rounded, covered with equal keeled
scales. Male with a double row of anal pores. Grey-brown or sandy coloured above,
with or without more or less distinct quadrangular dark grey, brown, or reddish spots
arranged symmetrically on the back ; some of the larger scales may be lighter; lower
parts white; the breeding male’s throat with blue longitudinal lines, or blue with
white spots.

From snout to vent 90 millim., tail 130.

This species varies considerably in the scaling of the back, a large series of specimens
showing almost every transition between a nearly equal lepidosis and one in which the
enlarged scales, scattered far apart, are as much as three or four times as large as the
others. Such extremes are, however, not of frequent occurrence.

Numerous specimens were obtained by M. Lataste in the Algerian Sahara at Biskra,
Hadjira, Wargla, in the Mzab at Tibremt, Laghouat, Bou-Saada, and on the High
Plateaux at Msila; by Dr. André in the district of the Tunisian Chotts ; in Tunisia by

VOL. XIII.—Part 111. No. 4.—Octover, 1891. T

118 MR. G. A. BOULENGER ON THE REPTILES

M. Lataste at Sidi Guenao, on the hills between Limagues and Kebili, at Tamesmida,
Bir-el-Ahmar, Zarzis, Sidi Haich, El Hammam des Beni Zid, Nebech el Dib, Wed
Zitouna, and Gafsa.

The range of A. inermis extends eastwards to Egypt.

A good figure accompanies Reuss’s description, Mus. Senckenb. i. 1854.

2. AGAMA TOURNEVILLII, Lataste, 1880. (Plate XIII. fig. 4.)

Habit more slender than in the preceding. Head one fourth longer than broad,
snout sloping gradually. Nostril directed upwards and backwards, pierced on the
canthus rostralis in the posterior part of a small flat nasal. Upper head-scales smooth,
convex; occipital not enlarged ; no spinose scales; a very slight fringe of pointed scales
on the upper border of the ear, which is smaller than the ear-opening. Male with a
large gular pouch. Body not or but feebly depressed, covered above with equal,
rhomboidal, strongly keeled, not mucronate scales; lateral and ventral scales a little
smaller, strongly keeled. Tibia a little shorter than the skull; third finger slightly
shorter than fourth, fifth not extending quite as far as first; third toe much shorter
than fourth, fifth not extending quite as far as first. Tail twice and a half as long as
the distance from gular fold to vent, distinctly compressed, covered with equal keeled
scales. Male with a row of anal pores. Sandy coloured above; a transverse brown
band between the eyes, two longitudinal ones from the occiput along the nape, and two
others on each side of the head; back with regular longitudinal series of quadrangular
transverse brown spots separated by rather indistinct longitudinal light lines ; tail with
dark annuli; lower parts white; gular pouch grey.

From snout to vent 94 millim., tail 162.

Only two specimens are known, a female collected by M. Lataste at Wargla, and a
male, labelled ‘‘ Sahara,” which has been for many years in the British Museum. The
latter differs from the former in the less depressed body, the slightly larger dorsal
scales, the rather larger ear-opening, and the presence of a gular sac.

3. AGAMA BIBRoNI, A. Duméril, 1851. (Plate XIV. fig. 1.)
A. colonorum, Gervais, Guichenet, Strauch.

Head a little longer than broad. Nostril tubular, directed outwards and backwards,
pierced just below the canthus rostralis. Upper head-scales smooth or indistinctly
keeled; occipital scale considerably larger than the surrounding ones; eleven to
fourteen upper labials; sides. of head and neck with groups of spines; ear-opening
larger than the eye-opening. Throat strongly plicate; no gular pouch. Body depressed,
covered above with large, rhomboidal, mucronate, imbricate scales, with strong keels
converging towards the vertebral line; a slight nuchal crest; ventral scales small,
smooth. Tibia as long as the skull; third and fourth fingers equal; third and fourth
toes nearly equal, fifth extending beyond first. ‘Tail about twice as long as the distance

AND BATRACHIANS OF BARBARY. ALG)

from gular fold to vent, round in the female, very feebly compressed in the male
covered with whorls of strongly keeled, spinose scales. Male with a row of anal pores.
Bronzy olive or leather-brown above, the vertebral line sometimes yellowish, sometimes
with darker and lighter spots; tail with dark annuli; whitish or greyish beneath, the
male’s throat bluish. Young with three dorsal series of whitish dark-edged ocelli.

From snout to vent 110 millim., tail 150.

This species was founded upon specimens from Mogador. Boettger records it from
Tangier (whence the British Museum has also received specimens through M. Henry
Vaucher) and between Mogador and Morocco, and states that itis very common on the
plateau between Ain-Umest and Sidi-Moktar. In Algeria it is on record from Tlemsen,
Bou-Saada, and the Mzab (Strauch); the Chotts of the Province of Oran (Gervais) ;
and Saida, eastern frontier of Algeria (Gwichenot). According to M. Lataste it is only
found in the stony parts of the Sahara in the Provinces of Algiers and Constantine ; he
obtained specimens in the Mzab, at Laghouat, at Bou-Saada, and at Cachrou, near Oran.
It has not yet been found in Tunisia.

2. Uromastix, Merrem, 1820.

Incisors large, uniting in the adult into one or two cutting-teeth, separated from the
lateral teeth by an interspace. Tail short, depressed, covered with whorls of large
spinose scales. Praanal and femoral pores.

These curious essentially phytophagous Lizards inhabit the arid tracts of North Africa
and Southern Asia. The Algero-Tunisian species is :—

1. URromastix ACANTHINURUS, Bell, 1825.
U. spinipes, Giinther.—U. temporalis, Valenciennes.

Anterior border of ear denticulated. Scales small, no enlarged ones on the back or
flanks. 9 to 11 femoral and 4 or 5 preanal pores on each side. ‘Tail above with large
spinose scales not separated by smaller ones; below with smaller scales, two or three
whorls corresponding to one of the upper surface. Yellowish, greenish, or greyish
above, dotted or vermiculated with blackish or brown.

Total length 250 millim., tail 150.

Very common in the stony Sahara, where it makes burrows in the firm soil to the
depth of two or three feet. It is found in abundance at Biskra, Bou-Saada, Geryville,
Mascara, Laghouat, and the Mzab; also in the Hodna district, in the south of the
High Plateaux (Msila and Magra); in Tunisia M. Lataste observed it at Tamesred,
Taoudjout (Matmata), Hadedj, Wed-el-Kreil, M. Valéry-Mayet at Gafsa and Kriz, and
Dr. André between Cabes and the Chott Fejej.

U. acanthinurus is not recorded west of Algeria or east of Tunisia. It has sometimes
been confounded with the more eastern U. spinipes, Daud., distinguished by much

q 2

120 MR. G. A. BOULENGER ON THE REPTILES

smaller scales and the presence of scattered small tubercles on the flanks, and which
does not occur in Barbary.

The only figure of this Lizard is that accompanying the original description by Bell,
Zool. Journ. 1. 1825, pl. xvii.

Fam. 3. ANGUID.

A single representative in Morocco, Ophisaurus koellikert.

Two other Lizards of this family have been ascribed to this fauna, but I have no
hesitation in erasing them from the list, viz., Anguis fragilis, L., and Ophisaurus
(Pseudopus) apus, Pall. ‘They were mentioned by Gervais as having been obtained,
together with Ophiomorus punctatissimus, Bibr., at Algiers by M. Marloy. The fact
that these Lizards, which occur together in the East (Greece, Asia Minor), have never
since turned up in Barbary, renders it absolutely incredible that they should have been
obtained at or near so well-explored a place as Algiers; I therefore believe the locality
to be altogether erroneous. It is true that the supposed Algerian specimen of
Ophiomorus is now entered in the registers of the Paris Museum, where, thanks to
Prof. Vaillant’s courtesy, I have been able to examine it, as from Bona, through
Dr. Guyon. But this is doubtless again an error, the specimen being surely the same
as was mentioned by Gervais, and presented by him to the Museum together with the
Anguis fragilis, as may be seen by referring to Aug. Duméril’s ‘ Catalogue Méthodique,’
pp. 189, 190, where the Anguis is referred to as from “ Bone: M. Marloy.”

1. Opuisaurus, Daudin, 1803.

A lateral fold. Scales squarish rhomboidal, forming straight longitudinal and
transverse series. Limbs absent externally, or reduced to rudiments of the hind pair.

1. OPHISAURUS KOELLIKERI, Gtinther, 1873.
Pseudopus apus, forma ornata, Boettger, 1881.

Azygous prefrontal large, quite as broad as the frontal, with which it forms a suture ;
interparietal much broader than the parietals or the occipital; two shields on a line
from the nasal to the azygous prefrontal; five supraoculars. Ear concealed. Dorsal
shields in 14 longitudinal and 120 transverse series, the median obtusely keeled, the
laterals smooth ; ventrals in 10 longitudinal series, smooth. Rudiments of hind limbs.
Upper and lower caudal scales keeled. Brownish above, with a darker lateral band
and small dark brown spots on the middle line of the anterior part of the back, or with
blackish transverse bands spotted with bluish ; pay yellowish.

From snout to vent 15 centim.

The type specimen is probably from Mogador; Dr. Boettger examined two specimens
from Casablanca.

Figured in Catal. of Lizards, vol. ii. pl. xv. fig. 2.

AND BATRACHIANS OF BARBARY. 121

Fam. 4. VARANIDZ.
A single genus, confined to the Old World and Australia.

1. Varanus, Merrem, 1820.

Neck much elongate. Dorsal scales roundish, juxtaposed, surrounded by rings of
minute granules ; ventral scales squarish, arranged in cross rows. No femoral pores.

1. Varanus @Riseus, Daudin, 1802.
V. arenarius, Gervais, Guichenot.— V. scincus, Strauch.

Teeth acute, compressed. Nostril an oblique slit near the eye. Scales of head
small, granular, subequal. ‘Tail round or slightly compressed. Greyish yellow, some-
times with more or less distinct brown cross-bands on the back and tail, and brown
streaks along the sides of the neck.

From snout to vent 56 centim., tail 71.

Inhabits the Sahara, extending eastwards to the Caspian Sea, Afghanistan, and
North-western India. Pretty common in the Algerian Sahara, and also found in the
Hodna district, in the south of the High Plateaux. Specimens were brought to
M. Lataste at Wargla, Laghouat, Bou-Saada, Biskra, Msila, and in Tunisia M. Lataste
obtained it at Tozeur, and M. Valéry-Mayet at Majura.

Figured in the ‘ Description de Egypte,’ Rept. pl. iii. fig. 2.

Fam. 5. AMPHISBENIDZ.
Worm-like Lizards, represented in this fauna by two genera :-—

Preeanal pores; nostril pierced in the first labial. . . . . . . . . . . L. Buanus.
No preanal pores ; nostrilin a separate nasal. . . . . . . . . . . . 2. TROGONOPHIS.

1. Buanus, Wagler, 1830.

Teeth anchylosed to the sides of the jaws. Nostril pierced in the first labial; a
large frontal, forming a suture with the rostral. A well-marked lateral line; a curved
postgular fold. No limbs. ‘Tail pointed. Preeanal pores. —

1. BLanus cINEREUS, Vandelli, 1797.

Snout rounded, not at all projecting beyond the lower jaw. Four upper and four
lower labials ; 110-125 annuli on the body and 20—22 on the tail; 4—6 preanal pores.
Brownish flesh-colour, the segments of a more or less dark brown.

Total length 250 millim.

This species occurs in Spain and Portugal and Morocco, where it has been obtained
at Tangier, Tetuan, and between Mogador and Morocco. From Algeria it is recorded

122 MR. G. A, BOULENGER ON THE REPTILES

by Lallemant from the western part of the Province of Oran, and from Tebesa, in the
Province of Constantine. Dr. Strauch purchased specimens stated to have been obtained
at Batna. M. Lataste did not come across it in Algeria, and thinks some of the reports
of its occurrence may be due to confusion with the uniform fuliginous variety of
Trogonophis wiegmanni.

Figured by Gervais, Mag. Zool. 1836, pl. x.

2. Trogonoruis, Kaup, 1830.

Teeth anchylosed to the parapet of the jaws. Nostril pierced in a large nasal; two
pairs of upper head-shields. A vertebral line and a stronger lateral line. No limbs.
Tail conical. No preanal pores.

1. TROGONOPHIS WIEGMANNI, Kaup, 1330.
Amphisbena elegans, Gervais.

Snout slightly projecting. Five upper labials, second and fifth smallest ; four lower
labials. 136-151 annuli on the body and 12-14 on the tail. Yellowish white,
chequered with black. A variety, of which a specimen was obtained at La Chiffa by
Dr. Bavay, and another at Biskra by M. Hénon, is of a uniform fuliginous grey, a little
lighter below.

Total length 259 millim.

This genus is peculiar to Barbary. From Morocco Trogonophis is known from
Tangier, Tetuan, the Zaffarine Islands, Larache, Casablanca, Coreina, Mogador, and
between Mogador and Morocco. It is found in the three provinces of Algeria, and
occurs as far south as Biskra. The only Tunisian locality is Tamesmida, between
Ferriana and Tebesa, where specimens were collected by M. Lataste.

Figured by Gervais, Mag. Zool. 1836, pl. xi.

Fam. 6. LACERTID.

The range of this family is restricted to the Old World; the five following genera
are represented in Barbary :—

A. Movable eyelids.
a. Nostril above the first labial, from which it is separated, if at all, by
a narrow rim.
a. Digits not denticulated laterally.

Dorsal scales small 3 1. Lacerta.
Dorsal scales large and imbricate . 2. PsAMMopROMUS.
8. Digits denticulated laterally 3. ACANTHODACTYLUS.
6. Nostril well separated from the labials . 4, HEREMIAs.
B. No movable eyelids . 5. Opniops.

AND BATRACHIANS OF BARBARY. 123

1. Lacerta, Linneus, 1766.

Nostril bordered by two or three nasals and the first labial. Eyelids movable.
Collar well marked. Dorsal scales much smaller than caudals, not or but feebly
imbricate ; ventral shields tetragonal, feebly imbricate. Digits with smooth or tuber-
cular lamelle, not denticulated laterally. Femoral pores.

Three species in Barbary :—

Rostral entering the nostril; collar-edge serrated. . . . . . . ..- . « LI. ocellata.
Wisely stile joel 6 6 5 6 6 066 6 6 oo oo a ee RES
Lower eyelid with atramsparent disk. . . . ....... =.=. =. . & perspicillata.

1. Lacerta ocELLata, Daudin, 1802. (Plate XV.)

Lower eyelid scaly. Rostral shield entering the nostril; two superposed postnasals ;
occipital shield large ; temple covered with irregular, rather large scales. Collar with
serrated edge. Dorsal scales small, roundish-rhomboidal, feebly keeled, 60 or more
across the middle of the body. Ventral plates broader than long, in 6 to 10 longitudinal
series. Green above, with or without blue or white ocelli, or reticulated black and
yellow. Size large (grows to 3 feet).

The typical form of this, the largest species of the genus Lacerta, inhabits the south
of France, Liguria, and the Pyrenean Peninsula, but is not found in Barbary, where it
is represented by the two following races or subspecies.

Var. PATER, Lataste, 1880.
L. viridissima, Rozet.—L. ocellata, Schlegel, Strauch.—L. viridis, Gervais.

The large Algerian green Lizard was long confounded with both the typical L. ocellata
and LZ. viridis, until M. Lataste showed that it formed a distinct race or subspecies,
which, though much nearer the former, presents some points of affinity to the latter.
The form has since even been raised to the rank of a species (Bedriaga), but I hold
that M. Lataste was well advised in treating it as subordinate to L. ocellata. Exami-
nation of a good number of specimens from the Spanish Peninsula has even convinced
me that the distinctive characters between the two forms are by far not so well marked
as M. Lataste thought. But on the whole, taking the ensemble of the characters of the
Algero-Tunisian form, a separation from the European Z. ocellata is justified, and the
name proposed by Lataste is well chosen, as L. pater may be looked upon as most
nearly allied to, if not the actual survivor of, the ancestral stock from which the allied
L. ocellata and L, viridis are descended. The affinity to L. viridis is shown in the
usually more distinctly keeled dorsal scales; the smaller occipital shield, which is as
broad as or a little narrower than the frontal; the ventrals in 8 longitudinal rows
(8-10 in L. ocellata, 6-8 in L. viridis); and, in some specimens, the absence of ocellar
spots. Not one of these characters, taken by itself, is, however, absolutely distinctive..

124 MR. G. A. BOULENGER ON THE REPTILES

The following notes are taken from the specimens in the British Museum :—

The occipital is constantly broader than the interparietal, and usually narrower than
the frontal; however, in a male specimen from Tunis it is quite as broad as the
frontal, not larger than in a specimen of the typical form from Ferrol (Spain)'. The
dorsal scales are small, oval-subrhomboidal, feebly keeled, and their number across the
middle of the body varies from 70 to 80. The ventral shields form 8 longitudinal
series, of which the two median and the outermost are the narrowest. Femoral pores
vary from 14 to 16 on each side, 14 being the usual number. ‘Two or three semicircles
of small shields on the anal region.

The coloration varies greatly, green being the ground-colour. ‘The young are usually
ornamented all over with large bluish-white black-edged ocelli, like the young of the
typical L. ocellata; others are uniform green. The adult may be uniform or speckled
with black, like typical LZ. viridis ; or with black rings or small black and white ocelli,
thus very similar to the Iberian L. viridis, var. schreiberi; or with large blue ocelli on
the sides like the typical LZ. ocellata. Lower parts uniform greenish yellow.

The largest specimen measures 165 millim. from snout to vent.

This fine Lizard is found all over Algeria, as far south as the northern border of the
Sahara. M. Lataste found it very common at Algiers, Aumale, Setif, Bona, Batna,
Lambesa, El Guerrah, Bougia; and he received it from the Plateau of Sersou.
Dr. Strauch records it from Oran and Constantine; J. von Fischer from Boghar,
Blidah, La Chiffa, Tiaret, El Kantarah, El Rouached, St. Arnaud; and Dr. Kobelt
collected specimens at Tlemsen and Biskra. In Tunisia, it is known from the city of
Tunis, and M. Lataste observed it in the northern parts, between Ferriana and ‘Tebesa.
Marquis Doria found it on Galita Island.

Figured on Plate XV. figs. a—e.

Var. TANGITANA, Boulenger, 1887.

This interesting form was recently discovered at ‘Tangier by M. H. Vaucher, from
whom I received 11 specimens. There was no previous record of either L. ocellata or
LL. viridis in Morocco. ‘The var. tangitana comes very near the Algerian var. pater, but
diverges from it as well as from L. ocellata typica in the still smaller dorsal scales, of
which there are 77 to 100 across the middle of the body, and the greater number of
femoral pores, viz. 17 to 21. In the usually smaller size of the occipital and the
number (6 or 8) of longitudinal rows of ventral shields, it approaches nearer still the
Spanish-Portuguese form of ZL. viridis (var. schreiberi, Bedriaga), from which some
specimens are with respect to these two characters undistinguishable.

In some specimens the occipital is not or but slightly broader than the interparietal ;
in others its greatest width equals that of the frontal. 24 to 28 scales on a line
between the chin and the collar; latter with 11 to 13 shields. There are usually eight

* The head of this Spanish specimen is figured on Plate XY. fig. g.

AND BATRACHIANS OF BARBARY. 128

bo
Or

longitudinal rows of ventrals, but sometimes only six. Two or three semicircles of
small shields on the anal region. Green above, with whitish or blue black-edged
ocelli, which may be confined to the sides; lower parts uniform greenish yellow.

From snout to vent 140 millim., tail 300.

I have given a figure of this variety in the third volume of the British Museum
‘Catalogue of Lizards,’ pl. iii. fig. 1. The upper surface of the head is represented on
Plate XV. fig. f of this Memoir.

2. LACERTA MURALIS, Laurenti, 1768.
L, agilis, Gervais.

Lower eyelid scaly. Rostral shield not entering the nostril; usually a single post-
nasal; occipital shield small ; temple granular, usually with an enlarged circular shield
in the middle. Collar with even or very slightly serrated edge. Dorsal scales small,
granular, smooth or feebly keeled, 40 or more across the middle of the body. Ventral
plates broader than long, in 6 (rarely 8) longitudinal series.

A very variable species, inhabiting Western and Southern Europe, South-western
Asia, and Barbary. It is recorded from Cyrenaica by Reichenow.

In the specimens from Tangier, the scales are very small, obtusely keeled, 61 to 73
across the middle of the body, three or four transverse series corresponding to one
ventral shield. Four labials anterior to the subocular (one specimen has five on one
side, another only three). An enlarged shield, sometimes broken up into two or three,
is present in the middle of the granulate temple in five specimens, altogether absent in
a sixth. Upper caudal scales strongly keeled. Femoral pores 13 to 19, usually 17.
Olive-grey above, with small black spots or reticulations ; a more or less defined dark
lateral band, bordered above by a whitish streak or series of white spots; hinder side
of thighs with round white spots; two series of white, black-edged spots along each
side of the tail; belly uniform white in both sexes, or with a few scattered black dots ;
throat with black dots.

The largest male measures, from snout to vent 52 millim., tail 95 ; the largest female,
from snout to vent 50, tail 79.

In specimens from the plains of the Spanish Peninsula (Valencia, Lisbon), the dorsal
scales are smooth and the caudals very feebly keeled ; 55 to 60 scales across the middle
of the body. Specimens from the Serra de Gerez have 54 to 63 scales across the body,
and these are obtusely keeled, and the caudals have well-marked keels. In Persian
specimens I count only 45 to 50 scales. In European specimens of the typical form
generally, the number of scales across the middle of the body varies from 48 to 65.

In Morocco, this Lizard is as yet only known from Tangier. In Algeria it is widely
distributed north of the Sahara, though far less common than on the opposite shores of
the Mediterranean. M. Lataste collected specimens at Aumale, Rorfa des Beni Slimam,
Constantine, Setif, Chabet-el-Acra, Beni Mansour, Tebesa, and received some from Daya

VOL. X111.—Part 1. No. 5.—October, 1891. U

126 MR. G. A. BOULENGER ON THE REPTILES

and the Plateau of Sersou. Strauch obtained it from’Tlemsen. It is thus recorded from
the three provinces. In Tunis, M. Lataste met with Z. muralis at Guelaat-es-Sinam, and
on the route from Ferriana to ‘Tebesa.

The larger green form, var. tliguerta, Gm. (neapolitana, Bedriaga), so common in
the south of Italy and Malta, is recorded from Tunis by Camerano (Atti Acc. Torin.
xili. 1877, p. 87).

3. LACERTA PERSPICILLATA, Dum. & Bubr. 1859,

Lower eyelid with a transparent disk. Rostral shield not entering the nostril; two
superposed postnasals; occipital shield small; temple uniformly granular. Collar with
even edge. Dorsal scales small, granular, 50 or more across the middle of the body.
Ventral plates square, subequal, in 10 longitudinal series. Green or bronzy above,
uniform or with light ocelli; uniform greenish white beneath.

This Lizard is well distinguished from all other members of the genus Lacerta by
the presence of a transparent disk in the lower eyelid, which permits the animal to see
when the eyes are closed. As far as we know at present, its habitat is restricted to
the environs of Oran, where it is common on Mount Santa Cruz; it was also found
in the town of Oran by M. Gazagnaire; its habits are, like those of L. muralis,
essentially saxicole. The various statements of the occurrence of L. perspicillata
outside the Province of Oran appear to be erroneous’.

The following descriptive details are taken from the specimens in the British
Museum :—

A series of granules between the supraoculars and the supraciliaries; occipital
smaller than the interparietal, which is at least twice as long as broad ; five upper
labials anterior to the subocular. <A distinct fold across the throat, from ear to ear;
32 to 34 scales on a line between the chin and the collar; latter with 9 or 10 shields.
54 to 56 scales across the middle of the body; three transverse series of scales correspond
to one ventral shield. 29 transverse rows of ventrals in the male, 55 in the female. 18
or 19 femoral pores on each side. A single semicircle of small shields borders the anal.
Upper caudal scales smooth or very feebly keeled, truncate posteriorly. The hind limb
reaches the axil in the male, the elbow of the adpressed fore limb in the female.

From snout to vent 58 millim.

L. perspicillata is figured by Guichenot, Explor. Sc. Alg., Rept. pl. i. fig. 3.

2. Psammopromvs, Fitzinger, 1826.

Nostril between two nasals, in contact with the first labial or separated only by a
narrow rim. Eyelids movable, scaly. Collar absent or very feebly marked; a short
fold in front of the arm. Back covered with large, rhombic, strongly keeled and
imbricate scales. Ventrals imbricate. Digits with tubercular or keeled lamelle
inferiorly. Femoral pores.

* Cf. Lataste, C. R. Assoc. Frang. Ay. Se., Congrés d’Oran, 1888, i. p. 191.

AND BATRACHIANS OF BARBARY. 127

Three species in Barbary :—

Collar ahsiimsmisnmoles 6 6 S66 6 o 6b 6 6 oo oo oo od 0 0» dl Alemaai
No trace of a collar; ventral plates of the median and outer series narrower

WM TNS OWNS 6 6g 6 6 6 6 6 oe oo 8 ooo Oo 6 9 ee POKGROCIEOS UTS.
INo}trace ofjarcollar;) ventral platesisubequally j=) )|-0u-) ee dee ee oe Om algenus:

The fourth and type species of this genus, P. hispanicus, Fitz., inhabits the Spanish
Peninsula and the south of France. It has been recorded from Tangier in the ‘ List of
Animals living in the Gardens of the Zoological Society of London,’ no doubt through
confusion with P. microdactylus ; and from Oran by F. Miller (Verh. nat. Ges. Basel, vi.
1878, p. 625), through confusion with P. blanci.

1. PsaMMoDRoMUS BLANCI, Lataste, 1880. (Plate XIV. fig. 2.)

A more or less distinct gular fold, connecting the ears; collar distinguishable.
Ventrals broader than long, the median and outer series narrower than the two others.
Digits inferiorly with a double series of more or less strongly or obtusely keeled
tubercles. Olive or bronzy brown above, with two yellowish streaks along each side,
bordered with small black spots; the upper streak commences from the upper border
of the temple, the lower commences on the upper lip and passes through the ear;
sometimes a light vertebral band ; lower parts uniform yellowish.

The suture between the rostral and the first labial usually falls below the anterior
border of the nostril. 28 to 32 scales round the middle of the body, ventrals included.
10 to 12 femoral pores on each side.

From snout to vent 40 millim., tail 61.

This small Lizard is easily mistaken for the young of the much commoner P. algirus,
which occurs together with it in many localities. When examined, it is, however,
found to differ in the gular fold, the collar, and the arrangement and shape of the
ventral plates, which much resemble the same in true Lizards, whilst those of P. algirus
remind one to a certain extent of the Scinks.

Psammodromus blanci is as yet only known from Algeria, but is on record from
the three provinces; it does not occur south of the Atlas. M. Lataste found it not
uncommon in and near Algiers and near Aumale, very common at Lambesa, and between
Tafrant and Meriana, and also obtained it at Youkouss, near Tebesa, and at Oran; his
collection contains specimens from Batna, received from M. Henri Martin; and J. y.
Fischer records it from Blidah (Prov. Algiers) and Rouached (Prov. Constantine).

2. PSAMMODROMUS MICRODACTYLUS, Boettger, 1881.

No gular fold; no trace of a collar. Ventrals broader than long, the median and
outer series narrower than the two others. Digits beneath with a double series of
strongly but obtusely keeled tubercles. Upper caudal scales more strongly keeled than
the dorsals. Olive or pea-green above, with or without brown or black spots, which

may be mixed with white; a more or less distinct brown or reddish lateral band ; lower
v2

128 MR. G. A. BOULENGER ON THE REPTILES

surfaces white, outer row of ventrals lemon-yellow; throat of males bluish. Young
with small black and white ocelli on the sides; upper lip pure white; a pure white,
black-edged streak from the eye to above the tympanum, and a second from the angle
of the mouth through the lower half of the tympanum to the axil.

The suture between the rostral and the first labial falls below the centre of the
nostril. 28 to 30 scales round the middle of the body, ventrals included. 19 to 13
femoral pores on each side, the usual number being 12.

From snout to vent 119 millim., tail 75.

This elegant little Lizard has hitherto been found only in Morocco. Dr. Boettger
received it from Tangier and Tetuan. In addition to the types in the Senckenberg
Museum, I have examined 23 specimens, obtained by M. Vaucher at Tangier, and by
Lieut. Quedenfeldt at Mogador.

For a detailed description and figure, cf. Boettger, Abh. Senck. Ges. xiii. 1883,
Delle pl sa ene 2

3. PSAMMODROMUS ALGIRUS, Linneeus, 1766.
Algira barbarica, Gervais.

No gular fold; no trace of a collar. Ventrals subequal, little broader than long,
roundish hexagonal, strongly imbricate. Subdigital lamelle smooth, tubercular, or
feebly bicarinate. Upper caudal scales like the dorsals. Bronzy above, with one or
two golden, dark-edged lateral streaks; male with a pale blue ocellus above the
shoulder, sometimes followed by one or two more; lower surfaces whitish.

From snout to vent 76 millim., tail 190.

This species, which occurs also in the Spanish Peninsula and the south of France,
is common and generally distributed in Morocco, Algeria, and Tunisia north of the
Sahara. A variety (var. nollii) from the Sahara (Tuggurt and Southern Tunisia) has
recently been described by J. v. Fischer (Zool. Gart. 1887, p. 69); this variety, which is
distinguished by having two additional yellowish stripes along the back, is represented
in M. Lataste’s collection by a specimen from Founasse, in the south of the Province
of Oran, received from M. Maury in 1888. A melanotic form, blackish above, with
bluish dots, bluish grey beneath, obtained by Marquis Doria on Galitone Island, near
Galita, has been described by Bedriaga as var. doriw (Beitr. Kenntn. Lacert. 1886,

. 409).

A good figure of this pretty Lizard is given by Bonaparte in his ‘ Fauna Italica.’

3, ACANTHODACTYLUS, Wiegmann, 1834.
Nostril pierced between two nasals and the first labial. No occipital shield.
Eyelids movable. Collar more or less distinct. Dorsal scales juxtaposed or imbricate ;

ventrals tetragonal, feebly imbricate. Digits keeled inferiorly and more or less strongly
fringed laterally. Femoral pores.

AND BATRACHIANS OF BARBARY. 129

Four species in Barbary :—

Posterior dorsal scales very much larger than the anterior, strongly imbricate,

sharply keeled, passing gradually into the caudals . . L. boskianus.
Ventrals in 14 to 18 longitudinal series, not or but slightly roast en long ;

digital denticulations at least as long as the diameter of the corresponding

part of the toe 2. scutellatus.
Ventrals in 12 or 14 bavetarhaall series, ee wean Libader chen one 3. pardalis.
Ventrals in 8 or 10 longitudinal series . 4. vulgaris.

For fuller details upon the Lizards of this genus, cf. Lataste “ Les Acanthodactyles de
Barbarie et les autres espéces du genre,” Ann. Mus. Genova, (2) ii. 1885, pp. 476-516.
The only important point in which I differ from M. Lataste with regard to the four
species of Barbary is his identification of Lacerta savignyi, Audouin (1829), with the
species which, from an examination of the types of Lacerta pardalis, Lichtenstein
(1823), I call A. pardalis. I identify the figure of the Egyptian L. savignyi with the
Lizard from Somaliland described by Lataste as A. vaillanti.

These Lizards are only found in sandy localities; they range over the Sahara and
the bordering countries as far south as Senegambia and Somaliland, one species ex-
tending to the Spanish Peninsula and the south of France. Eastwards they extend
through Arabia, Syria, Persia, and Baluchistan to Sind and the Punjab.

1. ACANTHODACTYLUS BOSKIANUS, Daudin, 1802.

Snout obtuse. Four supraoculars, the fourth very seldom broken up into small :
scales; front edge of the ear usually distinctly denticulated. Dorsal scales strongly
keeled, very much larger on the hinder part of the back than between the shoulders
and on the flanks, rhomboidal, strongly imbricate. Ventral plates considerably broader
than long, in straight longitudinal and transverse series, 10 (rarely 12) across the
middle of the body. Digital denticulations strong, usually shorter than the correspond-
ing diameter of the toe, much more developed on the outer than on the inner edge of
the fourth toe. Young with whitish longitudinal lines separated by blackish inter-
spaces with series of round whitish spots ; these markings become more indistinct or
disappear with age, the adult being greyish, brownish, or buff, with or without small
blackish spots; lower parts white ; tail of young pink.

The Algerian specimens belong to the var. asper, Audouin, distinguished from the
typical form (hitherto found only in Egypt) by its larger dorsal scales. These scales
form 34 to 42 longitudinal series (exclusive of the ventrals) round the middle of the
body, and there are 10 to 14 longitudinal rows of large keeled scales between the hind
limbs. The subocular does not reach the lip, but is wedged in between the fourth and
fifth or fifth and sixth labials. 19 to 24 femoral pores on each side. A median series
of broad preanal plates, posterior largest.

The largest specimen examined by me measures 80 millim. from snout to vent; tail

not quite twice as long.

150 MR. G. A. BOULENGER ON THE REPTILES

A. boskianus occurs from Southern Algeria and Tunis through ‘Tripoli and Egypt to
Syria and Arabia; it is also found in Abyssinia. It is common in the Algerian and
Tunisian Sahara, and rare on the Algerian Plateaux. M. Lataste found it on the
plateau at Bordj Medjez, Msila, and Ngaous; in the Algerian Sahara at Bord) Tayer
Rason, Ghardaia, Berrian, Tilremt, Laghouat, Bou-Saada, and on the route from Bou-
Saada to Biskra, and received specimens obtained by M. Maury at Kreder, Prov. of
Oran, near the Sahara and the Moroccan frontier ; in Tunisia at Cabes, Zarzis, Djerba
Island, Limagues, and Ferriana.

Figured, ‘ Description de Egypte,’ Rept., Suppl. pl. i. fig. 9.

2, ACANTHODACTYLUS SCUTELLATUS, Audouin, 1829.

Snout acute. Four supraoculars, fourth sometimes broken up; front edge of the
ear usually strongly denticulated. Dorsal scales small, rhomboidal, keeled, slightly
enlarged towards the posterior part of the body. Ventral plates as long as broad, or a
little longer than broad, rarely a little broader than long, in irregular longitudinal and
angular transverse series; 14 to 18 plates in a transverse series in the middle of the body.
Digital denticulations strong, at least as long as the diameter of the corresponding part
of the toe, much more developed on the outer than on the inner edge of the fourth toe.
Greyish or pale buff above, dotted or reticulated with darker ; white beneath.

This is a very variable species. ‘The Saharian examples are smaller, more slender
and lighter-coloured than the typical form from Egypt, and the name var. exiguus
(=Scaptira inornata, Gray) has been bestowed upon them by Lataste.

The following notes are taken from Algerian and Tunisian examples :—

Dorsal scales very distinctly keeled, juxtaposed, usually passing very gradually
into the marginal ventral shields, the exact number of longitudinal rows of which is,
therefore, often difficult to ascertain; there are at least 14 well-defined ventrals in a
transverse series across the middle of the body, and the number may rise to 18. 61 to
74 scales (ventrals included) round the middle of the body. Praanal shields often all
small and subequal. Subocular never reaching the lip, from which it is usually sepa-
rated by one or two additional labials intercalated between the (normally) fourth and
fifth. Femoral pores 18 to 26.

From snout to vent 50 millim., tail 90.

The range of A. scutellatus extends from Senegambia through the Sahara to Egypt,
Somaliland, the Sinaitic Peninsula, and Syria. It is only found in the sandy parts of
the Sahara, just penetrating to some points of the Plateaux. M. Lataste obtained it in
Algeria, at Biskra, Mraier, Tuggurt, Bled Ahmar, Hadjira, N’gousa, Tilremt, Laghouat,
Ain-el-Hel, and Bou-Saada, and received from M. Maury specimens at Ain Sefra, in the
south of the Province of Oran; in Tunisia, at Wed-el-Kreil, Kebili, Tozeur, and Nefta.
It is not on record from Morocco, although it surely exists in the Saharian parts; we
have specimens obtained not far from the southern limit of Morocco, viz. from Cape
Jubi, near Cape Nun.

Figured, ‘ Description de Egypte, Rept., Suppl. pl. 1. figs. 7 & 11.

AND BATRACHIANS OF BARBARY. 151

53. ACANTHODACTYLUS PARDALIS, Lichtenstein, 1823.

A. savignyi, Gervais, Guichenot, Strauch, Lataste—Zootoca deserti, Giinther.—A. bedriage,
Lataste.

Snout obtuse, or, at any rate, less acute than in A. scutellatus. Three supraoculars,
the anterior frequently divided into two or three; front edge of ear more or less
distinctly denticulated. Dorsal scales very small, granular, or subrhomboidal, smooth
or more or less feebly keeled. Ventral plates broader than long, sometimes very
slightly, in 12 or 14 regular longitudinal series. Digital denticulations feebly or
moderately developed, much shorter than the diameter of the corresponding part of the
toe, slightly more developed on the outer than on the inner side. Young longitudinally
streaked black and white on the body, with round white spots on the limbs; coloration
of adult very variable, usually with longitudinal series of blackish and yellowish or
brick-red spots, sometimes with yellowish longitudinal bands; the desert specimens
pale greyish or reddish.

A. pardalis is a transitional form between A. scutellatus and <A. vulgaris. It is
represented in Algeria and Tunis by two ill-defined varieties, which have been
named by Lataste var. bedriage, the larger, stouter form, approaching A. vulgaris in
structure as well as coloration, and var. deserti (=Scaptira maculata, Gray, Zootoca
deserti, Giinther), the smaller form from the Sahara, which often closely approaches
A. scutellatus.

The subocular rarely reaches the lip; its lower border is usually wedged in between
the fourth and fifth or fifth and sixth labials. In Algerian and Tunisian specimens I
count 66 to $2 scales round the middle of the body, ventrals included. Femoral pores
15 to 265.

From snout to vent 70 millim., tail 100.

This species occurs from Algeria to Egypt, Somaliland, and Syria. It is the most
common and generally distributed species in Algeria and Tunisia; it is common in the
Sahara, the Plateaux, and also occurs in the southern parts of the Tell. M. Lataste
obtained it in numerous localities both in Algeria and Tunisia, and received it from
Kreder, on the Moroccan frontier, from M. Maury in 1888.

Figured, Proc. Zool. Soc. 1881, pl. Ixiii. fig. 1.

4, ACANTHODACTYLUS VULGARIS, Dum. & Bibr. 1839.
A. lineomaculatus, D. & B.

Snout obtuse. Two supraoculars, the first and fourth being broken up into small
scales or granules ; front edge of the ear not or but feebly denticulated. Dorsal scales
smooth, or more or less distinctly keeled, small, rhomboidal, feebly imbricate on the
back. Ventral plates much broader than long (the largest nearly or quite twice as
broad as long), in straight longitudinal and transverse series; 8 or 10 plates across the
middle of the body. Digital denticulations very feebly developed, about equally on

132 MR. G. A. BOULENGER ON THE REPTILES

both sides. Young longitudinally streaked black and white on the body, with round
white spots on the limbs; tail pink; adult greyish or brownish, with more or less
distinct traces of light and dark longitudinal lines and longitudinal series of black and
pale spots; sometimes with large blue ocelli on the sides.

This species might be divided into two forms—the typical form from Europe and
Algeria, with smooth or faintly keeled scales, and the var. lineomaculatus, D. & B.,
from Morocco and Algeria, with strongly keeled scales,—were it not that there exists so
complete a transition between the two that it is almost impossible to draw the line.
It is a fact, however, that all specimens from Morocco, of which very many have been
examined, have strongly keeled scales, and, with rare exceptions, the subocular does not
reach the lip, but forms an angle inferiorly, wedged in between the fourth and fifth,
rarely fifth and sixth, labials. In Algerian specimens the dorsal scales may be perfectly
smooth, or more or less distinctly keeled, and it is but exceptionally that the subocular
does not border the lip. Out of 14 Spanish and Portuguese specimens preserved in the
British Museum, the subocular borders the lip in two specimens on both sides, and in a
third on one side only, the 11 other specimens being in this respect like those from
Morocco.

The number of scales round the body (ventrals included) varies from 70 to 85, and
that of femoral pores on each side from 22 to 27.

The largest Algerian specimen measures from snout to vent 70 millim., tail 110.

A. vulgaris inhabits the south of France (where very few specimens have been found),
the Spanish Peninsula, Morocco, Algeria north of the Sahara, and, probably, Northern
Tunisia. It is very common at Tangier, and is on record from Casablanca, Mogador,
Morocco, and the Plateau of Chiodma. In Algeria it is found on the coast and rarely
on the Plateaux. M. Lataste obtained specimens from Oran, Daya, Plateau of Sersou,
Algiers, Wed Sedeur (between Laghouat and Djelfa), Setif, between Aumale and Beni
Mansour, Bordj-Bou-Arrerij, and Tebesa.

4, Eremias, Wiegmann, 1834.

Nostril pierced between three (or four) nasals. Eyelids movable. Collar more or
less distinct. Dorsal scales small ; ventral shields tetragonal, feebly imbricate. Digits
with keeled lamelle inferiorly, not denticulated laterally. Femoral pores.

A single species in Barbary.

1. Eremias GurruLata, Lichtenstein, 1823.

E. pardalis, Gervais, Guichenot, Strauch.—Podarces simoni, Boettger.

Snout rather pointed ; nasals more or less swollen. Lower eyelid with a more or
less transparent disk formed of one, two, or several scales. A small occipital, in contact
with the interparietal; no auricular denticulation. Collar curved or more or less
angular, free or attached in the middle. Dorsal scales granular, smooth. Ventral
plates in straight longitudinal and transverse series, broader than long; 10 longitudinal

O32

AND BATRACHIANS OF BARBARY. 1)

series, the outer composed of very small plates. Coloration very variable, usually pale
greyish or brownish above, with blackish dots or series of black spots and white ocelli,
or with a broad dark grey vertebral band, &c. ; young with dark and light streaks along
the sides; lower parts white.

This is a very variable species, especially with respect to the scaling of the lower
eyelid and the development of the collar. Specimens with imperfectly transparent
eyelid, in which the median disk is broken up into numerous scales, and with the
collar distinct only on the sides, have been named £. simoni by Boettger, who has
kindly enabled me to examine the type specimens preserved in the Senckenberg Museum,
two of which are now in the British Museum. ‘Specimens with similar eyelids, but
with the collar free right across the throat, were named £. guttulata by Duméril and
Bibron, and such as have the eyelid perfectly transparent and formed of two scales,
and the collar free only at the sides, represent EH. pardalis of the same authors. The
large number of specimens examined by M. Lataste convinced him, however, that no
such division can be carried out, and I arrived at the same conclusion when working
out the extensive series in the British Museum. ‘The following table shows the varia-
tions in the Algerian and Tunisian specimens in the Museum :—

-¢, Aumale. ©, Tilremt. 3, Susa. @, Cabes |
os A Nee ae eS |
(Civ bie ee cain cameo eae eta Distinct, attached | Distinct only at | Distinct, attached | Free all round. |
in the middle. the sides. in the middle. |
Number of scales in trans- |
parent palpebral disk .. 3 2 6 4
Number of scales round
| middle of body (ventrals)
maCkWEel) Losatogddooe 68 52 | 59) 52
Hlemoralyporesierrteitork 17-17 10-10 14-13 11-10

The largest Barbary specimen examined by me measures 45 millim. from snout to
vent, tail 72.

The range of EH. guttulata is a very wide one, extending in Africa from Morocco to
Egypt, and in Asia from Arabia and Syria to Sind. In Morocco, the species is recorded
by Boettger from Casablanca and between Mogador and Morocco. In Algeria, it is
recorded by Strauch from Oran, Algiers, and Laghouat, and by F. Muller (Verh. nat.
Ges. Basel, vi. 1878, p. 625) from the Plateau of Sersou; M. Lataste obtained it at
Biskra, Mraier, Tuggurt, Gardaia, Berrian, Bou-Guelfaia, Tilremt, Laghonat, near
Aumale, between Aumale and Beni-Mansour, and at Bordj-Bou-Arrerij. In Tunisia,
the species is known from Susa (Brit. Mus.), Cabes, and the Chotts (André), and was
collected by M. Lataste at Cabes, Hadedj, Tamesred and Matmata, Mettamer, Plateau
of Haonaia and Djebel Domeur, Zarzis, Kebili, and Djerba Island.

Figured, ‘ Description de l’Eypte, Rept., Suppl. pl. ili. figs. 1, 2.

VOL. XI.—PaRT 111. No. 6.—October, 1891. x

134 MR. G. A. BOULENGER ON THE REPTILES

5. Opuiors, Ménétries, 1832.

Nostril between two to four nasals. Eyelids immovable, the lower united with the
upper, with a very large transparent disk. Collar absent or very indistinct. Dorsal
scales imbricate and strongly keeled. Digits with sharply keeled lamelle inferiorly,
not denticulated laterally. Femoral pores.

A single species in Barbary.

1. OPHIOPS OCCIDENTALIS, Boulenger, 1887.
Ophiops elegans, Lataste, Boettger.

Head-shields smooth. Nostril between an upper and a lower nasal, followed by one
or two postnasals ; four supraoculars, first and fourth small, all in contact with the
supraciliaries ; occipital very small or absent. Dorsal scales very large and sharply
keeled, larger than the laterals, and but little if at all smaller than the caudals ; 26 to
30 scales round the middle of the body, ventrals included. 6 to 11 (usually 7 or 8)
femoral pores on each side. Olive or bronzy above, with black spots, and one or two
light longitudinal streaks on each side ; lower parts white.

The North-African Ophiops is closely allied to, but perfectly distinct from the South-
Eastern Asiatic O. elegans, with which it was at first confounded. The absence of
granules between the supraoculars and the supraciliaries at once distinguishes it. The
scales are also larger. 26 to 30 round the body instead of 30 to 40.

The largest specimen measures 44 millim. from snout to vent, tail 98.

I have given a figure of this species in the third volume of the British Museum
‘Catalogue of Lizards,’ pl. iil. fig. 2.

The occurrence of Ophiops in North Africa was first recorded by Lataste, in 1880 ;
but two specimens from Susa, Tunisia, collected by Mr. Fraser, had been for many years
in the British Museum. In 1880, also, Peters recorded it from ‘Tripoli (Mon. Berl. Ac.
1880, p. 308). And in 1885, Boettger (in Kobelt, Reis. Alg. u. Tunis, p. 467) described
a specimen obtained by Dr. Kobelt at Biskra. M. Lataste obtained it in Algeria, at
Batna, Wed Sedeur, Yukuss (near Tebesa), and Portes de Fer, and examined specimens
collected by the Roudaire expedition at the Eastern Chotts ; in Tunisia, at Hadedj des
Matmata, and between the latter locality and Cabes, Medina, and at Ferriana and
Tamesmida.

or

AND BATRACHIANS OF BARBARY. 15

Fam. 7. SCINCID.

This large and cosmopolitan family is represented im Barbary by four genera,
distinguishable as follows :—

A. Palatine bones in contact on the median line of the palate; nostril in a
Sime eenaSall eee Mies) Uric Niche emtersbee, a yiglhes iti ate a Je cesta goer eaeb liens PAV Uae
B. Palatine bones separated on the median line.
a. Rostral not entering the nostril.

Nostril pierced in the nasal; digits not denticulated laterally . . . . . . . 2. Eumects.
Nostril between an upper and a lower nasal; digits denticulated laterally . . . 3. Scrncvs.

6. Nostril pierced between the rostral and a very small nasal, in an emargina-
HOM, Ol UNS nox SMOlL 5 5c 6 6 5 @ 8 6 6 8 o 5 o o o 4 Chamoru,

The genus Ophiomorus, Bibron, is erased from this fauna for reasons stated above,
p- 120.

1. Masur, Fitzinger, 1826.

Palatine bones in contact mesially. Eyelids movable. Nostril pierced in a single
nasal; supranasals present; preefrontals and frontoparietals present. Limbs well
developed, pentadactyle ; digits subcylindrical or compressed.

A single species in Barbary.

1. Masuta virrata, Olivier, 1804.

Lower eyelid with an undivided transparent disk. Normally no postnasal; parietals
usually in contact behind the interparietal. Ear-opening oval, about as large asa lateral
scale, with two or three pointed lobules, one of which usually is long. Nuchal and
dorsal scales strongly tricarinate ; 52 or 84 scales round the middle of the body, dorsals
a little larger than laterals or ventrals. ‘The adpressed limbs meet or slightly overlap,
or fail to meet. Olive or brown above, with a more or less distinct lighter vertebral
stripe ; and two narrow whitish lines on each side, the lower commencing below the
eye and passing through the ear; these light streaks may be edged with black lines or
bands; lower parts yellowish or greenish white.

From snout to vent 75 millim., tail 105.

Strauch records this Scink from the Mzab, and Lallemant adds the Souf. M. Lataste
received specimens from the plateaux of the Province of Constantine and collected
two specimens at Biskra. He also found it in Tunisia, at the Oasis of Tozeur,
and in abundance at Cabes. ‘The range of J. vittata extends eastwards to Egypt
and Syria.

Figures: Olivier, Voy. Emp. Othom. pl. xxix. fig. 1; Savigny, Descr. Egypte, Rept.,
Suppl. pl. 11. figs. 5 & 6.

136 MR. G. A. BOULENGER ON THE REPTILES

2. Eumecrs, Wiegmann, 1854.

Palatine bones not meeting on the median line of the palate; pterygoids toothed.
Eyelids movable, scaly. Nostril pierced in the nasal (which may be divided); supra-
nasals present ; preefrontals and frontoparietals present. Limbs well developed, penta-
dactyle; digits subcylindrical or compressed, not serrated laterally.

Two species in Barbary.

1. EuMscrs scHNEIDERI, Daudin, 1802.

Nasal in contact with the two anterior upper labials ; parietals entirely separated by
the interparietal. Ear-opening rather large, with four or five long pointed lobules
anteriorly. 22 to 28 scales round the middle of the body, perfectly smooth, those of
the two median series very broad. The adpressed limbs just meet or fail to meet.
Olive-grey or brownish above, uniform or with irregular golden-yellow spots or longi-
tudinal streaks ; a yellowish lateral streak, extending from below the eye to the hind
limb ; lower parts yellowish white.

From snout to vent 160 millim., tail 205.

This large and fine Scincoid inhabits Egypt, Syria, Armenia, Persia, and Baluchistan,
and extends westwards to Southern Tunis, where specimens were found by Dr. André
at Cherb Berrania, and by M. Lataste at Matmata, Wed Kebiriti (north of Chott
Fejej), and Gafsa. The specimen from the south-eastern frontier of Algeria, mentioned
by A. Duméril, probably belongs to this species.

Figured in Geoffroy, Descr. Egypte, Rept. pl. iii. fig. 3, and pl. iv. fig. 4.

2. EUMECES ALGERIENSIS, Peters, 1864. (Plate XVI.)

Scincus (Plestiodon) cyprius, Gervais, Strauch.—? Plestiodon aldrovandii, Guichenot.—Eumeces
pavimentatus, Boettger.

This occidental form is distinguished from the preceding by a somewhat heavier
build, the nasal in contact with the first labial only or with a very small portion of the
second, shorter ear-lobules, and 30 to 32 scales round the body, the dorsals distinctly
striated. Brown above, with small, yellowish, black-edged ocelli and large orange-red
spots, which often form irregular transverse bands on the body.

From snout to vent 200 millim., tail 220.

Common in Morocco, at Casablanca and Mogador. In Algeria it appears to occur
only in the Province of Oran, where it is recorded by Strauch from St. Cloud, Le Sig,
and Arzew. Specimens from Fleurus, near Oran, were examined in the Oran Museum
by M. Lataste.

3. Scincus, Laurenti, 1768.

Palatine bones not meeting on the median line of the palate; pterygoids toothed.
Eyelids moyable, scaly. Nostril pierced between two nasals; supranasals present ;

—

AND BATRACHIANS OF BARBARY. 137

prefrontals and frontoparietals present. Limbs well developed, pentadactyle ; digits
flattened, serrated laterally.
Two species in Barbary.

1. Scincus rascratus, Peters, 1864.

Head oval, pyramidal, with broad, obtusely truncated snout; nostril lateral ; loreal
region rounded; eye Jarge; ear-opening large, almost entirely covered by two very
large scales. Rostral moderately large, separated from the frontonasal by the supra-
nasals ; six supraoculars; parietals as long as interparietal. Body cyclotetragonal, the
sides of the belly rounded. 24 or 26 scales round the middle of the body; dorsal
scales striated, the two vertebral series largest, at least twice as large as the ventrals ;
a double series of large transverse nuchals. Digits feebly depressed, feebly denticulated
laterally. Yeilowish or orange above, with seven transverse black bands, of which the
first is on the nape, the second and third on the back, and the fourth on the sacrum.

From snout to vent 147 millim., tail 77.

The type specimen, from Geryville, which I have seen in the Berlin Museum, is
described in detail by Peters. The following notes were taken by M. Lataste from his
Tunisian specimen :—

Three prefrontals, middle one trapezoid and in contact with the truncate posterior
border of the frontonasal ; five pairs of nuchals; two superposed anterior loreals, the
lower deeper but shorter than the upper; parietals considerably shorter than the
frontal ; nine upper labials on each side, eighth largest, fifth and sixth under the eye ;
second and third postmentals divided into two; the two auricular scales much larger
than the temporal scales immediately preceding them. 26 rows of scales.

millim.
BRON OnGie GO WaMe 6 6 6 5 @ a 8 6 -o . 6) oll
IROL. SOMO. GRIP 6 deo) ake eo beeen woes)
AVVavch Glam Obatnicercl het cen ah) enti erate. .0TR oP oc eae ARS AIT t 2 6)

This is a very rare Lizard; only four specimens are actually known. The type
specimen noticed by Strauch under Scincus officinalis is from Geryville, in the Sahara
of the Province of Oran; one specimen was picked up, in a mummified condition, in
the plain of Arad, south of Cabes, near Sidi-Guenao, by M. Lataste, and is now preserved
in the Paris Museum; a third specimen, from Khartoum, is in the St. Petersburg
Museum, as well as the fourth, which is without locality, and served as the type of
Strauch’s Cyclodus brandti in 1866.

2. ScINCUS OFFICINALIS, Laurenti, 1768.

Snout cuneiform, truncate, strongly projecting; nostril pierced in the canthus
rostralis; loreal region concave ; eye very small; ear-opening distinguishable, covered

138 MR. G. A. BOULENGER ON THE ‘REPTILES

by two fringed scales. Rostral very large, forming a suture with the frontonasal or in
contact with the anterior angle of the latter shield; six supraoculars; parietals shorter
than the interparietal, followed by three to five pairs of nuchals; seven to nine upper
labials. 26 to 28 (very rarely 30) scales round the middle of the body, all pertectly
smooth ; dorsals usually not or scarcely larger than ventrals. Sides of belly angular.
Digits much depressed and strongly serrated laterally. Yellowish or brownish above,
each scale with small brown and whitish spots or shafts; frequently with more or less
marked dark transverse bands across the body; lower parts uniform whitish.

From snout to vent 120 millim., tail 59.

This species inhabits the Sahara and the borders of the Red Sea. It is not recorded
from Morocco, but was found by M. Maury at Ain-Sefra, in the south of the Province
of Oran, close to the Moroccan frontier. M. Lataste found it common in the Algerian
and Tunisian Sahara; its occurrence further north, at Dyelfa, for instance, whence it is
recorded by Strauch on the authority of Loche, is doubtful. It is only found in the
sand, in which it burrows with great rapidity. It is eaten by the Arabs.

Very good figure in ‘ Description de Egypte,’ Rept., Suppl. pl. ui. fig. 8.

4, Cuarciprs, Laurenti, 1768.

Palatine bones not meeting on the median line of the palate, which is toothless.
Hyelids movable, lower with an undivided transparent disk. Nostril pierced between
the rostral and a very small nasal, in an emargination of the former shield; supranasals
present; preefrontals and frontoparietals absent. Limbs short or rudimentary.

Six species in Barbary, distinguishable as follows :—

A. Snout conical.
a. Har-opening much larger than the nostril.
Limbs peutadactyle

yim bs, 4d 1. ocellatus.
Limbs tridactyle; second and third toes equal 2. lineatus.
Limbs tridactyle; second toe longer than third 3. tridactylus.

6. Har-opening not or scarcely larger than the nostril.
Limbs tetradactyle (rarely pentadactyle)
Fore limb didactyle, hind limb tridactyle

4. mionecton.
5. mauritanicus.
6

B. Snout wedge-shaped ; sides of belly angular; limbs penta- or tetradactyle. 6. sepoides.

1. CHALCIDES OCELLATUS, Forskal, 1775.

Snout obtuse, scarcely projecting beyond the labial margin ; ear-opening much larger
than the nostril. Nostril pierced just above the suture between the rostral and the
first labial; supranasals distinct; frontal longer than broad; usually the fifth labial
entering the orbit. Body cylindrical or rather depressed. 24 to 40 scales round the
middle of the body. Limbs short but well developed and pentadactyle; length of the
hind limb thrice and one third to four anda half times in the Jength from snout to

AND BATRACHIANS OF BARBARY. 139

vent. The coloration of the upper parts varies considerably ; lower parts uniform
whitish.

Four forms occur in Barbary :—

A. Forma TyPica.

28 or 30 scales round the body. Olive or brown above, ocellated with black spots,
sometimes confluent into irregular transverse bands, bearing central white dots or
longitudinal shafts. Measures up to 140 millim. from snout to vent.

Figured by Savigny in the Descr. Egypte, Suppl. pl. ii. fig. 7.

Ranges from the Algerian Sahara to Eeypt, Syria. Cyprus, Arabia, Persia, and Sind.
It is only found south of the Atlas; M. Lataste’s specimens are from Tuggurt,
Ghardaia, and Cabes.

B. Var. TiLIGUGU, Gmelin, 1788.

28 to 54 scales round the body (usually 30 or 82). Above olive or brown, with black
and white ocelli, and a more or less distinct lighter lateral band sometimes edged with
black inferiorly. Stouter and larger than the preceding, reaching a length of 170
millim. from snout to vent.

An excellent figure is given in Bonaparte’s ‘ Fauna Italica.’

Inhabits Sardinia, Sicily, and South Italy, Algeria and Tunis, and the intermediate
islands; also Tripoli, Egypt, North-Western Arabia, and Abyssinia, It is common
and generally distributed in Algeria and Tunis north of the Sahara.

C. Var. virtatus, Boulenger, 1890. (Plate XVII. fig. 1.)

30 to 34 scales round the body (usually 32). Bronzy brown above, without ocelli;
a light upper and a black lower lateral band. From snout to vent 115 millim.
All the specimens from Tangier belong to this variety.

D. Var. PoLYLEPIS, Boulenger, J890. (Plate XVII. fig. 2.)

34 to 40 scales round the body (usually 36 or 38). Dark brown above, usually each
scale with a small round yellowish spot; sides of neck with vertical black and white
bars, which disappear in the adult. From snout to vent 150 millim.

First noticed by Boettger from Casablanca, Mogador, and the city of Morocco.

2. CHaLcipEs Lingatus, Leuckart, 1828. (Plate XVII. fig. 3.)

Snout obtuse, scarcely projecting ; ear-opening much larger than the nostril. Nostril
pierced entirely in advance of the suture between the rostral and the first labial ;
supranasals distinct ; frontal longer than broad; fourth labial entering the orbit.
Body cylindrical, much elongate ; 22 to 26 scales round the middle of the body. Limbs
very small, tridactyle; the second toe as long as the third; the length of the hind limb
equals at least the distance between the ear and the fore limb, and is contained 12 to

140 MR. G. A. BOULENGER ON THE REPTILES

15 times in the distance from snout to vent. Bronzy olive above, uniform or with nine
or eleven dark brown longitudinal streaks, as broad as or broader than the interspaces
between them, which occupy the middle of each seale.

From snout to vent 126 millim., tail 154.

C. lineatus inhabits the south of France, Liguria, the Iberian Peninsula, Morocco
(whence I have received three specimens from M. H. Vancber), and Algeria. Whether
the specimen of “ Seps chalcides” from Casablanca, recorded by Dr. Boettger, belongs
to this or the following species, is unknown; but Dr. Boettger kindly informs me that
a second specimen from Casablanca recently examined by him belongs to C. lineatus.
‘The only Algerian specimen I have seen is one obtained by M. Lataste at El Guerra,
and now in the collection of Dr. J. de Bedriaga. It is uniform olive above, and has
26 scales round the middle of the body; its length from snout to vent is 140 millim.,
tail 92, fore limb 10, hind limb 12.

3. CHALCIDES TRIDACTYLUS, Laurenti, 1768.
Seps chalcides, Guichenot, Strauch.

Very closely. allied to the preceding. Limbs weaker still, the hind one usually
shorter than the distance between the ear and the fore limb, and contained 15 to 24
times in the length from snout to vent; third toe shorter than second. Olive or
bronzy above, uniform or with darker and lighter longitudinal streaks, which are
constantly in even number.

From snout to vent 183 millim., tail 200.

Inhabits Italy, Sardinia, Sicily, Tunis, and Algeria. Gasco records it from Egypt
(Alexandria). M. Lataste found it in Algeria at Oran, Bona, Aumale, between Azesga
and Tifrit, and between Bou-Saada and Biskra, and received it from Maison-Carrée and
the Plateau of Sersou. Strauch mentions it from the Mzab. M. Lataste states in
his notes that specimens of this species were obtained near Tunis by M. M. Sédillot
in 1883.

Well figured in Bonaparte’s ‘ Fauna [talica.’

4, CHALCIDES MIONECTON, Boettger, 1873.

Spout rather conical than wedged-shaped, but with distinctly projecting labial edge ;
ear-opening not or scarcely larger than the nostril, on a line with the mouth. Nostril
pierced entirely in advance of the suture between the rostral and the first labial; supra-
nasals distinct or united; frontal longer than broad; fifth labial entering the orbit.
Body much elongate; sides of body not distinctly angular, although somewhat more
so than in the preceding species; 24, rarely 26, scales round the middle of the body.
Limbs short, tetradactyle, rarely with a rudimentary fifth digit: the length of the hind
limb equals or a little exceeds the distance between the anterior border of the orbit
and the fore limb; the length of the latter equals about three-fourths its distance

AND BATRACHIANS OF BARBARY. 141

from the ear-opening. Brown above, usually with small yellowish black-edged ocelli ;
a broad yellowish stripe on each side of the back; a blackish streak from nostril to
eye; lips spotted with blackish ; lower surfaces white.

From snout to vent 80 millim., tail 62.

‘This Moroccan species is intermediate between C. ocellatus and C. sepoides.
Boettger’s specimens are from Tangier, Casablanca, and between Mogador and Morocco.
I have examined numerous specimens from Tangier, Larache, Casablanca, and
Morocco.

Figured by Boettger, Abh. Senck. Ges. ix. 1873, pl. —. fig. 6.

5. CHALCIDES MAURITANICUS, Dum. & Bibr. 1839.

Allied to the preceding. Snout conical, slightly projecting; ear-opening minute,
scarcely distinguishable ; nostril entirely in advance of the suture between the rostral
and the first labial; fourth upper Jabial entering the orbit; supranasals distinct.
Body much elongate. 18 scales round the body. Limbs very short; the anterior
didactyle, the posterior tridactyle; third toe nearly twice as long as second; the hind
limb equals the length of the head, and the fore limb the distance between the end of
the snout and the posterior border of the eye. Yellowish or greyish above, with a
lateral band formed of closely-set large black dots.

From snout to vent 71 millim., tail 45.

I have examined the two type specimens, from Oran, preserved in the Paris Museum.
A third specimen, from Nemours, Province of Oran, obtained by M. Gazagnaire in 1888,
is in M. Lataste’s collection.

Figured by Guichenot, Explor. Sc. Alg., Rept. pl. ii. fig. 1.

6. CHALCIDES SEPoIDES, Audouin, 1829.
Sphenops capistratus, Gervais, Strauch.

Snout wedge-shaped, with projecting labial edge; eye very small; ear-opening
appearing as an oblique slit at the commissure of the mouth, covered with a fringe of
three or four pointed scales. Nostril pierced entirely in advance of the suture between
the rostral and the first labial; supranasals fused to a single shield; frontal as broad
as or a little broader than long; fourth labial entering the orbit. Body much elongate;
sides of belly angular; 24 scales round the middle of the body. Limbs weak, penta-
or tetradactyle, the hind pair more developed than the front pair; the length of the fore
limb equals half its distance from the centre of the eye, that of the hind limb about
the distance between the nostril and the fore limb. Yellowish above, with more or
less distinct light brown longitudinal streaks; a dark brown streak on each side of the
head, beginning from the nostril and passing through the eye ; lower surfaces white.

From snout to vent 95 millim., tail 75.

VOL. XIII.—PART 111. No. 7.—October, 1891. Y

142 MR. G. A. BOULENGER ON THE REPTILES

C. sepoides occurs from Senegambia and Algeria to Kgypt, Arabia, and Syria. Like
the true Scink, it is essentially a sand Lizard, adapted for burrowing. It is scarce in
Algeria and Tunisia. Gervais records it from the Souf and Strauch from the Mzab.
M. Lataste received from Major Oudri a specimen from Mraia, and found a single one
himself at Tuggurt. He obtained another specimen in Tunisia, at Mettamer, and records
it from Tozeur (collected by Dr. André).

Figured in the ‘ Description de l’Egypte,’ Rept., Suppl. pl. ii. figs. 9 & 10.

Suborder II. RHIPTOGLOSSA.

Fam. 1. CHAM ASLEONTID.

1. CuamM@LEon, Laurenti, 1768.
Eye large, covered by a thick granular lid pierced with a small opening for the
pupil. No ear-opening. Body compressed. Digits arranged in bundles of two and
three; claws simple ; scales on soles smooth. ‘Tail prehensile, at least as long as head

and body. A large genus, inhabiting Africa, Madagascar, the south of Spain, Asia
Minor, Syria, Arabia, and India and Ceylon.

1. CHAM#LEON VULGARIS, Daudin, 1802.
C. africanus, Schlegel.—C. cinereus, Strauch.

Casque raised posteriorly, with strong curved parietal crest; the distance between
the commissure of the mouth and the extremity of the casque nearly equals the
length of the mouth; a strong lateral crest, becoming indistinct as it ascends towards
the extremity of the parietal crest ; a small but very distinct occipital dermal lobe on
each side, extending to the extremity of the parietal crest. Body uniformly granulate;
no dorsal crest ; a more or less distinct series of conical scales on the anterior part of
the vertebral keel ; a series of conical, slightly enlarged granules on the median line
of the throat; no ventral crest. No tarsal process. ‘Tail usually a little shorter
than head and body. A white line from chin to vent ; usually two or three series of
pale spcts along each side.

Total length 274 millim., tail 155.

The Chameleon inhabits North Africa, the South of Spain, Asia Minor, and Syria.
It is more or less common all over Barbary, where shrubs or trees occur, and it is also
found in the oases of the Sahara. A good whole figure is given in the ‘Description de
VEgypte,” Rept. pl. iv. fig. 8; and of the head in the British Museum Catalogne of
Lizards, iii. pl. xxxix. fig. 1. A specimen from Bou-Saada has recently been described
as C. saharicus by F. Muller, Verh. nat. Ges. Basel, vil. 1885, p. 715, pl. si., and viii.
1887, p. 295, but it is nothing but a hali-grown C. vulgaris.

AND BATRACHIANS OF BARBARY. 1438

Suborder III. OPHIDIA.

The Snakes of Barbary belong to three families, distinguished as follows :—

Mandible with coronoid element ; maxillary horizontal; prefrontal bones forming

a suture with nasals; vestiges of pelvis terminating in a claw-like spur visible on

HON MGS OH UNO MAME cog po en Ge, Woe va Gv ode bY om eo well, wetokike,
Mandible without coronoid element; maxillary horizontal; prefrontals not

forming a suture with nasals ; no ndbiannts Of pelvis) 3. & . . 2. Colubride.
Mandible without coronoid element; maxillary vertically erect fpetnendicnlag

to transpalatine; prefrontals not forming a suture with nasals; no rudiments

CHINE: ight oh) SPCR CATER Wy EN tree kt Eo a roti Abacos is Hier cE ING wuipmint bu n'i5 Xa gC)

Fam. 1. BOID/.

A single genus.
1. Eryx, Daudin, 1805.

Anterior maxillary and mandibular teeth a little longer than posterior. Head
covered with small scales; a mental groove. Hye very small, with vertical pupil.
Scales very small, smooth or keeled. ‘ail very short, not or but very slightly pre-
hensile ; subcaudals simple.

1. Eryx gacutus, Linneus, 1766.

Rostral large, with angular horizontal edge; nine to eleven upper labials, separated
from the eye by one or two series of scales. Scales smooth, in 40 to 50 longitudinal
rows. Pale greyish or yellowish brown above, with large dark brown transverse
blotches or alternating spots; belly white, uniform or with black dots.

Total length 510 millim., tail 45.

Inhabits North Africa, Greece, Turkey, and South-western Asia. Strauch records
the snake from Oran and the Algerian Sahara, and M. Lataste obtained it at Wed
Magra, Barika, and Ngaous, between Msila and Batna. A specimen was obtained in
Tunisia at Bir-oum-Ali, south of Tebesa, near Tamesmida, by M. Sédillot. Not on
record from Morocco.

Figured, ‘ Description de l Egypte,’ Rept. pl. vi. fig. 2.

Fam. 2, COLUBRID.
The Colubride of Barbary belong to 8 genera, falling into three divisions or series :—
A. Aglypha. No grooved fangs. Harmless.
a. Scales smooth or feebly keeled.
Head short; pupilround; nosubocular. . ....... =. =. . . J. Coroner.

Snout cuneiform ; pupil vertically elliptic . . . . .... . =. . . 2. LyrorHyNnonus.
Y2

144 MR. G. A. BOULENGER ON THE REPTILES

Head elongate ; one or more suboculars below the preocular . . . . . 3. Zaments.

ouScales\strongly, keeled yar) yh cunt etn ne ee OPNDONOMURS

B. Opisthoglypha. Grooved fangs behind the series of maxillary teeth.
Suspected, or poisonous to a slight degree.

a. Some of the anterior maxillary teeth enlarged and separated from

those following by an interspace.

Head short ; eyes rather small, pupil vertically subelliptic . . . . . . 5. Macroproropon.
Head elongate; eye large, with round pupil . 6. PsaMMopPHIs.
6. Only the grooved maxillary teeth enlarged . . . . . . . ).) . 67. Ca@nopetris.
C. Proteroglypha. Grooved fangs anteriorly. Poisonous.
Week clamor mo lowalgielel os 6 5 5 6) 56 6 6 6 6 6 6 6 6 o eb INA

1. CoronELia, Laurenti, 1768.

Maxillary teeth 12 to 14, increasing in size posteriorly ; mandibular teeth subequal.
Head short, scarcely distinct from neck; eye rather small, with round pupil. Body
cylindrical ; scales smooth, in 19 to 32 rows, with apical pits; ventrals rounded ; tail
moderate; subcaudals in two rows.

‘wo species in Barbary.

1. CoRONELLA AMALL&, Boettger, 1881. (Plate XVIII. fig. 1.)

Snout prominent ; rostral as deep as broad, produced posteriorly between the inter-
nasals, the portion seen from above about half as long as its distance from the frontal ;
suture between the internasals one-third the length of that between the preefrontals ;
frontal a little longer than its distance from the end of the snout, a little shorter than
the parietals ; loreal longer than deep; one pre- and two postoculars; temporals
2+3; eight upper labials, fourth and fifth entermg the eye; four lower labials in
contact with the anterior chin-shields ; posterior chin-shields three fourths the length of
the anterior. Scales in 2] rows. Ventrals 190-193; anal divided; subcaudals 63-64.
Grey-brown above, with reddish-brown spots and four rather indistinct dark longi-
tudinal bands; vertebral region light; a pair of elongate dark brown spots on the
nape ; a black streak on each side of the head, from the nostril, through the eye, to
the angle of the mouth; a dark band between the eye crossing the preetrontals; a
black line below the eye, on the suture between the fourth and fifth upper labials,
Lower surfaces coral-red, with quadrangular black spots.

Total length 590 millim., tail 72.

This species is intermediate between C. gustriaca and C. girondica, agreeing with the
former in the size and shape of the rostral shield, with the latter in all other respects.

The two specimens from which the above description is taken were obtained by
M. H. Vaucher in the Benider hills, near Tangier. The type specimen of Fhinechis
amalie, Boettg., is from between Tetuan and Tangier. M. Lataste received a single
specimen from Bona, through Dr. Hagenmiiller.

AND BATRACHIANS GF BARBARY. 145

2. CoRONELLA GiRONDICA, Daudin, 1803.

Snout scarcely prominent ; rostral much broader than deep, just visible from above ;
suture between the internasals half as long as that between the prefrontals; frontal a
little longer than its distance from the end of the snout, a little shorter than the
parietals ; loreal longer than deep; one pra- and two postoculars; temporals 2+3 ;
eight upper labials, fourth and fifth entering the eye; four lower labials in contact
with the anterior chin-shields, which are as long as the posterior. Scales in 21 rows.
Ventrals 200 ; anal divided ; subcaudals 59. Grey-brown above, with blackish spots;
a pair of elongate blackish spots on the nape; a black streak on each side of the head,
from the nostril, through the eye, to the angle of the mouth; a dark band between
the eyes, crossing the prefrontals ; a black line below the eye, on the suture between
the fourth and fifth upper labials. Lower surfaces yellowish (in spirit), with quadran-
gular black spots. ,

Total length 450 millim., tail 80.

The above description is taken from a single female specimen from the city of
Morocco, described in 1874 by Dr. Boettger, to whose kindness I am indebted for
examining it. The species was recorded from Tangier by Gervais, who probably took
for it the common Macroprotodon cucullatus, which does not appear in his list. Boettger
records it from Tlemsen, in the Province of Oran. Nothing more can be said as to its
distribution in Algeria since the records of authors may be based on the closely-allied
C. amalie. M. Lataste did not meet with it in Algeria. C. gérondica inhabits the
south of France, Spain and Portugal, and Italy.

Figured in Bonaparte’s ‘Fauna Italica,’ and in Jan’s lcon. Gén. Ophid. livr. 17, pl. iii.
figs. 1-3.

2. Lyroruyncuus, Peters, 1862.

a

Maxillary teeth 6 to 9, posterior much longer than anterior; mandibular teeth
subequal. Head slightly distinct from neck, with cuneiform projecting snout ;
eye moderate; pupil vertically elliptic; rostral large, four-sided, projecting, concave
beneath ; nostril an oblique slit between two nasals. Body cylindrical; scales smooth,
in 19 rows, without apical pits; ventrals obtusely angulated laterally; tail mederate ;
subcaudals in two rows,

A single species in Barbary.

1. LyrorHyncHUS DIADEMA, Dum. & Bibr. 1804.

Rostral angularly bent, with straight horizontal edge, detached on the sides, the
portion visible from above as long as its distance from the frontal ; suture between the
internasals much shorter than that between the prefrontals ; frontal nearly as long as its
distance from the end of the snout, slightly shorter than the parietals; a small, squarish
loreal; one or two preoculars, with or without a subocular below; two postoculars ;
temporals 1+2 or 2+3; seven or eight upper labials; fourth, fifth, or fourth and

146 Mk. G. A. BOULENGER ON THE REPTILES

fifth, entering the eye; three lower labials in contact with the anterior chin-shields ;
posterior chin-shields as long as or a little longer than the anterior, and separated
from each other by two series of scales. Scales in 19 rows. Ventrals 160 to 188; anal
divided ; subcaudals 86 to 46. Pale buff or cream-colour above, with a series of large
transversely rhomboidal dark spots; a dark median band along the head and nape,
sometimes confluent with an interocular transverse band; an oblique dark band from
the eye to the angle of the mouth; lower parts uniform white.

Total length 450 millim., tail 60.

This sand-snake was first described from a specimen obtained in Algeria in the
Western Desert. Gervais records another from the Souf. A specimen from Batna is
in the St. Petersburg Museum. M. Lataste received it from Mraier through Major
Oudri, and from Ferriana, in Tunisia, through Dr. Robert. M. Valéry-Mayet found it
at Gourbata, near Gafsa, and Marquis Doria has it from Kairouan. It is also known
to occur in the Sennaar, in Arabia, Syria, and Persia. Figured by Jan, Icon. Gén.
Ophid. livr. 10, pl. vi. fig. 2.

3. ZaMENIS, Wagler, 1850.

Maxillary teeth 10 to 20, increasing in size posteriorly; mandibular teeth subequal.
Head elongate, distinct from neck ; eye moderate or rather large, with round pupil ;
one or more suboculars. Body elongate, cylindrical ; scales smooth or feebly keeled,
in 15 to 31 rows, with apical pits; ventrals rounded or with an obtuse lateral keel;
tail long; subcaudals in two rows.

The study of the species of this genus is a most perplexing one, owing to the
complete passage which exists between many forms which it seems nevertheless
necessary to distinguish. As regards the Barbary Zamenis, however, of which only
three species are to be distinguished, the only difficulty will be in the definition, other-
wise than by coloration, of Z. algirus from Z. hippocrepis; I am afraid we must at
present content ourselves with the statement that the former has the scales disposed in
25 longitudinal rows and usually one labial entering the eye, whilst the latter has the
scales usually in 27 rows (at the point where they are most numerous) and usually the
eye completely separated from the labials by a series of suboculars.

Three other species have been ascribed to the fauna of Barbary, but must be erased :
1. Z. florulentus, Schleg., noticed and head figured by Gervais (Mém. Ac. Montpellier,
i. 1857, p. 512, pl. v. fig. 4), from Laghouat, is, in my opinion, the Z. algirus, and
therefore widely different from Schlegel’s species. 2. Zamenis atrovirens, Shaw, recorded
from Algiers by Giinther (Cat. Colubr. Snakes, 1858, p. 102) from a specimen purchased
of a dealer in Paris, and mentioned from Mogador (specimen purchased, 1870) in the
list of the animals having lived in the London Zoological Gardens. 3. Zamenis ater,
Giinther (Ann. & Mag. N. H. (4) ix. 1872, p. 22), stated to be from Biskra, is a melanotic
variety of the South-American Liophis regine.

AND BATRACHIANS OF BARBARY. 147

The three species described below may be distinguished as follows :—

Scales perfectly smooth, in 25 rows; usually one labial entering the eye; a pair

of internasals and a pair of prefrontals . . . . Rees che melemalounes.
Scales perfectly smooth, in 27 (rarely 25 or 29) rows; ustally no val entering

the eye; a pair of internasals and a pair of prefrontals . . . . . . =. 2 hippocrepis.
Scales more or less distinctly keeled, in 25 to 33 rows; no labial entering the

eye; usually three or more prefrontals; analentire . . . . . . . . 38. diadema.

1. ZAMENIS ALGIRUS, Jan, 1863.
Z. florulentus, Gervais.

Scales smooth, in 25 rows. One preocular, with a subocular below; two post-
oculars and a subocular; temporals 24+3; nine upper labials, rarely ten, fitth or sixth
usually entering the eye, but sometimes separated by an additional subocular. Ventrals
214 to 252; anal divided, rarely entire; subcaudals 92 to 100. Pale olive, yellowish
brown, or greyish above, with three alternating series of darker transverse bars, and a
series of dark spots along each side of the belly; a more or less distinct blackish
crescentic band on the nape, extending to the sides of the throat; a blackish spot
below the eye; lower parts white. Young specimens may have the head entirely
black above.

Total length 940 millim., tail 250.

This species appears to be restricted to the Algerian and Tunisian Sahara. M.
Lataste obtained specimens in Algeria at Biskra, Laghouat, between Bou-Saada and
Biskra, and in Tunisia at Cabes, Raz-el-Oued, Ain-Zerig, Hadedj, Djebel Domeur,
Djerba Island, Tozeur, Ferriana, Gafsa; and M. Valéry-Mayet on Kerkenna, and at
Djebel Berda and Madjoura; specimens were obtained on the Tunisian Chotts by
Dr. André.

Of 14 examples examined by M. Lataste, all have 25 rows of scales; two have the
anal single. In 22 out of 28 cases the fifth labial enters the eye, in two cases the
sixth, in four the eye is completely separated from the labials by suboculars.

Figure: Jan, Icon. Gén. Ophid. livr. 48, pl. iv. fig. 2.

2. ZAMENIS WIPPOCREPIS, Linnzeus, 1766.

Scales smooth, in 25 to 29 rews, usually 27. One or two preoculars; two post-
oculars ; a series of three or four suboculars usually completely separates the eye from
the labials; temporals 2-3 or 3+3; eight or nine (rarely ten) upper labials, fifth or
sixth very rarely entering the eye. Ventrals 222 to 248; anal divided, rarely entire ;
subcaudals 79 to 107. Brown, reddish, yellow, or pale olive above, with a dorsal series
of large dark brown, black-edged rhomboidal spots, on each side of which is a series of
smaller alternating spots; these spots may be black and so large as to reduce the
ground-colour to a mere chain or series of X’s of pale colour; a dark cross-band
between the eyes, and a (A- or horseshoe-shaped band on the back of the head, which

148 MR. G. A. BOULENGER ON THE REPTILES.

may be confluent with an elongate spot on the nape; the spots confluent into three
longitudinal streaks on the tail; yellowish or red beneath, with or without black dots,
but constantly with a lateral series of black spots.

Total length 1340 millim., tail 270.

One of the commonest snakes in Morocco and Algeria north of the Sahara, and
extending to the northern parts of Tunisia. Obtained by M. Lataste in Algeria, at
Oran, Batna, M’sila, and Tebesa, and received by him from the Plateau of Scrsou and
from Bona. Found also in Spain and Portugal and in Sicily. Recorded from Egypt
through confusion with Z. nwmmifer, Riippell.

Figured in Bonaparte’s ‘ Fauna Italica.’

3. ZAMENIS DIADEMA, Schlegel, 1837.
Z. cliffordii, Strauch.

Scales more or less obtusely keeled, in 25 to 33 rows. Head-shields more or less
broken up, there being frequently three transverse series of shields between the
rostral and the frontal ; three or more preefrontals; three to five loreals; two to four
pre- and three or four postoculars; a series of suboculars separates the labials from
the eye; temporals small, scale-like; 10 to 15 upper labials. Ventrals 210 to 278 ;
anal entire; subeaudals 65 to 110. Pale buff or sandy grey above, with more or less
marked dark spots, of which the median usually form a series of rhombs; lower parts
white, rarely with small blackish spots.

Total length 1800 millim., tail 340.

Ranges from the Algerian Sahara eastwards to North-western India. M. Lataste
collected specimens at Biskra, Wargla, and Wed Magra in Algeria, and at Mettamer
and Ferriana in Tunisia. The specimen from Biskra has 33 rows of scales, two from
Wargla 25, two others from Wargla 27, the one from Wed Magra 32, the one from
Mettamer 27, and the one from Ferriana 32.

Figures: Jan, Icon. Gén. Ophid.. livr. 20, pl. ii.; Geoffroy, Descr. Egypte, Rept.
pl. vii. fig. 1.

4. 'Troprponorus, Kuhl, 1824.

Maxillary teeth 12 to 22, posterior longest; mandibulary teeth subequal. Head
distinct from neck ; eye moderate or rather large, with round pupil. Body cylindrical ;
scales keeled (rarely smooth), in 15 to 29 rows, with or without apical pits; ventrals
rounded ; subeaudals in two rows. Nasal bones very small; vertebral hypapophyses
distinct throughout the vertebral column.

Two species in Barbary.

1. TRopIpoNoTUS NaTRIX, Linnens, 1766.

Scales in 19 rows. Seven upper labials, third and fourth entering the eye; usually
one pre-and three postoculars. Greyish, brownish, or olive above, uniform or with

AND BATRACHIANS OF BARBARY. 149

black spots; young with a yellow, black-edged collar, which may disappear in the
adult. A variety (persa, Pall., murorum, Bp.) with two whitish or yellowish stripes
along the back is common in Italy, but has not been found in Algeria.

Total length one metre or more.

Found all over Europe, Western and Central Asia. Rare in Algeria; not recorded
from Morocco or Tunisia. Dr. Strauch found it at Algiers, M. Lataste at La Chiffa,
Algiers, and Tifret, and Dr. Hagenmiiller on Mt. Edough, near Bona.

Figures: Bonaparte, ‘ Fauna [talica’ (Natrix torquata).

2. TRoprponotus vipERINUS, Latr., 1802.

Scales in 21 rows, rarely 23 (one specimen from L’Arba and one from Cabes,
Lataste). Seven upper labials, third and fourth entering the eye; one or two prae-
and two postoculars. Greyish, brownish, or reddish, with a more or less distinct
black zigzag stripe along the back and ocelli on the sides; frequently (var. ocedlata,
Wael., aurolincata, Gervais) two pale stripes along the back.

Total length 850 millim.

Common all over Morocco, Algeria, and Tunisia, wherever water occurs, for this
species is still more aquatic than the preceding. Inhabits also the Iberian Peninsula,
France, Switzerland, and Italy.

Figure: Bonaparte, ‘ Fauna Italica.’

5. Macroproropon, Guichenot, 1850.

Maxillary teeth 10 or 11, fourth and fifth or fifth and sixth enlarged, followed by an
interspace, the two posterior grooved ; mandibular teeth increasing in size to the sixth,
which is followed by an interspace, the posterior teeth small. Head short, slightly
distinct from neck; eye small, with vertically subelliptic pupil. Body moderately
elongate, cylindrical; scales smooth, in 19 to 25 rows, with apical pits; ventrals
rounded ; tail moderate; subcaudals in two rows.

1. Macroprotopon cucuLLatus, Geoffroy, 1827.

M. mauritanicus, Guichenot.—Lycognathus teniatus et textilis, Dum. & Bibr.—Coronella brevis,
Ginther.—M. maroccanus, Peters.

The unique species of this genus bears a general similarity to the Palzarctic species
of Coronella, from which it is, however, easily distinguished by the much depressed
snout, the subelliptic pupil, the very broad and low rostral shield, and the presence of
a single anterior temporal, which is usually separated from the postoculars, the sixth
upper labial touching the parietal. Hight upper labials, fourth and fifth entering the
eye; one pre-and two postoculars. ‘The number of rows of scales varies from 19 to 25.
Specimens from Tangier and Tetuan appear to have constantly 21 rows of scales, like

VOL. XI1I.—PaRT 11. No. 8.—October, 1891. Z

150 MR. G. A. BOULENGER ON THE REPTILES

those from the south of Spain (Badajos, coll. Lataste; Andalusia, Brit. Mus.). The
number varies from 21 to 25, 23 being the usual number, in those from Casablanca,
Mogador, and Morocco (Coronella brevis, Gthr., Macroprotodon maroccanus, Peters) ;
out of 7 specimens from the city of Morocco, I find 5 with 23 rows, one with 21, and
one with 25, and variation in a nearly equal proportion is recorded by Boettger on 44
specimens from Casablanca. In Algerian and Tunisian specimens there are usually
19 rows, as in all those from the Baleares, Tripoli, and Egypt; but I note 21 rows in
one from Tunis, and M. Lataste found the same number in one specimen from Batna.
Ventrals 153 to 192; anal divided ; subcaudals 40 to 51. The coloration also varies
greatly, and irrespective of the scaling. The large black blotch on the head, whence
the name cucullatus, is not very frequent in occidental specimens; but a dark collar,
descending to the sides of the neck, is usually present, and an oblique dark streak
below the eye is constant. The belly may be uniform yellowish or almost entirely
black; it is usually yellowish, with quadrangular black spots, as in Coronella girondica
and amaliv ; usually a dark streak along the middle of the subcaudal region.

The largest specimen examined by me measures 490 millim., tail 85.

This species inhabits the south of the Iberian Peninsula, the Baleares, and the
whole of North Africa, penetrating into the Sahara. It is common and generally
distributed in Morocco, Algeria, and Tunisia.

Figured by Guichenot, Explor. Sc. Alg., Rept. pl. ii. fig. 2.

6. PsamMMopuis, Boie, 1827.

Maxillary teeth 10 to 13, one or two of the middle ones much enlarged, fang-like, and
preceded and followed by an interspace, the two posterior grooved ; anterior mandibular
teeth long, posterior small. Head elongate, distinct from neck, with angular canthus
rostralis; eye rather large, with round pupil. Body elongate, cylindrical; scales
smooth, in 15 or 17 rows, with apical pits; ventrals rounded or obtusely angulated
laterally ; tail long ; subcaudals in two rows.

A single species in Barbary.

1. PsaAMMOPHIS SIBILANS, Linneeus, 1766.
Ps. punctatus, D. & B., Gervais.

Head narrow and elongate; internasals much shorter than the prefrontals; frontal
very narrow, in contact with the preeocular; loreal much elongate; one or two pre-
and two or three postoculars; eight or nine upper labials, fourth and fifth, or fifth and
sixth * entering the eye. Scales in 17 rows. Brown or greenish above, with longitu-

All the specimens obtained in Algeria and Tunis by M. Lataste have 9 upper labials, fifth and sixth
entering the eye (var. punctata, D. & B.).

AND BATRACHIANS OF BARBARY. 151

dinal series of darker dots, or with two or three yellowish longitudinal streaks; upper
lip and lower parts yellowish white, uniform or dotted with black.

Total length 920 millim., tail 290. Reaches a length of 1} metre.

Inhabits North Africa, Arabia, Syria, Asia Minor, and Southern Russia. Recorded
by Gervais from Sefissifa, near the Moroccan frontier of Algeria, and by Strauch from
the Mzab. M. Lataste found this snake common in the Algerian and Tunisian Sahara,
and in the southern parts of the Plateaux; in Algeria, at Biskra, Tuggurt, Laghouat,
Bou-Guelfaia, and Bou-Saada; in Tunisia at Raz-el-Wed, Cabes, Djebel Domeur, El
Hamman des Beni-Zib, Fratis, Taferma, Tamesred, Bougrara, Mettamer, and Nebech
el Dib.

Figured in ‘ Description de !Egypte,’ Rept. pl. vii. fig. 4.

7. Caopettis, Wagler, 1830.

Maxillary teeth 10 to 16, subequal, followed by a very long grooved fang; anterior
(especially 3rd to 5th or 6th) mandibular teeth long, posterior small. Head more or
less elongate, distinct from neck, with angular canthus rostralis and projecting supra-
ocular; eye large, with round pupil; nostrila crescentic slit between two nasals. Body
elongate, cylindrical; scales smooth, more or less distinctly grooved longitudinally in
the adult, with rather indistinct apical pits, in 17 or 19 rows ;-ventrals rounded ; tail
rather long ; subcaudals in two rows.

Two species in Barbary.

1. Ca@Lopettis LacertiIna, Wagler, 1824.
C. monspessulanus, Rozet.

Snout rather prominent, obtuse ; forehead and loreal region concave. Internasals
much shorter than the prefrontals; frontal narrower than the supraocular, in contact
with the preocular ; two loreal shields; one pree- and two or three postoculars ; eight
upper labials, fourth and fifth entering the eye. Scales more or less distinctly grooved,
in 19 (rarely 17) rows. Olive or brown above, with or without dark spots; sides often
blackish ; yellowish white beneath, uniform or spotted or clouded with brown or olive.

Total length 1580 millim., tail 550.

This fine snake inhabits Southern Europe, the whole of North Africa, north and
south of the Atlas, and South-western Asia. It is one of the commonest snakes all
over Morocco, Algeria, and ‘Tunisia.

Excellent figures are given by Bonaparte, ‘ Fauna Italica,’ and by Savigny, ‘ Descrip-
tion de ’Egypte, Rept., Suppl. pl. v. figs. 2 & 3.

2. C@LOPELTIS PRODUCTA, Gervais, 1857.

Snout very prominent, obtusely pointed; forehead flat or slightly convex ; loreal
72

4

152 MR. G. A. BOULENGER ON THE REPTILES

region concave. Rostral wedged in far between the internasals, which are a little
shorter than the preefrontals ; frontal as broad as the supraocular; a single loreal; one
pre- and two or three postoculars ; eight upper labials, fourth and fifth entering the
eye. Scales not very distinctly grooved, in 17 rows. Pale yellowish brown or sandy
grey above, with brown or blackish spots; two oblique brown bars on each side of the
head behind the angle of the mouth ; lower parts white.

Total length 600 millim., tail 110.

Originally described from the Algerian Sahara, this species has since been found in
Tunisia, Tripoli, Egypt, Nubia, and Arabia.

Gervais’s specimen was obtained on the borders of the Sahara, between Bou-Alam
and the Arbas, in the Province of Oran. A single specimen was obtained in Tunisia,
at Bou-Hedma, near Gafsa, by M. Valéry-Mayet.

Figured by Jan, Icon. Gén. Ophid. livr. 34, pl. u1. fig. 2.

8. Nara, Laurenti, 1768.

Poison-fangs with a distinct groove anteriorly, followed by one to three small solid
teeth. Head distinct from neck; no loreal. Hye rather small, with round pupil.
Neck dilatable. Body cylindrical; scales disposed obliquely, smooth, in 15 or more
rows; tail moderate ; subcaudals in a single or double row.

A single species in Barbary.

1. Nata watz, Linneeus, 1766.

Seven upper labials, sixth largest and in contact with the postoculars. 21 to 23
scales across the neck, 19 to 21 across the middle of the body. All the specimens
hitherto found in Barbary have the eye completely separated from the labials by a series
of suboculars (var. annulifera, Peters, 1854) and the coloration of the upper parts is a
uniform dark or blackish brown.

Reaches a length of 2 metres.

The African Cobra is found all over Africa south of the Atlas.

A. Duméril (Rev. Mag. Zool. 1856, p. 554) mentions a specimen received by the
Paris Museum from Morocco, and Boettger records it from the interior of Morocco.
M. Lataste obtained specimens near Biskra, and one at Zarzis, in Tunisia; a second
specimen was obtained in Tunisia by M. Valéry-Mayet at the well of El Aia, near Wed
Leben, and a third at Raz-el-Aioun, between Gafsa and Tameghza, by M. Sédillot.

The var. annulifera is beautifully figured in the ‘ Expédition de /Egypte,’ Rept.,
Suppl. pl. in.

AND BATRACHIANS OF BARBARY. 15

Poi)

Fam. 3. VIPERIDZ.

The Vipers of Barbary belong to three genera :—

Lateral series of scales running in straight longitudinal lines; subcaudals in two

TOMS pir iene rst SO suLdey Newest Siena eater te enor alae part ee Use ne ews eee DVDR ECAR
Lateral scales disposed obliquely ; subcaudalsin tworows . . . . . . . . 2. Cwrasres.
Lateral scales disposed obliquely ; subcaudals ina single row . . . . . . . 3. Ecurs.

1. Vipera, Laurenti, 1768.

Upper surface of head covered with scales or small shields. Scales keeled, in 21 to
38 straight longitudinal rows. Subcaudals in two rows.
The three species found in Barbary may be distinguished as follows :—

A. Nostrils lateral.

Scales in 21 rows ; end of snout turned up 1. latastiz.
Seales in 23 to 27 rows; snout blunt BU a es 2. lebetina.
B. Nostrils directed upwards ; scales in 29 to 31 rows . 3. arietans.

1. Vieera LavAstil, Bosca, 1878.
V. aspis, Strauch.

Snout turned up, terminating in a low erect appendage; rostral twice as deep
as broad; a large supraocular shield, separated from its fellow by five to eight longi-
tudinal series of smooth scales; two or three series of scales between the eye and the
labials; 9 to 11 upper labials. Scales in 21 rows. Pale brown above, with a zigzag
or scalloped dark brown band and a lateral series of dark brown spots; a dark band
from the eye to the neck ; lower parts blackish, dotted or spotted with white or grey,
spotted black and white.

Total length 530 millim., tail 60.

This species, first described from Spain and Portugal, where it had long been con-
founded with the more oriental V. ammodytes, is of particular interest as forming a
complete passage between the latter species and V. aspis ; in fact, the lesser development
of the rostral appendage as compared with V. ammodytes, and the greater development of
the same as compared with I’. aspis, are the only characters which distinguish V. latastii
from its two allies. _

It was first recorded in Barbary from Guyotville, near Algiers, under the name of
V. aspis, by Strauch, and it has since been found on Mt. Edough, near Bona, by
Dr. Hagenmiiller and M. Lataste. Dr. Kobelt collected two specimens near Tangier,
which have been described by Boettger. __

Figured by Bosca, Bull. Soc, Zool. France, 1878, pl. iv.

154 MR. G. A. BOULENGER ON THE REPTILES

2. VIPERA LEBETINA, Linneeus, 1766.

V. brachyura, Schlegel. Echidna mauritanica, Guichenot.

Snout obtuse. Nostril between three shields; rostral a little broader than deep; a
narrow supraocular shield is present or absent; upper surface of head covered with
small, imbricate, strongly keeled scales, 9 to 12 across the forehead, from eye to eye ;
three or four series of scales between the eye and the labials; 10 to 12 upper labials.
Scales in 23 to 27 rows. Pale grey-brown above, with darker spots or cross-bands,
which are very distinct in the young, but feebly marked or absent in the adults;
lower parts whitish, powdered with grey.

Total length 14 metre.

Inhabits North Africa and South-western Asia, and the Greek Island Milo, to
Northern Baluchistan, Afghanistan, and Cashmere. It is not uncommon near Oran ;
has been recorded from the interior of Morocco by Boettger; M. Lataste saw specimens
from near Batna and Mt. Edough, Bona’, in the collection of M. Hénon, and captured
others in Tunisia at El Hammam des Beni-Zib, Djebel Domer, Taferma, and Tames-
mida, and received a specimen from ‘Tadjera, near Mettamer, through Capt. Rebillot.

Figured by Guichenot, Explor. Sc. Alg., Rept. pl. iii.

3. VIPERA ARIETANS, Merrem, 1820.

Snout very short and broad. Nostrils large, directed upwards, pierced between three
shields; rostral more than twice as broad as deep; no supraocular shield; upper
surface of head covered with small, imbricate, strongly keeled scales, 9 to 11 across the
forehead, from eye to eye; three or four series of scales between the eye and the
labials; 12 to 15 upper labials. Scales in 29 to 31 rows. Yellowish or pale brown
above elegantly marked with large 4/-shaped, dark brown, black-edged spots; a large
dark blotch covers the crown, separated from a smaller interorbital spot by a transverse
yellow line; an oblique dark brown band below, and another behind the eye.

Reaches a length of 1220 millim.

This large and deadly snake is found over the greater part of Tropical and South
Africa. From West Africa it penetrates into Southern Morocco, a specimen from the
Valley of Sous having been recorded by Boettger.

A good figure is given by Wagler, Icon. Amph. pl. xi.

2. Crrastes, Wagler, 1830.
Upper surface of head covered with scales. Scales keeled, in 23 to 35 rows, the
laterals disposed obliquely. Subcaudals in two rows.
‘Two species in the Algerian and ‘Tunisian Sahara.

* This locality, M. Lataste informs me, is somewhat doubtful. The Batna specimens are uniform greenish
above.

AND BATRACHIANS OF BARBARY. 155

1. Cerastes vipers, Linneus, 1766. (Plate XVIII. fig. 2.)
Vipera avicenne, Strauch.

Snout very short, rounded; eye very small. Head-scales small, tubercularly keeled ;
nostril between two small shields; no horn-like scales over the eye; three or four
series of keeled scales between the eye and the labials. Scales in 23 to 25 rows. A
strong keel on each side of the ventrals. Yellowish brown, sand-colour above, with or
without darker spots; lower parts white.

Total length 240 millim., tail 30.

This small Viper inhabits the desert of Algeria, Tunisia, Tripoli, and Egypt. It was
first recorded in Algeria from the Western Desert by Duméril and Bibron (Hehidna
atricauda),and M. Lataste collected specimens at Bou-Saada, Biskra, and between Bou-
Saada and Biskra. M. Lataste did not come across it in Tunisia, but was informed that
Marquis Doria possesses several specimens from the southern parts.

The specimen figured by Jan under the name of Vipera avicenne is a Cerastes cornutus
without horns. As there exists no figure of this viper, I have supplied one on
Plate XVIII. fig. 2.

2. CERASTES cornuTUS, Forskal, 1775.
Vipera cerastes, Schlegel, Gervais, Strauch.

Snout very short and broad ; eye rather small, usually with a large, ribbed, horn-like
scale above. Head-scales small, tubercularly keeled ; nostril in a single small nasal ;
four or five series of keeled scales between the eye and the labials. Scales in 29 to 35
rows. An obtuse keel on each side of the ventrals. Yellowish-brown sand-colour
above, with or without brown spots forming regular longitudinal series, white below.

Total length 630 millim., tail 75.

Inhabits the Sahara, extending eastwards to Arabia. Schlegel records it from Biskra,
Strauch from Djelfa, Laghouat, Saida, Biskra, and Batna, and Boettger from Geryville.
M. Lataste obtained it in Algeria, at Bou-Saada and at Bordj-Tayer-Rasson, and in
other localities in the sandy Sahara, and at Wed Magra, in the southern parts of the
High Plateaux; he found it common throughout Southern Tunisia.

Figures: Jan, Icon. Ophid. livr. 45, pl. v. (Vipera cerastes and V. avicenne), and
Geoffroy, Descr. de !Egypte, Rept. pl. vi. fig. 3.

3. Ecuis, Merrem, 1820.
Upper surface of head covered with scales. Scales keeled, in 25 to 35 rows, the
laterals disposed obliquely. Subcaudals in a single row.
A single species in Barbary.
1. Ecuis carrnata, Schneider, 1801.

Snout very short and rounded. Nostril between three shields, the anterior and upper

156 MR. G. A. BOULENGER ON THE REPTILES

of which are in contact with the rostral ; head covered with small keeled scales, among
which an enlarged supraocular is sometimes present ; two series of scales between the
eye and the Jabials; 11 or 12 upper labials. Scales in 29 to 35 rows. Pale buff,
greyish, reddish, or pale brown above, with three series of whitish spots edged with
dark brown; a zigzag dark brown band may run along each side: a cruciform or
Ax-shaped, whitish, dark-edged marking on the head; lower parts whitish, uniform or
with brown dots.

Total length 600 millim., tail 60.

Inhabits the desert sandy districts of North Africa, South-western Asia, and India.
M. Lataste examined several specimens from Biskra in M. Hénon’s collection at Con-
stantine, and a single specimen was found at Tadjera, near Mettamer, in Tunisia, by
M. Letourneux.

Figure: ‘ Description de ’Egypte,’ Rept., Suppl. pl. iv. fig. 1.

BATRACHIA.

The Batrachians are very poorly represented in North Africa. They fall into two
Orders :—

I. ECAUDATA, Frogs and Toads, in which the tail is present only during the larval period.
II. CAUDATA, Newts and Salamanders, in which the tail persists throughout life.

Order I. ECAUDATA.

Four Families :—

1. Ranide. Upper jaw toothed; diapophyses of sacral vertebra not dilated.

2. Bufonide. Jaws toothless ; diapophyses of sacral vertebra dilated.

3. Hylide. Upper jaw toothed; diapophyses of sacral vertebra dilated ; digits expanded at the
end, the distal phalanx claw-shaped and swollen at the base.

4. Discoglosside. Upper jaw toothed; diapophyses of sacral vertebra dilated ; anterior dorsal
vertebrze with short ribs.

Each of these four Families is represented in Barbary by a single genus. “The four
genera may be distinguished as follows :—

I. Pupil horizontal ; vomerine teeth, if present, between the choane.
A. Digits not dilated at the end.
Tongue deeply notched, bifid posteriorly ; teeth in upper jaw and on palate. 1. Rana.
Moan UTNE GR MOWING 6 6 6 6 6 6-0 6 4 56 6 co co BORO,
A. Digits dilated at the end. 3. Hyza.

II. Pupil round or subtriangular; vomerine teeth behind the choane . . . . 4. Discoenossus.

AND BATRACHIANS OF BARBARY. Lo7

Fam. 1. RANIDZE.

1. Rana, Linneus, 1766.

Pupil horizontal. Vomerine teeth. Tongue forked and free behind. Fingers free,
toes webbed.
This almost cosmopolitan genus is represented in Barbary by a single species.

1. Rana EscuLenta, Linneus, 1766.
R. viridis, Guichenot.

Vomerine teeth between the choanz. Interorbital space narrower than the upper
eyelid; tympanum distinct, about two thirds the size of the eye. Toes entirely webbed.
A glandular lateral fold. Green, olive, or bronzy brown above, usually with black spots
and a pale green vertebral line. Male with two external vocal sacs.

This species is distributed over nearly the whole of the Palearctic Region. The
form found in Barbary, which has been named var. latastii by Camerano, I regard as
inseparable from the var. ridibunda, Pall., which is found in Western and Central
Asia, Eastern Europe and Germany, the south of France and the Pyrenean Peninsula,
Tripoli, Egypt (?), and the Sinaitic Peninsula. It is distinguished from the typical
form by the smaller size of the inner metatarsal tubercle, which is blunt, not com-
pressed. It is found throughout Morocco, Algeria, and Tunis, penetrating into the
desert, where it was obtained as far as Wargla by M. Lataste, who also found it
everywhere in Tunisia.

I append measurements in millimetres of several specimens (females) in the collection
of the British Museum :—

Tangier. Constantine. Tunis.
pal tI Be SSG | a
Hromsnoub/ to vent js) 3.5 <=). 85 69 69 65 90 53 75 75
eng tbrotatibiaeer erect 45 37 35 35 44 27 37 37
Length of foot (from outer meta-
tarsal tuberele) .......:...- 45 37 34 34 46 27 38 38
Length of inner toe .............. 13 11 10 10 13 > 11 103
Length of inner metatarsal tubercle. 4 3 3 3 4 22 33 33

From which we see that the length of the inner metatarsal tubercle is contained from
3 to 3% times in the length of the inner toe, and from 11 to 12 times in the length of
the foot, which nearly equals that of the tibia.

The var. ridibunda is figured, from German specimens, in Proc. Zool. Soc. 1885,
pl. xl.

VOL. XUI.—pParT tl. No. 9.—October, 189i. 2A

158 MR. G. A. BOULENGER ON THE REPTILES

Fam. 2. BUFONID.
1. Buro, Laurenti, 1768.

Pupil horizontal. ‘Tongue elliptic or pyriform, entire and free behind. Fingers
free, toes webbed.

The true Toads form a large genus, represented over the greater part of the world.
Three species are found in Barbary, distinguishable as follows :—

A. Subarticular tubercles under the toes all single; a tarsal fold; tympanum
about half the diameter of the eye eh NEO vc 1. viridis.
B. Subarticular tubercles under the median toes in pairs.

A tarsal fold; first finger much longer than second

2. mauritanicus.
No tarsal fold 3.

vulgaris.

1. Buro viripis, Laurenti, 1768.

B. variabilis, Gervais.

Interorbital space narrower than the upper eyelid ; tympanum about half the diameter
of the eye. First finger extending a little beyond second; toes half or two-thirds
webbed, with single subarticular tubercles ; a tarsal fold. Above with olive or greenish

spots on a greyish or pinkish ground; sometimes a yellow vertebral line. Male with a
subgular vocal sac.

From snout to vent 95 millim.

This Toad has a very wide distribution, being found over the greater part of Europe,
though not west of the Alps, Central and Western Asia, the Himalayas, and North
Africa. Common in Italy and the Baleares, but absent from the Iberian Peninsula.
It is known in Morocco, from Casablanca and between Mogador and Morocco, but has
not been found in the Northern Promontory. Strauch says it is common near Oran,

and Eichwald found it at Musaya, on the Algerian Atlas. Lataste received it from the

Plateau of Sersou, collected specimens at Oran, El Guerah, Bona, and Ghardaia, and
Tilremt, and found it common everywhere in Tunisia.

The characters upon which Bufo boulengeri, Lataste (Rev. Int. des Sciences, 1879,
p- 438), was founded, upon a single specimen from the Plateau of Sersou, have proved
to be individual, and M. Lataste now entertains no doubt that the name should be
regarded as a synonym of B. viridis.

2. Buro mauriranicus, Schlegel, 1841.
Bufo pantherinus, Guichenot, Strauch.—B. arabicus, Gervais.

Interorbital space concave, broader than the upper eyelid ; tympanum very distinct,
vertically oval, its greatest diameter not much more than half that of the eye. First
finger much longer than second; toes webbed at the base, with double subarticular

AND BATRACHIANS OF BARBARY. 159

tubercles ; a tarsal fold. Above usually with large insuliform, dark-edged, reddish-
brown spots. Male with a subgular vocal sac.

From snout to vent 140 millim.

This fine large Toad appears to be peculiar to Barbary. It has been found in every
part of Morocco yet investigated, and is very common all over Algeria as far south as
the limit of the Sahara, M. Lataste having obtained it at Biskra, Bou-Saada, and
Laghouat. It is found near Tunis, and M. Lataste met with it at Tozeur, Gafsa, and
Ferriana, but not at Cabes, where Bufo viridis and Rana esculenta are abundant.

I have given a figure of 5. mauritanicus in Proc. Zool. Soc. 1880, pl. li.

3. Buro vuLGARIS, Laurenti, 1768.

Interorbital space broader than the upper eyelid ; tympanum more or less distinct.
First finger extending scarcely beyond second; toes at least half webbed, with double
subarticular tubercles; no tarsal fold. Brown or dull olive above, with darker spots;
parotoid glands with a dark outer margin. Male without vocal sac.

From snout to vent 130 millim.

The Common Toad of Europe and Palearctic Asia is rare in Algeria, and at present
only known from Algiers (Strauch, Lataste), Tlemsen (Boettger), and Bona, where
Dr. Hagenmiiller states it is not unfrequent. It is undoubtedly the rarer of the three
toads of Barbary. Camerano has recorded it from Larache in Morocco. It has not
yet been obtained in Tunisia.

Fam. 5. HYLID.
1. Hyta, Laurenti, 1768.

Pupil horizontal. Vomerine teeth. ‘Tongue subcircular, entire or slightly nicked
and more or less free behind. Fingers free or webbed, toes webbed.
Of this large genus a single species occurs in Kurope and round the Mediterranean.

1. Hyta arporea, Linneeus, 1766.
H. viridis, Guichenot.

Fingers webbed at the base; tympanum distinct; upper parts perfectly smooth, belly
granular. Male with a large subgular vocal sac.

From snout to vent 50 millim.

The Tree-Frog of Barbary belongs to the var. meridionalis, Boettg., 1874 (=perezi,
Boscaé, 1880, darytonus, Héron-Royer, 1884), which is uniform green, without a dark
stripe along the side of the body, and with the green colour extending to the sides of
the throat. It inhabits the south of France, North Italy, Spain and Portugal, the
Canary Islands and Madeira, and Barbary, where it is common all over the Tell. It has
been found everywhere in Morocco.

This variety is figured by Bosca, Ann. Soc. Esp. x. 1881, pl. ii. figs. 7-10, and by
Héron-Royer, Bull. Soc. Zool. France, 1884, pl. ix.

2a2

160 MR. G. A. BOULENGER ON THE REPTILES

Fam. 4. DISCOGLOSSIDZ.

1. Discoetossus, Otth, 1836.

Pupil roundish-subtriangular. Tongue circular, entire, scarcely free behind. Vomer-
me teeth in long transverse series behind the choane. Fingers free, toes webbed.

1. Drscoauossus Pictus, Otth, 1836.

Snout longer than the diameter of the orbit, without canthus rostralis; tympanum
hidden or slightly distinct. First finger shorter than second; three metacarpal tuber-
cles, the inner very much developed in the male; toes webbed at the base in the female,
almost entirely in the male; a small inner metatarsal tubercle. Skin smooth or with
small flat warts above. Brownish, yellowish, reddish, or olive above, with dark light-
edged spots, sometimes confluent into longitudinal bands; some specimens with three
light dorsal stripes. Male without vocal sacs; during the breeding-season with black
rugosities on the chin, the inner metacarpal tubercle, the inner digits, and on the free
border of the web between the toes.

From snout to vent 60 millim.

North-African specimens have been distinguished by Camerano (Atti Acc. Torin. xiii.
1878, p. 548, and xiv. 1879, p. 447, figs.) under the name of D. scovazzi, chiefly on
account of their distinct tympanum; and this distinction has been recently upheld by
Héron-Royer with the new name of D. auritus (Bull. Soc. Angers, 1889, p. 177). A
renewed examination of the rich series in the British Museum has convinced me that,
as shown by Lataste (“‘ Etude sur le Discoglosse,” Act. Soc. Linn. Bordeaux (4) ii. 1879,
p. 275), but one species of Discoglossus can be admitted. A male specimen, collected
at Algiers by Mr. Sclater, has the tympanum completely concealed, and, on the other
hand, the organ is perfectly distinct ina male collected by M. Bosca on the Serra
Morena. The shape and extent of the dark temporal band are subject to much varia-
tion in specimens from the same locality, and I have failed to find any constancy in the
other very trivial distinctive characters pointed out by M. Heron-Royer.

D. pictus abounds in the Tell, from Moroeco to Tunis, and is also found in the
intermediate zone between the Tell and the Sahara, M. Lataste having obtained it as
far south as Batna. Moroccan localities are Tangier, Tetuan, Casablanca, Mogador,
and Morocco. Common round Tunis, and found on Galita Island by Marquis Doria.
Also recorded by Lataste from the Tunisian Chotts (Tozeur and Nefta). In Europe
D. pictus is known from Spain and Portugal, Corsica, Sardinia, Malta, and small neigh-
bouring islands,

M. Lataste has observed that, as in the European Alytes, the breeding-season extends
from the early spring to the end of summer.

Excellent figures of this Batrachian accompany Lataste’s memoir quoted above.

AND BATRACHIANS OF BARBARY. 161

Order II. CAUDATA.
Fam. 1. SALAMANDRID.

Lizard-like Batrachians breathing by lungs in the perfect condition, with well-
developed eyelids. Represented in Barbary by two genera.

1. Satamanpra, Laurenti, 1768.

Tail subcylindrical. ‘Terrestrial.

1. SALAMANDRA MaAcULOSA, Laurenti, 1768. (Plate XVIII. fig. 3.)

Skin smooth, shining, porous above; a distinct parotoid gland on each side of the
neck ; a lateral series of large warts; a strong gular fold. Black, with yellow markings.
Inhabits Central and Southern Europe, Syria, Morocco, and Algeria.

The Moroccan and Algerian specimens are separable as a variety which has been
named var. algira, by Bedriaga (Arch. f. Nat. 1883, p. 245). It is distinguished from
the European Salamander, including the Corsican form, S. corsica, Savi, to which the
Algerian specimens have been erroneously referred by some authors, by the more
slender build, the longer digits, and the longer tail, as may be seen from the following
measurements, in millimetres, of specimens in the British Museum :—

Tangier. Bona.
poe ; a aS
From snout tovent. . . 86 fie 62 AT 89 60
Tail . ae 65 55 43 80 52
Third toe . patie A 8 6:5 5 4 8 3)
ertthstoere a te: en aa 3 3 2 2 379 2

The yellow spots are few, round or oval, and disposed alternately im two series, but
never form longitudinal bands as is frequently the case in European specimens.

The Salamander is very local in Algeria, but abundant where it occurs. Guichenot
found it at Oran, and states that Col. Levaillant got it at Constantine. Lallemant
records it, on the authority of Letourneux, from Kabylia and Bona (Mt. Edough). It
has been found in great numbers in the latter locality by Dr. Hagenmiiller. M. Lataste
received it from L’Arba throngh M. Lallemant, and collected himself larval specimens
in that locality. Hanoteau and Letourneux record it from the forest of Akfadou,
Bougia, and Fort National, and Dr. Boettger from Bougia. The only locality in
Morocco where the Salamander has yet been found is the Benider Hills, near Tan-
gier; we are indebted to M. H. Vaucher for this discovery.

The North-African variety of Salamandra maculosa has not been figured betore.
Fig. 3a on Plate XVIII. represents a specimen from Mt. Edough, near Bona, and
figs. 3b, c two young from the Benider Hills, near Tangier.

162 MR. G. A. BOULENGER ON THE REPTILES

2. Mo.tar, Merrem, 1820.
Tail compressed. Aquatic during the breeding-season.
Three species in Barbary.

A. Palatine teeth not extending forwards beyond the line of the choane.

Palatine teeth forming a N; contour of lower jaw semicircular. . . . . . IL. potreti.
Palatine teeth forming a A; contour of Jower jaw semielliptic . . . . . . 2. hagenmuelleri.
B. Palatine teeth extending forwards beyond the line of the choane. . . . 3. waltlii.

1. MoLcE porreti, Gervais, 1825.
Euproctus rusconii, Guichenot. Triton nebulosus, Guichenot.

Palatine teeth in two slightly curved series, approximating in front, forming af.
Tongue subcircular or oval, free behind and on the sides. Head much depressed, as
long as broad or a little longer; snout broad, rounded, the contour of the jaws semi-
circular. ‘Tail longer than head and body. Skin tuberculate; a more or less distinct
gular fold. Olive above, yellow or orange beneath, with or without small black spots.

Total length 150 millim.

Common near Aigiers and at L’Arba, whence M. Lataste received specimens. He
also observed it at Guyotville, and received a specimen from Bougie, through Dr.
Hagenmiiller, in 1882. Guichenot records it from Oran, and Giglioli from near Tunis.

Figured under two different names by Guichenot.

2. MoLGE HAGENMUELLERI, Lataste, 1881. (Plate XVIII. fig. 4.)

Distinguished from the preceding by the series of palatine teeth being less far apart
anteriorly, forming a A; head longer, contour of jaws semielliptic. Fingers and toes
more slender.

Size smaller, 100 millim.

This species was founded on numerous specimens sent from Bona and its environs
by Dr. Hagenmiiller to M. F. Lataste. It has since been recorded from Biskra by
Dr. Boettger. Specimens received from Constantine through M. Hénon in 1881, induce
M. Lataste te regard MW. hagenmuelleri as a variety of MW. poireti, as there appears to be
a complete passage between them and the specimen from Bougie noticed above.

I have given figures of typical specimens from Bona, on Plate XVIII. fig. 4.

3. Mouee waLtiu, Michahelles, 1830.

Palatine teeth in two slightly curved series approximating in front, commencing in
frent of the line of the choanz. ‘Tongue subcircular, free on the sides and behind.
Head much depressed, as long as broad. ‘Tail longer than head and body. Ribs
very long, ending in a sharp point, which frequently perforates the skin. Skin
tuberculate; a strong gular fold. Olive-brown above, yellowish beneath, with
blackish markings.

AND BATRACHIANS OF BARBARY. 163

Total length 224 millim.

Inhabits the south of Spain and Portugal, and has been found in Morocco at
Tangier, Ceuta, and between Tangier and Tetuan. Specimens recently sent by
M. Vaucher were obtained in the marshes of Charf-la-Kaab, a few miles south of
Tangier.

Molge wailtlii has been several times figured ; the least unsatisfactory figure is that
given by Bonaparte, ‘ Fauna Italica.’

EXPLANATION OF THE PLATES.

PLATE XIII.
Fig. 1. Saurodactylus mauritanicus, D. & B. Mogador. (Page 109.)
a. Natural size.
6. Side view of head, x 3.
c. Lower view of head, x 3.
. Ptyodactylus lobatus, var. oudrii, Lataste. Bou-Saada. (Page 114.)
a. Natural size.
6. Side view of head, x 2.
c. Lower view of head, x 2.
Fig. 3. Tarentola mauritanica, var. deserti, Lataste. Wargla. (Page 116.)
a. Natural size.

eS
o
bo

6. Side view of head, natural size.
c. Lower view of head, natural size.
Fig. 4. Agama tournevillii, Lataste. Sahara. (Page 118.)
a. Natural size.
b. Side view of head, natural size.

PLATE XIV.

Fig. 1. Agama bibronii, A. Dum. Tangier. (Page 118.)
a. &,natural size.
6. 3, side view of head, natural size.
c. 2, natural size.
d. Young, natural size.
Fig. 2. Psammodromus blanci, Lataste. Lambesa. (Page 127.)
a. @, natural size, upper view.
b. Lower view.
c. Upper view of head, x 2.
d. Side view of head, x 2.

164

ON THE REPTILES AND BATRACHIANS OF BARBARY.

PLATE XV.

Lacerta ocellata, Daud. (Page 123.)

Figs. a, b,c, d. Adult ¢. Tunis. (Var. pater, Lataste.)
Fig. e. Young. Algiers. (Var. pater, Lataste.)
Fig. f. Upper view of head of 9. ‘Tangier. (Var. tangitana, Blgr.)
Fig. g. Upper view of head of ¢. Ferrol, Spain.
All the figures natural size.

PLATE XVI.

Pumeces algeriensis, Peters, natural size. Mogador. (Page 136.)

PLATE XVII.

Fig. 1. Chalcides ocellatus, var. vittatus, Blgr. Tangier. (Page 139.)
And (a) side view of head and anterior part of body, natural size.
Fig. 2. Chalcides ocellatus, var. polylepis, Blgr., natural size. City of
Morocco. (Page 139.)
And (a) side view of head and anterior part of body, x 2.

Fig. 3. Chalcides lineatus, Leuck. Tangier. (Page 139.)
PLATE XVIII.
Fig. 1. Coronella amalie, Boettg. Tangier. (Page 144.)

a. Natural size.
6. Upper view of head, x 2.
c. Side view of head, x 2.
Fig. 2. Cerastes vipera, L. ‘Tripoli. (Page 155.)
a. Natural size.
b. Upper view of head, x 2.
c. Side view of head, x 2.
Fig. 3. Salamandra maculosa, var. algira, Bedr. (Page 161.)
a. @, natural size. Bona.
6, c. Young. ‘Tangier.
Fig. 4. Molge hagenmuelleri, Lataste. Bona. (Page 162.)
a. &, Natural size.
6. 2, natural size.
c. Open mouth, X 2.

?
C

(ES ee ZA wy, EVE
VOM. LOO Bc (OC CACE?

@

eh

fo

5c.

ee

ie

Miritern Bros. imp.

1.SAURODACTYLUS MAURITANICUS 2.PTYODACTYLUS LOBATUS, var. OUDRII
3. TARENTOLA MAURITANICA,vaR.DESERTI. 4.AGAMA TOURNEVILLII.

NL

Peter Smit del.et hth.

ear

ii
ne

ack
eae aT

i

pata) ‘ n wen
Ws editor DEA ee ht
Tos i

Wey
PO aan abe

i
aS
Tie eear?. AS

Peter-Smit det. et lith

acess
a

ape

la

<e

IL,

AGAMA

BIBRON

T
{

I

At

0 <n,
rerereitt

a
ie}

2.PSAMMODROMIUS BLANCI

Mintern

Bros

imp

PS lion ae
FAIA E

i
Hania
saa

I

Peter Smit del. et lith.

6 + ~+—

ee

ad

Seve

ce
ie Sp

Pots

LAV.

Sf

C

AU

iL

CC

y
yy

@)

a

ten

imp -

Mintern Bros

del. et lith.

Smit

Peter

i

Maca ye bios
*

AL es

imp.

Mintern Bros.

Smut del. et hth.

Peter

PIS.

US, van. POLYI.E

OCELLA

2.CHALCIDES

JS

OCELLATUS var. VITTAT

CIDES

CHAL

alee

US).

Ht

CHALCIDES LINEA

oO.

AA eZ y 7 Wy ar G)P r
ans Loo. Boe Volt AM SC. SVUL

Ist

LEX
Sc

Re

Peter Smit del. et hth. Mintern Bros imp

1.CORONELLA AMALIA 2. CERASTES VIPERA
3 .SALAMANDRA MACULOSA,var . ALGIRA 4. MOLGE HAGENMUELLERI

[ 165 ]

VI. On a Skull of TYrogontherium cuvieri from the Forest Bed of East Runton,
near Cromer. By K.'T. Newton, F.G.S., F.Z.S., Geological Survey.

Received March 17th, 1891, read April 21st, 1891,

[Puate XIX.]

ConTENTs. Page
aalintroductoryn -tc.cs, tatelota cine setae asters Se eseae chornte aioatae clone eee 165
2. Description of the Skull of Vrogontherium from East Runton.......... 166
3. Dental Characters of the Trogontherium of the Forest Bed .......... 170
4, Comparison of the Forest Bed Skull with Fischer’s Vrogontherium cuviert 171
5. Comparison of the Forest Bed Skull with that of Conodontes boisvillettii.. 172
GreiMieagurem ents noes Searcy een ate ctiscot Newari en are tear epesvortienere om eeaiey Are eA te 173
eMC ONCLUSIONS tate ee ycreMenS oom ceauecedte hc ene OA Aar acne neta shai ac ate lore Reger ee hoe 174
Suebibliographiyarnpecgesr es citt tease whch sre eeacrccr ae Lie ee treterte ser als 174
Q; IbquemenwGn Gr PAO IIIS Gosencoeaoqhoosonvnondomoonansoeucenor 175

1. Introductory.

AT the time when the Geological Survey Memoir on the Vertebrata of the Forest Bed
was published (1882), the parts of the skull of Zrogonthertwm, which were known in
this country, were the maxille and premaxille, with incisors and cheek-teeth ; the only
example, however, of a last upper molar was the half tooth preserved, with the first
and second molars, in the maxilla described by Sir R. Owen", which is preserved in the
King Collection in the Museum of Practical Geology. The imperfect condition of this
last upper tooth deprived us of one of the chief characters by which Trogontherium is
distinguished from Castor, and rendered the determination of the British specimens less
certain than it would otherwise have been; the available materials, however, were
sufficient to justify Sir R. Owen’s reference of them to Trogontheriwm cuvieri, a form
described by Fischer” from near the Sea of Azof.

A skull of this rodent, from British deposits, has long been a desideratum, in order
that a close comparison might be made with the nearly perfect type skull; and such a
specimen has now been secured from the Forest Bed of East Runton by Mr. A. Savin, of
Cromer, who has been kind enough to send it to me for identification and description.
In many particulars this specimen (Plate XIX.) is in a more satisfactory condition for
examination than Fischer’s type, which is so much obscured by hard sandy matrix that
the relations of the bones cannot be clearly seen, while Mr. Savin’s specimen is almost
free from matrix, and the surfaces of the bones, as well as the sutures, are well

displayed.

* Geol. Mag. vol. vi. p. 49 (1869). 2 Moscow Soc. Nat., Mém. vol. ii. p. 250 (1809).
VOL. XIl.—Part Iv. No. 1.—April, 1892. 28

166 MR. E. T. NEWTON ON A SKULL OF TROGONTHERIUM CUVIERI

It is proposed in the first place to describe this new specimen in detail, at the same
time comparing it with the skull of the Beaver, to which it is nearly allied, and then
to institute a comparison between it and Fischer’s type; and finally to consider its
relation to the skull from the Pliocene of Saint-Prest, called by M. Laugel in 1862
Conodontes boisvillettii', which is believed to be referable to the same genus and species.

There has been much diversity of opinion as to the genus to which these Forest Bed
remains should be referred, and consequently the synonymy is very perplexing; this
has, however, been explained in the Memoir of the Geological Survey already referred
to, and the new specimen seems to me fully to confirm the opinions there expressed.
References to the authors who have written upon the subject will be found at
the end of this paper.

2. Description of the Skull of Trogontherium from East Runton.

This Forest Bed skull has very nearly the same parts preserved as Fischer’s type; it is
of a dark brown colour, and is much impregnated with iron. The nasal bones, as might
be expected, have fallen away; and both the incisor teeth are absent, but their alveoli
are intact. The jugal arch on both sides is broken off close to the cranium, leaving
only the bases of the maxillary and squamesal buttresses. ‘The bony external auditory
meatus is wanting on the right side, and the thin lamine of the palatines and ptery-
goids are broken away. There has evidently been a strong supraoccipital crest,
probably as well developed as in the Beaver; but it is now much abraded.

As this skull (Plate XIX.) bears a very close resemblance to that of a Beaver, its
description will be most intelligible if given in terms of comparison with that animal ;
and it is so little longer than the specimen from the fens, which will be used for
comparison, that the relative proportions, indicated by the measurements given in the
table on page 173, will be very obvious.

When viewed from above (fig. 1) the difference in general proportion between the
two skulls is clearly seen. The length of the cavity exposed by the removal of the
nasal bones is about the same in both specimens, but is much the widest in the
Trogontherium. ‘The fronto-premaxillary suture in the Zrogontherium is opposite the
middle of the maxillary buttress for the jugal arch, and the premaxilla is consequently
almost wholly in front of the maxillary buttress. In the Beaver this buttress is further
forwards, and the premaxilla extends for some little distance behind it. ‘The frontals
are longer and wider in Zrogontherium than in the Beaver; and the postorbital process
is about halfway between the anterior and posterior buttresses of the jugal arch, while
in the Beaver this process is further forwards. ‘The parietals are much shorter than
they are in the Beaver, and the same may be said of the interparietal. On each
parietal, near the sagittal suture, there is a ridge, which, running backwards, meets its
fellow at their junction with the triangular interparietal, and they are continued as a

' Bull. Soc. Géol. Fr. sér. 2, vol. xix. p. 167 (1862).

FROM THE FOREST BED OF EAST RUNTON. 167

prominent median crest along the latter bone. ‘The position of these parietal ridges,
depending as they do upon the development of the temporal muscles, would no doubt
vary in form in older skulls. The sagittal suture is even shorter than it is in the
Beaver, the parietals only joining for about half an inch. ‘The squamosal bone has
much the same form that it has in the Beaver; but its parietal margin is more deeply
sinuous, and the suture forms a strongly impressed groove along the side of the
brain-case.

On the side of the skull (fig. 5) the squamosal sends down a process from its hinder
part between the auditory meatus and mastoid, and it has a large foramen a little
behind the jugal buttress ; while anteriorly it ends at the postorbital process.

The jugal buttress of the maxilla descends from the upper surface of the skull about
halfway towards the alveolar margin; while in the Beaver this buttress extends as a
distinct ridge quite to the lower edge of the maxilla.

The suborbital foramen is situated as it is in the Beaver, but the plate of bone by
which it was covered is partly broken away on both sides. ‘This plate of bone in the
Beaver is developed anteriorly into a strong ridge for the attachment of muscles; and
judging from the still larger process remaining in the fossil below the foramen, the
muscles attached in this region were more strongly developed than they are in the
Beaver.

The premaxilla in the fossil commences directly in front of the suborbital foramen,
and is relatively larger than in the Beaver, being altogether more strongly developed,
evidently in relation to the powerful incisor, indicated by the large size of the alveolus.
The two premaxille form in front of the snout a vertical, roughened, triangular area,
the upper part of which is separated by a deep transverse groove, and forms a pair of
tubercles which encroach upon the lower part of the external nasal opening, and
give it a nearly horizontal lower margin: the Beaver’s nasal aperture is pointed
inferiorly. Seen from below (fig. 2) the premaxille are broader than in the Beaver, and
especially so at their front part: where the two bones meet on the palate they form
a median crest, which, dividing, is continued around the palatal foramina, and again
uniting extends along the entire length of the palate to the posterior nares. ‘The
palatal foramina are elongated slits as in the Beaver, but are differently situated, being
about twice as far from the incisors as they are from the anterior grinders, while in the
Beaver they are placed about midway between these two points; and further, in the
fossil the maxillo-premaxillary sutures divide these foramina at about their middle into
anterior and posterior moieties, the maxillze and premaxille taking about equal shares
in the formation of the apertures; in the Beaver, on the other hand, these sutures are
quite at the hinder part of the foramina, which are thus almost wholly formed by the
premaxille.

The palatal aspect of the maxille shows very clearly the large size of the processes
below the suborbital foramina (sof.), as well as two deep depressions at their bases,

168 MR. E. T. NEWTON ON A SKULL OF TROGONTHERIUM CUVIERI

just in front of the anterior grinders. These depressions are not present in the Beaver.
The two rows of cheek-teeth diverge more rapidly than in the Beaver: the anterior
pair of teeth being nearer together, and the hinder ones further apart.

The palatine bones end anteriorly in a more broadly truncated extremity than do
those of the Beaver; but in both skulls there is a similar pair of foramina, placed one
at each angle of this truncation. Posteriorly the palatines are somewhat broken ;
but at the hindermost angle of each may be seen a large pit (or foramen filled with
matrix) placed just on the inner side of the last molar. This pit is not present in
the Beaver.

The notch of the hinder palatal margin extends forwards to a little within the hinder
extremities of the last molars; it is narrower than in the Beaver, and there is no median
process. In the Beaver the palatine notch is altogether behind the last molars, and it
has a median process. The pterygoids seem to be entirely broken away.

The under surface of the basioccipital bone is characterized by a median longitudinal
ridge with an oval depression on each side. Posteriorly the ridge widens and then
dividing joins the two occipital condyles; these, however, are separated by a distinct
median groove. Anteriorly the ridge is lost in the wide groove which occupies the
front part of the basioccipital. In the Beaver the whole under surface of the basi-
occipital is occupied by a deep round fossa. Only a small part of the basisphenoid can
be seen in the fossil, this region being obscured by the hard sandy matrix.

The tympanic bull (¢).) are not so rounded and inflated as they are in the Beaver,
and the ridge which runs up the external auditory meatus is not so strongly developed.
The mastoid portion of the periotic forms a well-marked prominence upon the side of
the occipital region (fig. 5), and is separated from the bony auditory meatus by a down-
ward process of the squamosal similar to, but not precisely like, that of the Beaver.

The foramen magnum (fig. 4) is triangular, its height and width being about the
same ; while in the Beaver it is rather more rounded and is proportionally higher.
The exoccipitals form lateral plates, which, extending downwards on each side, become
paroccipital processes; these are now denuded, but when perfect could not have been —
so large as those of the Beaver. The exoccipital plate also extends upwards and
outwards, covering much more of the mastoid than it does in the Beaver; and it may
be that it extends to the line seen near the outer margin of the occiput, thus almost
excluding the mastoid from the back of the skull; it is not quite certain, however,
that this line is really a suture, and possibly the exoccipital only extends as far as the
large foramina seen in this region. Each exoccipital is marked by a deep depression,
just above the occipital condyle. ‘The general contour of the back of the skull is
unlike that of the Beaver: for, even when allowance is made for the absence of the
occipital crest, the occiput is found to be much wider, the paroccipitals do not project
to the same extent, and the supraoccipital bone is certainly wider.

Dentition.—Vhe incisor teeth are absent, but the size and roundness of the alveoli

FROM THE FOREST BED OF EAST RUNTON. 169

leave no doubt that the teeth they supported were much larger and rounder than those
of the Beaver, even if this fact were not known from other specimens.

All the upper cheek-teeth are preserved (figs. 2, 3), and, notwithstanding the chipping
and rounding of the enamel edges, their patterns may be distinguished (fig. 5). ‘The
grinding-surface of each tooth is of a triangular form. Although the teeth have not
been much worn by use, it is only the single internal enamel fold of some of them
which remains open on the inner side; the external enamel folds having all become
isolated, and it is only from other specimens that we know they were originally open
to the exterior of the teeth.

The anterior tooth (pm. 4), on each side, is distinctly larger than any of the others;
it has one inner and three outer enamel folds. The second tooth (m. 1) has one inner
and two outer folds, the antero-external fold being very small. No doubt there was
originally a small third outer fold, as in the next tooth, which may have been worn
away during life, or has perhaps been broken away since it became fossilized. The
third tooth (m. 2) has one internal and three external folds, the anterior and posterior
external folds being very small. ‘The fourth or hindermost tooth (m. 5) is about the
same width as the one next in front of it, but is longer from before backwards; it has
one internal and four external folds. In the Beaver the anterior upper grinder (pm. 4)
is not so much larger than the second tooth (m. 1) as it is in the fossil, and the entire
series gradually decreases in size from before backwards; while in this Trogontherium
the last tooth (m. 3) is larger than either of the two (m. 1, m. 2) next preceding it.
In the Beaver both the internal and external enamel folds of all the grinders are open
to the outer surface down to the bases of the teeth, so that they always appear as folds
ou the grinding-surface, while in Trogontherium they very early become isolated.

Summary of the Distinctive Characters of the Forest Bed Skull.—The chief points of
difference between this skull and that of the Beaver may be thus summarized :—

The premaxillary region is wider and stronger. ‘The frontals are longer, the parietals
are shorter, and the postorbital processes are further back.

The anterior jugal buttress descends only about halfway down the maxilla, and not
to the alveolar margin as it does in the Beaver.

The palatine foramina are much nearer to the front grinders than to the incisors,
and they are formed about equally by the maxille and premaxille; in the Beaver
they are midway between the incisors and grinders, and are almost wholly formed by
the premaxille.

The posterior palatine notch is further forwards than in the Beaver, and the pits in
the hinder extremities of the palatines are not seen in the Beaver.

The under surface of the basioccipital has a median ridge with a fossa on each side
of it; in the Beaver the basioccipital is entirely occupied by one large and deep fossa.

The tympanic bulle are less inflated; and the foramen magnum is more triangular
than in the Beaver.

170 MR. E. T. NEWTON ON A SKULL OF TROGONTHERIUM CUVIERI

3. Dental Characters of the Trogontherium of the Forest Bed.

In order rightly to compare the dentition of the skull above described, it is necessary
to bear in mind the changes which the patterns of the teeth undergo during wear, as
shown by a study of the many teeth which have been obtained from the Forest Bed. It
will be desirable, therefore, to give a brief account of the entire dentition as at present
known.

The incisors of both upper and Jower jaws are larger than in the Beaver, the
enamelled fronts are much more convex, and the enamel itself is rough and not smooth.
The upper incisors are nearly circular in transverse section, and seem to have been
thicker than the lower ones, although the entire tooth forms a segment of a smaller
circle. The lower incisors are elliptical in transverse section, and, in some eases at
least, are worn to a point (or nearly so) and not to a chisel edge like the flat-fronted
incisors of the Beaver.

The cheek-teeth are all characterized by having the enamel folds connected with the
exterior of the tooth for only a short distance from the summit, and consequently at an
early stage of wear the folds become isolated.

Upper pm. 4.—This is the largest of the upper cheek-teeth ; it has one inner and
three outer folds, and all the examples of this tooth I have yet seen retain these four
areas of enamel.

Upper m. 1 and m. 2.—These teeth have each one inner and three outer folds (m. 2,
fig. 5); but the posterior outer fold is soon isolated and then worn out, reducing the
folds to three, as in m. | (fig. 5) of the skull above described; and then the anterior
outer enamel area is lost, leaving only two folds, as in the maxilla described by
Sir R. Owen (doc. cit.) (m. 1 & 2, fig. 6).

Upper m. 3 has one inner and four outer folds, as shown in the skull above described
(m. 3, fig. 5); but the anterior outer fold is small and soon lost, as exemplified in the
maxilla described by Sir R. Owen (m. 3, fig. 6). This tooth is longer from before
backwards than either of the other two molars, but its width is about the same.

Lower pm. 4.—This is the largest of the lower cheek-teeth; it has one outer and
three inner enamel folds; sometimes there is an additional, very small enamel area at
the front of the tooth.

Lower m. 1 is the smallest of the lower true molars; it has one outer and three inner
folds, but these may be reduced by wear to two, as in some of the upper molars.

Lower m. 2 is a little larger than m. 1, but not so large as m. 5; it has four folds,
the same as in m. 1, and probably by wear they may be reduced to two.

Lower m. 3.—Yhis tooth is perhaps a little larger than m. 2, but not nearly so large
as pm. 4. ‘There are four folds as in the other true molars, and these doubtless become
reduced in number by wear, but definite evidence on this point is wanting. In one
example some of the folds of molars 2 and 5 have become broken up into small islands.

A miik-tooth preceding the lower pm. 4 has been found in two or three instances.

FROM THE FOREST BED OF EAST RUNTON. ell

4. Comparison of the Forest Bed Skull with Fischer's Trogontherium cuvieri.

It will be obvious, from the description of the Forest Bed skull given above, that it
differs widely from that of the Beaver, and confirms the view held by Sir R. Owen and
other writers, as to the Forest Bed rodent belonging to a genus distinct from Castor. It
now remains to be seen whether we are also justified in referring it to the Trogontherium
of M. Gotthelf Fischer, and finally to ascertain whether Conodontes boisvillettii is, or is
not, to be included in the same species.

The skull from the Forest Bed agrees so closely with that of Trogontherium cuviert,
so far as one can judge from a comparison with the published figures of Fischer's type
skull and with a cast of the original specimen, kindly sent me from Moscow by the late
Dr. Kowalewsky, that our purpose will be best served by pointing out the differences
between them, rather than by alluding to their many points of resemblance.

Fischer’s type is certainly larger than the Forest Bed skull, a difference especially
noticeable in the premaxillary region; but this cannot be looked upon as of much
importance, and certainly not as an indication of specific distinction; for some of the
fragments from the Forest Bed, and more particularly a pair of premaxillary bones in
Mr. A. Savin’s collection, must have belonged to larger skulls than Fischer’s type, and
it is tolerably certain that the premaxille of this Pliocene rodent became elongated in
proportion to the gradually increasing size of the incisor teeth.

The cheek-teeth of Fischer’s type (fig. 7) are a little larger than those of this English
specimen (fig. 5), though not so large as some teeth found in the Forest Bed; the
patterns of the teeth, however, are essentially the same. ‘The first and second molars
in the maxilla described by Sir R. Owen (fig. 6), and alluded to above, having only two
folds instead of four, it was thought impossible for it to belong to Fischer’s genus; but
we now know that the loss of folds is due to the stage of wearing of the teeth. ‘The
Forest Bed skull forms an interesting link in the chain of evidence, for while the first
true molar (m. 1, fig. 5) has only three folds (that is, one more than in Owen’s specimen,
m. |, fig. 6), the second true molar (m. 2, fig. 5) still retains the four folds as in Fischer's
type (m. 1 & 2, fig. 7). Hitherto no English example of the upper third true molar
has been available for comparison; but the present specimen has both these teeth
preserved (fig. 2), and the folds are essentially the same as in Fischer's type—that is,
one inner and four outer folds. It is true that in Fischer’s specimen the hindermost
fold is represented by two small islands of enamel; but this 1 regard as an individual
peculiarity, having seen a similar tendency in teeth from the Forest Bed.

Much of the type specimen is obscured by the sandy matrix adhering to it, so that
the sutures and other details of structure cannot be compared. Nearly all the
minor differences which are seen on a close comparison seem to me to be due either to
this obscuring by matrix, or to the different condition of wearing. ‘There is, however,
one point about which I am in some doubt; in the description of Fischer's type
attention is called to a deep fossa, situated on the base of the skull between the

—
=I
bo

MR. E. T. NEWTON ON A SKULL OF TROGONTHERIUM CUVIERI

tympanic bulle. The cast of the specimen has a deep hole in this region, which has
evidently been made by some instrument, and it seems very doubtful whether it
properly represents a natural fossa of the specimen. If there is a single fossa in the
basioccipital it will be an important difference between this skull and the English one ;
but the basal region is much hidden by tie matrix, and it seems highly probable that
bone has been drilled away; for it is very unlikely that two skulls agreeing so closely
in other particulars should differ so materially in this one point, and I am constrained
to regard this peculiarity as due to accident. It may be that a too great confidence in
the similarity between this skull and that of the Beaver led to an injudicious clearing
away of the matrix, which resulted in the removal of some of the bone at the same
time. The large fossa in the basioccipital of the Beaver is quite unlike that in the
cast, aud the skull itself makes no nearer approach to Fischer’s specimen than it does to
that from the Forest Bed. Paying due regard to all the circumstances, it seems to me
that, until this fossa in Fischer’s type can be proved to be a natural one, Sir R. Owen’s
identification must be held to be correct, and these Forest Bed rodents regarded as
specifically identical with Trogontherium cuvieri.

5. Comparison of the Forest Bed Skull with that of Conodontes boisvillettii.

The type specimen of Conodontes hoisvillettii was found in the Pliocene deposit at
Saint-Prest, and is preserved in the Ecole des Mines, Paris. This skull has been
figured and described by Gervais; it has the three true molars in place, but wants the
anterior cheek-tooth (pm. 4) as well as the premaxillary bones. Some carefully
prepared casts of the surfaces of the teeth (fig. 8), which were kindly sent to me by
Prof. Daubrée some years since, serve well for comparison, and show the teeth to be
somewhat smaller than the figures of them given by Gervais; it seems, therefore,
that the entire figures given by the last-named author represent the skull a little
larger than its natural size, which would thus appear to be as nearly as possible the
same as the Forest Bed specimen.

The hindermost tooth of C. boisvillettii (m. 3, fig. 8) has four folds of enamel, all of
which are isolated from the exterior of the tooth, and differs from the corresponding
tooth of the Forest Bed skull (m. 3, fig. 5) only in the absence of the small anterior
outer fold, which has doubtless been lost by wear. ‘The first and second molars have
only two folds each, thus differing from the Forest Bed specimen, in which these teeth
have three and four folds respectively ; this difference, however, is due to the teeth of
the latter being somewhat less worn, the Saint-Prest specimen having lost by wear
the anterior and posterior outer folds; indeed the three teeth of this specimen agree
precisely with those in the maxilla already alluded to (fig. 6).

The palate presents the closest resemblance to that of the Forest Bed skull; there are
the same anterior grooves, just between the alveoli for the premolars, which run back-
wards about halfway along the palate ; each palatine bone has a similar pit, or foramen,

9
v

FROM THE FOREST BED OF EAST RUNTON. U7
at its hinder and outermost corner, the posterior palatine notch is likewise without a
median process; the tympanic bulle have the same form; and the basioccipital has a
median ridge.

When viewed from above, the median crest of the Saint-Prest skull is seen to be
narrower than that in the Forest Bed specimen, and to extend further forwards before
dividing and passing outwards to the postorbital processes; this, however, is a difference
due to age and the development of the masseter muscles. In other respects the two
skulls are remarkably alike; the interparietal is perhaps a little shorter in the Saint-
Prest specimen; but the general form and proportions, as well as the outlines of the
bones, agree as closely as could be expected in two skulls of the same species. The
parieto-squamosal suture has the same deeply sinuous form, and at its hinder part, in
both skulls, there isa foramen; there is also a large foramen on the outside of each
squamosal between the auditory meatus and the jugal buttress.

The backs of these skulls cannot well be compared, as the Forest Bed specimen is
somewhat denuded of its prominent parts, and the flatness of this region, in the figures
of the Saint-Prest skull, seems to indicate considerable obscurity in the original.
However, the general form is the same in both, and the depressions just above the
condyles are similar. The foramina seen on each side in the mastoid region are not
the same in the two specimens; but as neither of the skulls has the two sides alike,
this cannot be of very great importance, and the differences may be entirely due
to imperfections, and not to any structural difference between the specimens.

In a side view the two skulls also agree, in so far as they can be compared, and
perhaps the most important feature is their similarity in the non-extension of the
maxillary buttress downwards to the alveolar margin.

6. Measurements.
Beaver Forest Bed Fischer's |
from Fens. skull. type.
Basioccipital between condyles to point between incisors .........- m. °135 m. °133 m.°158
fe Toronerofethoshindersm Olanseemeae hare eiecneesacieie tenet “0475 “042 | <046)2
Series of cheek-teeth at alveolar margin ..........-.....-++e000- 034 038 044
Front cheek-tooth to point between incisors ............++++-05- 057 061 ‘075
as a pp. WO ollie) Mona, codons boo ooey novo Ge GdoounON “O19 -015 “016
Ralatalgtonameniyleng thie eencet tense ac levana re eaiate rel seul on seen trea ‘019 ‘021 021
ss P| EODITICLSOTS teeter rahe bean ean cles sor Rene gaits sib atten Men nese cele “023 -030 “O44
Distancesberweenkpremolarsrstlaeiys helene) slater) eter 0075 004 0065
a PA hindenmostam Olarsien emer ia eae rare tek: 022 “026 “027
Brain-case, width at temporal fossa ........... ce secrete eee ee 048 050 “050
oie a) WICC ONE, Goounooevoaoosde boob oUGo ap DODT } +0295 -03385 ‘034?
Pera, Wen: soon sacaco oo beeodonoUabooM dan doDedD ee DUUCOD 070 “056 |
e Wild thmotreac hve ied many ear wem ares nin Las. cu cucya tar eentar alone) ee siiatshei ¢ “021 021 |
IRoHEML, Ieayadl COUN, GoouarcodceougabeadocdoddoccuagoKbDKd “054 064 3005
39 WiAdthwot bochpumimronty OM OGM mi deneere sche sie: teieedencieelietctenenen: “O49 052 “054?
Premaxillee, width of both in front of maxilla.............+...--- ‘037 043
is J DEL MOKOE HIG 5 Goocodboobooondogoooo DOO 027/ “038
Height of skull, from palate between premolars to hinder end of} | :
MES os coe ob COeasd Gee UMS DION OM KD MoO onda ODN OA GUOe.co OP | “052 “055 | -059
2¢

VOL. XItI.—PART Iv. No. 2.—April, 1892.

174 MR. E. T. NEWTON ON A SKULL OF TROGONTHERIUM CUVIERI

7. Conclusions.

To summarize in a few words the results of the above examination and comparison
of the skull from the Forest Bed of East Runton :—In the first place, there can be no
question that, although presenting many points of resemblance to the Beaver’s skull,
the differences are nevertheless of greater importance, and indicate at least a generic
distinction.

In the second place, this skull, in all its essential characters, agrees so closely with
the type of Trogontherium cuvieri, described by M. Fischer, that there are no sufficient
erounds for regarding them as representatives of more than one species; the fossa in
the basioccipital bone of Fischer’s type having been, as I believe, artificially produced.

And thirdly, the skull from Saint-Prest, named Conodontes boisvillettu by M. Laugel,
also agrees so well with Fischer’s type and with the Forest Bed skull that it must be
included with them in the same species.

The earliest name given to this rodent was Trogontheriwm cuviert; it does not appear,
however, in M. Fischer’s first description of the skull, but seems to have been used by
him in a letter to Cuvier, who gave it in the heading of his article in the ‘Ossemens
Fossiles,’ but did not adopt it for the fossil. It was not until the year 1846 that
Sir R. Owen, in his ‘ British Fossil Mammals and Birds,’ used this name for the English
Forest Bed specimens; but fortunately the other generic and specific names proposed
for these fossils are of subsequent date, and cannot therefore be used.

Bones of other parts of the skeleton, which, for cogent reasons, are believed to belong
to this rodent, have been found in the Cromer Forest Bed, and were described by
Sir R. Owen in the ‘ Geological Magazine’ for 1869, and by the present writer, in the
‘Vertebrata of the Forest Bed, in 1882; no important additional specimens have
been obtained until the finding of the skull which has formed the subject of the
present communication.

8. Bibliography.

1809. Fiscner pe Watpueim, Gorrnetr. Sur I’ Elasmotherium et le Trogontherium. Moscow Soc.
Nat., Mém. vol. 11. p. 250.

1824-25. Cuvier, G. Recherches sur les Ossemens Fossiles. dit. 3, vol. v, p. 59.

1833-34. Scummriine, P. C. Recherches sur les Ossemens Fossiles découverts dans les Cavernes
de la Province de Liége.

1840. Lyrrt, C. On the Boulder Formation or Drift and the associated Freshwater Deposits
composing the Mud-cliffs of Eastern Norfolk. Phil. Mag. ser. 3, vol. xvi, p. 345.

1846. Owen, R. British Fossil Mammals and Birds, p. 184,

1847. Rouruer, C. Jubileum semisecularem Doctoris Gotthelf Fischer de Waldheim, celebrant
Sodales Scc. czs. nat. scrut. mosquensis &e.

1848. Pome, A. Diabroticus schmerlingi. Biblio. Univ. Genéve, Arehiv. Sci. vol. ix, p. 167.

1862. Laveen, A. Conodontes boisvillettii. Bull. Soc. Géol. France, sér. 2, vol. xix, p. 709.

1864. Larrer, HE. Comptes Rendus, vol. 58, p. 1201.

FROM THE FOREST BED OF EAST RUNTON, 175

1867. Larret, E. Comptes Rendus, vol. 64, p. 48.

1857-69. Gervais, Paut. Zool. Pal. Gén. p. 80.

1869. Owen, R. On the Distinction between Castor and Trogontherium. Geol. Mag. vol. vi. p. 49-

1876. Aston, E. R. On the Classification of the order Glires. Proc. Zool. Soc. 1876, p. 78.

1882. Newton, E.T. Vertebrata of the Forest Bed, p.65. Mem. Geol. Surv. United Kingdom.

1890. ——. Onsome New Mammals from the Red and Norwich Crags. Quart. Journ. Geol.
Soe. vol. xlvi, p. 447.

1891. Vertebrata of the Pliocene Deposits of Britain. Mem. Geol. Surv. United Kingdom.
9. EXPLANATION OF THE PLATE.

Trogontherium cuvieri, Fischer. All figures natural size.

bo., basioccipital. or., orbit.

€0., exoccipital. pa., parietal.

exa., external auditory meatus. | pl., palatine.

fr., frontal. plf., palatine foramen.

im., incisor alveolus. | pm. #, premolar 4.

ip., interparietal. | pamx., premaxilla,

m. 7, 2, 2, molar teeth. s0., Supraoccipital.

mt., mastoid portion of periotic. sof., suborbital foramen.

mx., maxilla. sq-, Squamosal.

oc., occipital condyle. tb., tympanic bulla.

Fic. 1. Skull from the Forest Bed of East Runton, near Cromer, in the possession of
Mr. A. Savin, seen from above.

Fig. 2. Same specimen, palatal view.

JEN Bs os as left side.
eee 5 3 seen from behind; the dotted line indicates the probable extent

of the occipital crest.
Fig. 5. Right cheek-teeth of same specimen, grinding-surfaces slightly restored to show
the enamel folds.
Fig. 6. Grinding-surface of three molars in the right maxilla described by Sir R. Owen
in 1869, and now in the King Collection, Museum of Practical Geology.
. 7. Grinding-surface of the right cheek-teeth of Fischer’s type skull from the Sea
of Azof, preserved in the Museum at Moscow. Copied from Rouillier’s

i
es

(6 [e)
=|

figure, and reduced to natural size.

Fig. 8. Grinding-surface of the right molars of the type skull of Conodontes borsvillettii,
from the Pliocene of Saint-Prest, preserved in the Ecole des Mines, Paris.
Drawn from wax impressions of the teeth kindly supplied by Prof. Daubree.

a

\

WG
1/2

Ali

Hanhart imp.

Lp.

Berjeau & Highley del. et lith

TROGONTAERIUM CUVIEET

imei

Ba rea ice

VII. Contributions to the Anatomy of the Anthropoid Apes. By Frank E. Bepparp,
M.A., Prosector to the Society, and Lecturer on Biology at Guy’s Hospital,

Received February 15th, 1892, read February 16th, 1892.

[Puates XX.—X XVIII. ]

TABLE OF ConTENTs. Page

Meat trodu cboryins sictaseras cine ae cite sca oeererentenor te rorereetaueia cic celscalensteta stone econ eters 177
II. External Characters and Anatomy of Troglodytes caluus........00.ee eee ee eeee 178
ITT. On the Orang reputed to be Simia morio. .... 00. eevee e cece n esac evs seee neve 201
1iVaeexplanationyofthesblates sy irse cis ciety sre eielaetere eich iia etersieivelelares 217

I. Iyrropuctory.

DURING the last year the Society has lost two of the most valuable and interesting
Anthropoid Apes from its collection—viz., the Bald-headed Chimpanzee “ Sally” and
the Lesser Orang “George.” As neither of these forms has been investigated anato-
mically, except as regards the skeleton, I desire to offer to the Society some account
of their structure as an addition to the existing knowledge of the Anthropoid Apes.
Indeed, these are about the only two forms of the structure of the soft parts of which
we have at present absolutely no knowledge. Unfortunately both animals were in
certain respects in an unfavourable condition for dissection. The viscera of the
Chimpanzee were very greatly diseased, while the enormous quantity of fat deposited
round the abdominal viscera of the Orang rendered their condition if possible still more
unfavourable for examination.

The Chimpanzee “ Sally,” as is well known, lived for a longer time in the Society’s
Gardens than has been recorded of any other Anthropoid Ape. She was purchased in
October 1883, and died in August 1891, having thus been with us for eight years and
some months. For notices of the character and intelligence of this animal, the reader
is referred to papers by Mr. A. D. Bartlett’ and Mr. Romanes*. Immediately
after the death of the animal Mr. Smit made some careful drawings of the hands and
feet and other external characters, which I now exhibit (Plates XX.—XXII.). A figure
used in illustration of Mr. Bartlett’s paper upon the Ape shows the general aspect and

“On a Female Chimpanzee now living in the Society’s Gardens,” P. Z. 8. 1885, pp. 673-675, pl. xli.
The plate is reproduced in the present paper.
> «On the Mental Faculties of the Bald Chimpanzee (Anthropopithecus calvus),” P, Z.S. 1889, pp. 316-321.

VOL. XII.—PaRT v. No. 1.—February, 1893. 2D

178 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

coloration of “Sally” just after her arrival. ‘The brain was carefully extracted and
preserved in alcohol, and the muscles, after being soaked in alcohol, were kept in
spirit vapour before being dissected. On opening the abdominal cavity the contained
viscera were found to be extensively diseased, so much so that I have not attempted
any description of them: so far as I could see, there were no differences from the
Common Chimpanzee ; I thought it would be worse than useless to give any accurate
measurements and description. In the liver, for instance, there may be differences in
the proportions of the lobes from the Chimpanzee; but, as the organ was greatly
enlarged and had undergone fatty degeneration, it would obviously be misleading to
give any measurements; its general form was, so far as I could judge, in no way
different from that of the liver of the Chimpanzee as described by Prof. Flower in his
published lectures upon the organs of digestion in the mammalian series*. Peritonitis
was so extensive that the intestines were greatly matted together. The lungs also
were diseased and were adherent to the pleural cavity. The bones of the vertebral
column were diseased, being very friable. The muscles were somewhat wasted and
there was no fat upon the body. ‘The skin was preserved, and has been stuffed by
Mr. Gerrard, Jun. ; it has been purchased by the Hon. Walter Rothschild for the
‘Tring Museum.

II. Externan CHARACTERS AND ANATOMY OF TROGLODYTES CALVUS.

When the Ape first arrived at the Society's Gardens she was, as I infer from the
teeth characters, about two years old; she was purchased from Mr. Cross, of Liverpool,
in 1883. Two years after she had been in the Gardens, a notice of the principal
external characters and of the habits by Mr. Bartlett appeared in the ‘ Proceedings’ ’,
illustrated by a coloured plate of the head and shoulders. ‘The principal features to
which Mr. Bartlett called attention as distinguishing this animal from 7roglodytes niger
were more evident at the time of her death. The face is quite black, as is also the hair
covering the body. ‘The hair on the top of the head only extends for a short distance
in front of the level of the ears; on the sides of the face there is a scanty growth on the
temporal region and on the cheeks ; the chin is sparsely covered with short hairs, many
of which are white, and there is also a still more sparse growth of short, chiefly white,
hairs upon the upper lip. From the prominent supraciliary ridges spring the long
scattered black hairs of the eyebrows, which do not meet in the middle line.

§1. The Ear.
The accompanying drawing (Plate XXVIII. fig. 3) shows the right ear of the natural
size ; the drawing was made immediately after the animal’s death. It may be compared

' ¢Medical Times and Gazette’ for 1872.
* * On a Female Chimpanzee now living in the Society’s Gardens,” P. Z. 8. 1885, p. 673.

ANATOMY OF THE ANTHROPOID APES. 179

with Hartmann’s figure of the ear of the Common Chimpanzee! and with Gratiolet’s
and Alix’s figures of the external ear of their supposed species 7’. aubryi. ‘The ear is
relatively quite as large as in these Chimpanzees, nor does it present any marked
features which could distinguish it. This organ, however, varies in the Common Chim-
panzee, even if we allow 7’. aubryi to be a distinct species. The margin of the ear of
T. calvus agrees with that of J. aubryi in not being folded for the greater part of its
length. Both the tragus and the antitragus, particularly the former, are very much
marked, and they are divided by a very deep fossa, ‘The margin of the ear is naked.
Superiorly the helix is continuous with the antihelix, the fossa of the helix (at the
letter H in the figure) becoming so shallow as to be practically non-existent. The fossa
of the antihelix (F) is well marked, and there is no lobule.

§ 2. The Hand.

The accompanying drawing (Plate X XII. fig. 1) illustrates the lines upon the palmar
surface of the hand, which were carefully drawn immediately after death. ‘The figure
itself does away with the need of any elaborate description ; it may be compared with
the drawing (Plate XXVII. fig. 1) of the palm of the hand of the Orang-Outang. I
have also had the opportunity, through the death of a specimen at the Gardens on
November 16th, of comparing the Bald-headed Chimpanzee with the Common Chim-
panzee. The latter has been carefully described and illustrated by Alix: his figure,
however, evidently gives only the chief lines, those lettered a, 6, c, in the accom-
panying drawing of the right hand of “Sally.” There are no important differences
that I can detect in the two Apes, except that in the Common Chimpanzee there are
two longitudinal lines running from @ to the roots of the fingers, such as occur in Man
and in the Orang. The line ¢, which is not figured by M. Alix, seems to be of import-
ance, as it occurs in both Chimpanzees. There is, no doubt, just as much variation in
the lines on the hands of Apes as in the human hand. I do not, therefore, think it
worth while at present to do much more than direct attention to the drawing. I have
seen in the human hand the lines a and 6 running at right angles to the axis, as in the
Chimpanzee, instead of somewhat obliquely, as in my own hand, for example.

§ 3. The Foot.

The accompanying drawing illustrates the plantar surface of the foot (Plate XXII.
fig. 2). There are two principal cross lines, lettered @ and 6, which appear to corre-
spond to those similarly lettered in the hand, and which by their direction show the
prehensile nature of the foot; so do the strongly marked lines upon the hallux, which
are hardly marked upon the pollex. The principal longitudinal line ¢ is more marked
than in the Orang.

> «Der Gorilla,’ pl. iv.

* « Des lignes papillaires du main et du pied,” Ann. Sci. Nat. t. ix. 1868,
2D2

180 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

M. Du Chaillu’s! description of the external characters of 7. calvus agrees nearly
absolutely with the above description of “Sally.” I should, however, be inclined to
dispute, on the hypothesis that “ Sally” is an individual of that species, M. Du
Chaillu’s statement that the ears, though large, are smaller than in 7. niger. Dr. Gray 2,
it should be remarked, was disinclined to allow the species, and remarked later? that
“the specimen received with the above name in the Museum was in too bad a state to
determine with accuracy if it is distinct from 7. niger. The baldness of the forehead
appeared to be accidental.’’ After this opinion, which certainly erred on the side of
caution, it is curious to find a brief diagnosis of what is perhaps the “ worst species ”
of Chimpanzee ever described, viz. 7’. vel/erosus, immediately following. The N’tschego,
according to Dr. Franquet, has a black face and small ears, thus contrasting with the
Common Chimpanzee, which has a flesh-coloured face and very large ears. He says,
as M. Du Chaillu has remarked, nothing about the baldness, which, if at all marked,
would hardly have escaped his attention. We may therefore assume that there was
no baldness; the smallness of the ears again distinguishes the animal described by
Franquet and by Duvernoy. With regard to the black face, Du Chaillu says that old
Chimpanzees of the common species are black in the face. The name also appears to
have misled Dr. Franquet. According to Du Chaillu, “the natives of the Camma
country call the 7. niger ‘Nschiego’ and the 7. calvus ‘Nschiego M’bouve,’ the latter
meaning something like ‘another tribe of the Nschiego.’ The Mpongwé called the
T. niger ‘ Nschiego,’ or the ‘ N’chiego’ of Dr. Franquet.” Du Chaillu considered that
the N’chiego of Franquet is simply 7. néger, and, as I point out below, there is nothing
in the osteology to contradict this opinion. I quite agree with Geoffroy St.-Hilaire
when he writes 4:—‘‘ Le Troglodytes tschego a été regardé avec raison par tous les
zoologistes comme une espéce au moins trés-douteuse.” Geoffroy St.-Hilaire remarks
also in this paper that a specimen of Franquet’s species was on its way to the Paris
Museum. In 1866° MM. Gratiolet and Alix described, in an elaborate and well-
illustrated memoir, the external characters and anatomy of a species of Chimpanzee
which they called Troglodytes aubryi. ‘The anatomical account given in the paper is
much fuller and more important than the section dealing with the external characters ;
for post-mortem changes had caused the almost complete loss of the hair and of the
skin itself in the palmar and plantar regions. Nevertheless it is possible to institute
comparison between that Chimpanzee and the animal which forms the subject of the
present communication, which will be more valuable inasmuch as both individuals are
females and of about the same size. This T’roglodytes aubryi, which has been con-

* Proc. Boston Nat. Hist. Soc. 1860. 2 P. Z.8. 1861, p. 273.

* «Catalogue of Monkeys, &c.’ 1870, p. 7.

* Arch. d. Mus. t. x. p. 17, footnote.

* «Recherches sur Anatomie du Z'roglodytes aubryt, Chimpanzé d’une espéce nouvelle,” Nouv. Arch. Mus.
@Hist. Nat. t, ii. (1866) 264 pp. 9 pls.

ANATOMY OF THE ANTHROPOID APES. 181

sidered * to belong to the same species as a Chimpanzee (“ Mafuca”) living in the
Dresden Zoological Gardens in 1876, has a black face,a more pronounced prognathism,
and a more massive form than the ordinary Chimpanzee; the hair is black with a red
reflection. In all these characters, except the last, it agrees with the Chimpanzee
dissected by myself. I did not observe any reddish tinge in the hair of “Sally;”
indeed the entirely black hair of this animal is one of the most marked differences
from the Troglodytes niger. As to the prognathism, I have gone into the matter more
fully below in describing the skull. On p. 35 of their memoir, MM. Gratiolet and
Alix describe in detail the pigmentation of the skin, which is not uniformly black. The
external characters mentioned above, of which a necessarily incomplete account is
given, hardly permit of a definite opinion as to the specific distinctions of 7, aubryi or
its identity with 7. calvus. Later in this paper (p. 183) I point out that the skull-
characters, at any rate, are not those of 7. calvus. The Chimpanzee called ‘* Mafuca”
was examined after its death by Dr. A. B. Meyer! and by Dr. Bischoff?. The
coloured sketch of the face shows no resemblances to 7. calvus; the skin is yellowish
brown, except on the nose and round the eyes, and there is no baldness on the top of
the head.

§4. The Skull.

I have carefully compared the skull of ‘Sally ” with a skull of a nearly adult male
Chimpanzee which arrived at the Society’s Gardens on the same day, and died
29th October, 1883. ‘This individual was slightly larger and had all its permanent teeth ;
the canines, however, had evidently only just cut the gums. The teeth of “Sally ” are
in the following condition ? :—In the upper jaw the permanent incisors and bicuspids
are present; the first molar is the only one of the molar series which is in place. The
canines are a long way from their definitive position ; the point of the tooth is fully half
an inch from the rim of the socket; the extremity of the root of the tooth is barely
half an inch from the rim of the orbit. Those teeth, which are not covered by bone,
look as if they had been artificially inserted into excavations of the maxilla, and are very
prominent and affect the contour of the face.‘ The milk-canines are the only repre-
sentatives of the milk-teeth which have not been replaced. The condition of the
teeth, were their possessor a human being, would suggest the age to be between ten
and eleven years. ‘The brain-caps of both apes belonging to the prosector’s stores had
been removed for the purpose of extracting the brain; this showed the greater density
as well as the greater thickness in front of the bones of 7". calvus.

In a preliminary account of Troglodytes aubryi communicated by M. Gratiolet to the

* “ Notizen tiber die Anthropomorphen Affen des Dresdener Museum,” Mitth. Mus. Dresd. Heft ii. p. 223 (1877).

* See a paper immediately following that of which the title has just been quoted.

* Many of them are diseased and defective, but I think that the above account is complete; I made
certain that the bicuspids are the permanent teeth by cutting into the jaw.

182 Mk. F. E. BEDDARD—CONTRIBUTIONS TO. THE

French Academy of Sciences !, the author distinguishes this species by the presence of an
additional cusp upon the last molar of the lower jaw. In the complete paper, on p. 73,
the following statement is made about the 5th molar of the lower jaw :—-* Elle a cing
tubercules, comme les deux dents qui la précedent. Ce caractére sépare le Troglodytes
aubryi du T. niger et du T. tschego, et le rapproche au Gorilla gina.” I quote the
passage in full in order to leave no room for any doubt of a misunderstanding on my part
in case my own statements are questioned. In the lower jaw of Vroglodytes niger
belonging to the individual mentioned above the last molar of the lower jaw, though
smaller than either M.1 or M. 2, has an identical pattern. I have not been able to
compare the corresponding molar of 7. calvus.

A comparison of these two skulls shows a good many points of difference ; but the
series of skulls in the Natural History Museum, comprising animals of various ages and
of both sexes, makes it a matter of considerable difficulty to draw up very definite
specific distinctions. The examination of a larger series would probably increase these
difficulties. On the whole it appeared to me that the prognathism of the face is more
marked in 7’. calvus; the face in profile shows a very much more concave outline, the
interorbital region being vertical or even (in ‘‘ Sally’s”” skull) slightly directed back-
wards, while in 7. niger the line passing from the top of the skull to the extremity of
the upper jaw is almost flat or slightly concave ; in an adult skull from Sierra Leone
the concavity of the profile outline was fully as marked as in 7. calvus. In none of
the specimens in the Natural History Museum was it possible to examine the interior
of the skull.

The inside of the skull of “ Sally ” showed a few points of difference in the two animals.
In 7. calvus the cribriform plate has a distinct crista galli; this ridge is wanting in
T. niger. The transverse groove for the optic chiasma is decidedly deeper in 7. calvus
than in 7’. niger; the foramen lacerum anterius is also more extensive, the outwardly
directed part being wider as well as larger in 7. calvus. ‘The orbits as seen from the
inside of the skull were rather different in the two Chimpanzees. In the Common
Chimpanzee a strong transverse ridge marks the highest part of the convexity formed
by these lines. In 7’. calvus there is no marked ridge, the surfaces sloping gradually,
and the concavity for the lodgment of the anterior part of the temporal lobes of the
brain being in consequence less deep. On the other hand, in the skull of 7. calvus
the petrous bone was produced into a much sharper edge, and the excavations for the
occipital Jobes of the brain were in consequence decidedly deeper. The vomer of
T. niger is bifurcate behind, the diverging plates being attached to the pterygoids. In
T. calvus the hinder part of the vomer is covered by the pterygoids, the two little
plates of bone which cover over the vomer being distinct from the rest of the ptery-
goid. The relative length of the basi-occipital and basi-sphenoid is different in the two

* This is reprinted at the end of the elaborated Memoir already referred to.

ANATOMY OF THE ANTHROPOID APES. 185

species: in 7’. niger the junction of these two bones is situated about a quarter of an
inch behind the anterior end of the petrosal; in 7. calvus the line of division is
farther forwards, nearly on a line with the anterior extremity of the petrosals: this
produces also a difference in the shape of the basisphenoid ; it is a triangular bone in
T. calvus ; in T. niger it ends in a point anteriorly in the same way, but it has parallel
margins from behind the points where the pterygoids lose their connection with it.

If the existence of the crista galli and the grooving for the optic chiasma prove to be
really differential characters, they are particularly interesting from the fact that they
occur in the Gorilla. The nasal bone in the skull of “Sally ” was distinctly ridged in
the middle line ; this character, however, was not found in the two skulls of 7. calvus in
the Natural History Museum nor in any of the skulls of 7. niger. This character is
also of some little interest, as it occurs in the Gorilla, and is indeed mentioned by
Owen as one of the points of distinction between these Anthropoids. In 7. calvus
the spheno-maxillary fissure is very wide and continuous with the groove lodging the
orbital branch of the trigeminus. In 7. niger this groove is not always open behind:
in the male Chimpanzee skull belonging to the Society the groove in question is
connected posteriorly into a tube by the complete closing in of the walls ; and in other
specimens of 7’. niger I found a partial indication of such a closing in. In the male
Chimpanzee just referred to, the spheno-maxillary fissure is remarkably narrow,
instead of presenting the appearance of a deep wide gash. And this apparent differ-
ence between the two species is very clearly marked in the two skulls belonging to
the Society; but it will probably require revising when a large number of skulls are
available.

No doubt it would be possible to institute a more detailed comparison between the
two species; but the above notes contain, I believe, the principal points of difference,
and I would submit that, taken collectively, they fully justify the separation of the two
Chimpanzees. It is now necessary to inquire how far 7. calvus agrees with any other
“ species ” of Chimpanzee.

Prof. Hartmann! does not consider that any case has yet been made out for
dividing up the Chimpanzees, though admitting “that there are not inconsiderable,
and perhaps even specific, varieties from the ordinary type.” He is inclined to allow
provisionally three varieties, viz. the typical form of 7’. niger, Giglioli’s variety (which
may be the same as MM. Gratiolet and Alix’s 7. aubryi), and thirdly Duvernoy’s
T. tschego, probably the same as the Chimpanzee “ Mafuca”’ already referred to. What
claims has 7. aubryi to be regarded as a distinct species? Of the external characters
our knowledge is far too incomplete to allow of an attempt to answer this question.
With regard to the skeleton, I do not find that the skull offers any character that one
can seize upon as being of specific value. ‘The skull of Z. aubryi is figured in four

* « Anthropoid Apes,” Int. Sci. Series.

184 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

aspects !, which enables one to get a very good idea of its principal characters, even
without reference to the careful description in the accompanying letterpress. J udging
from the figures, it seems to me that 7. aubryi does not differ so much from 7. niger
as does 7. calvus. In the text MM. Gratiolet and Alix do not call attention to
any points of difference between their species and the Common Chimpanzee, except as
regards the teeth. I have pointed out, on pp. 182, 185, the main cranial characters of
Troglodytes calvus which seem to distinguish it from Vroglodytes niger: none of these
characters are shown in the illustrations of the cranium of Troglodytes aubryi; in the
regular slope of the face (shown in the lateral view, /. c. fig. 4) and in the regular and
narrow palate the skull of this ape precisely resembles that of the typical Troglodytes
niger.

Another “species” of Chimpanzee was described nearly forty years ago by Duvernoy 2
in a paper dealing with the Anthropoids in general. This Chimpanzee was called
T. tschego, and the whole skeleton is described and for the greater part illustrated.
With regard to the skull, which measures in extreme length 20 cm., and is therefore
larger than that of “Sally,” I cannot make out either from the description or from
the plate (plate vi.) any marked differences from 7. niger. This species having been
identified by some with 7. calvus, I naturally paid attention to the differences indicated
by Duvernoy in order to see how far the characters agreed with those of the animal
described in the present communication. One of the characters upon which Duvernoy
lays some stress is the fusion of the temporal crests in 7. tschego and the separation of
those crests in 7. niger. ‘This supposed difference is indicated in the plate accom-
panying Duvernoy’s memoir. With regard to this matter, one cannot dissent from the
opinion of Geoffroy St.-Hilaire*, who writes: “ Les caractéres ostéologiques que donne,
a lappui, M. Duvernoy, sont-ils véritablement spécifiques? Ne peuvent-ils s’expliquer
par de simples différences de sexe et d’ige?” The skull of Z. tschego showing the
convergence and junction of the said crests is larger (and, having the complete
dentition, therefore older) than the skull of “Sally,” in which these crests do not meet.
Hartmann mentions the single crest in the adult Chimpanzee without suggesting that
this is a specific distinction. He speaks also of the muzzle being enlarged in front, the
greater width of the palate in front, and states that the alveolar border of incisors and
canines forms “un are assez bombé.” I do not think that anyone will be disposed to
lay great value upon these characters, and at all events they do not indicate any par-
ticular resemblance to the skull of 7’. caluus. The plate does seem to show a rather
thicker supraorbital ridge ; but this is not probably by itself a difference of great
importance.

* Loos. cit. pl. ii. figs. 1-4.
* “Des caractéres anatomiques des grands Singes pseudo-anthropomorphes,” Arch. d. Mus. t. viii.
* « Description des Mammiféres nouveaux ou imparfaitement connus,” &c., Arch. d. Mus. t. x.

ANATOMY OF THE ANTHROPOID APES. 185

§ 5. Muscular System.

As Troglodytes calvus is undoubtedly a distinct species, I have thought it worth
while to give such notes upon the muscles as I am able from my dissections of the
limbs. The other muscles I have not touched at all. There is no account of the
myology of this Chimpanzee, unless it be identical with 7. aubryi; but this identifi-
cation, as I have already pointed out, is hardly possible.

§ 6. The Muscular Anatomy of the Fore Limb.

(1) Pectoralis major.—This muscle has a clavicular origin as in Man; it is large and
powerful and undivided at its insertion, which is two and a half inches in length;
anteriorly a tendinous slip reaches the head of the humerus. The muscle where it is
inserted is tendinous on the under surface, but fleshy above ; posteriorly it is inserted
in common with a part of the deltoid, this part of the muscie being fleshy on both
faces.

(2) Pectoralis minor—The stout tendon of insertion of this muscle is upon the
coracoid process, just above the origin of the conjoined diceps and coraco-brachialis.

(3) Subclavius is largely ensheathed by the coraco-clavicular ligament; its fleshy inser-
tion is on to the proximal half of the clavicle.

(4) Coraco-brachialis—The coraco-brachialis arises, in common with the coracoidal
head of the biceps, from the coracoid process; the two muscles become separate at a
distance of three inches from their origin ; its insertion measures two inches in length;
near to its insertion some of the fibres of the triceps and of the brachialis anticus
arise from it.

(5) Biceps is composed of two very distinct portions, of which the scapular half is
rather, but not very markedly, the thicker ; the two parts of the muscle fuse together
about three and a half inches in front of their common insertion; the two halves of
the muscle are fleshy where they join. The tendon of insertion measures about two
inches up to its beginning on the “ coracoid” side of the muscle. It is inserted on to
the radius, the diameter at the insertion being three-fifths of an inch.

(6) Latissimus dorsii—This muscle gradually narrows towards its insertion, but
gets slightly thicker just before it gives off the dorso-epitrochlear; it has a diameter
of 7% inch. The dorso-epitrochlear slip is very large and fleshy; it ends in a tendon
two and a half inches in length, inserted on to the flexor condyle. The insertion of
the latissimus dorsi measures one inch in length; it is completely free from the teres,
and there is no division of the muscle into two parts such as occurs in the Orang.
The tendon of the muscle commences earlier on its ventral side; an inch and a half is
the length of the completely tendinous part.

(7) Trapezius.—The insertion of this muscle is on to the external half of the clavicle
and on to the greater part of the scapular spine.

VOL. X11I.—ParT v. No. 2.—February, 1893.

i)
&

186 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

(8, 9) The insertion of the two rhomboidet is a common insertion. I have found
it impossible to distinguish one from the other; the insertion extends along three
quarters of the base of the scapula, commencing more than an inch and a half on the
dorsal side of the point where the spine reaches the posterior margin.

(10) I could find no rhombotdeus occipitalis.

(11) Levator anguli scapule.—This muscle is unusually well developed and consists
of two distinct parts, which are perfectly separate from each other at their insertion.
The anterior of them is inserted along a line measuring one inch and a quarter,
commencing exactly at the angle of the scapula. The muscle becomes suddenly wider
just at the insertion; at a distance of one inch from the actual insertion it only
measures half an inch in diameter. This insertion is entirely of fleshy fibres. The
second half of the levator is inserted by an almost entirely tendinous attachment to
the next three quarters of an inch of the posterior border; it is mostly covered by the
rhomboideus }.

(12) The Omohyoid is of considerable size ; its insertion nearly fills up the concavity
on the inferior border ; the insertion is slightly tendinous on the lower surface.

(13) Serratus magnus.—The attachment of this muscle is precisely as in the
Orang.

(14) Deltoid.—The muscle is very large. It arises from the outer half of the
clavicle, from a considerable portion of the scapular spine, and from the septa between
itself and the infraspinatus, teres minor, scapular head of triceps, and teres magpor ;
also from the posterior portion of the scapula just in front of the origin of the teres
major. ‘The insertion is chiefly on to a rough triangular area upon the humerus; it is
also connected with the humeral heads of the triceps, and, as already mentioned,
with the insertion of the pectoralis major.

(15): Teres major—This large and fleshy muscle arises along a line measuring two
inches and three quarters. Its origin is chiefly from the scapular border, commencing
immediately behind the insertion of the serratus magnus and from the septum between
itself and the teres minor, deltoid, and subscapularis. ‘The line of insertion on to the
humerus measures one inch and three quarters; it commences just in front of the
insertion of the coraco-brachialis. The insertion is largely muscular, but that this is so
is not obvious on account of it being concealed by a sheet of tendon covering the muscle
on each side; the anterior half-inch or so of the insertion is entirely fleshy, the last
half-inch or so is nearly wholly tendinous.

(16) Zeres minor.—The teres minor arises partly from the lower border of the
scapula and also from the septa between itself and the following muscles, viz., deltoid,
infraspinatus, triceps, and teres major. The insertion of the muscle is partly on to

‘ It is possible that this muscle is really the rhomboideus minor. Not having its origin I cannot be certain
about the point.

ANATOMY OF THE ANTHROPOID APES. 187

the scapular ligament, and partly on to the head of the humerus, but chiefly on to
the greater tuberosity, and just below the insertion of the infraspinatus.

(17) Lnfraspinacius.—This muscle covers over the whole of the infraspinous fossa
It does not, however, arise from the middle of the fossa, only from the spine and from
the lower border, except posteriorly ; it also arises from the septa between itself and the
following muscles: deltoid, teres major, teres minor. It is largely tendinous at its
insertion: the tendon arises first just within the muscle, but subsequently reaches the
lower surface ; it is inserted partly on to the capsular ligament of the humerus, and
partly on to the humerus itself.

(18) Supraspinatus——This muscle, unlike the do eaetnatons arises from the greater
part of the supraspinous fossa; its tendon of attachment first commences within the
substance of the muscle.

(19) Subscapularis—This muscle is covered below (on the free surface) by a strong
aponeurosis, from which its fibres partly arise; this fascia has a specially strong
attachment to the lower angle of the aie The muscle arises from the greater
part of the subscapular fossa.

(20) Triceps.—This muscle has the usual three heads (not reckoning the dorso-
epitrochlear, which has been already described in connection with the latissimus dors?) ;
the scapular head is largely tendinous on the lower surface ; the tendinous aponeurosis
gradually diminishes in extent, and terminates just after the junction of the scapular
and humeral heads of the muscle. The scapular head arises chiefly from the lower
border of the scapula along a line measuring two and a half inches in length; the
origin is partially from the septa between itself and the teres major, teres minor,
and subscapularis. The origin of the outer humeral head commences just below the
head of the humerus; the inner humeral head commences much lower down; anteriorly
some of its fibres arise in common with those of the outer humeral head; further back
the origins of the two heads are quite distinct. Both humeral heads arise not only
from the humerus, but also from the septa between themselves and adjacent muscles.

(21) Brachialis anticus.—The origin of this muscle commences just below the
deltoid.

(22) Supinator radii longus arises just in front of the supinator carpi radialis longior
from the humerus; it also takes origin from the septa between itself and the trzceps,
the brachialis anticus, and the supinator; the line of origin measures two and a half
inches ; it is inserted by a broad flat tendon two and a half inches in length.

(23) Hatensor carpi radialis longior is the most external of the extensor muscles of
the hand; it arises chiefly from the lower part of the external condylar ridge of the
humerus just below the last muscle, and from the external condyle itself in common
with the extensor carpi radialis brevior. The muscle is short, not reaching halfway
down the forearm; the long tendon passing under the tendon of the extensors of the

thumb is inserted on to the inner side of the base of the second metacarpal.
252

188 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

(24) Eatensor carpi radials brevior.—This muscle is larger than the last ; it rises,
in common with the other extensors, from the extensor condyle of the humerus; it
does not become free from the eatensor communis until a point four inches distant from
the humerus ; it becomes entirely tendinous about two inches further on; the tendon is
quite twice the breadth of that of the last extensor ; it is inserted on to the base of the
metacarpal of the third digit.

(25) Hatensor communis digitorum.—This extensor is of course the largest of all; it
arises from the extensor condyle of the humerus, from the septa between itself and
adjacent muscles, and from the fascia covering the deep extensors; it divides into four
separate muscles, which end in tendons at various distances from their insertion; the
longest tendon is that supplying the third digit; each tendon supplies one of the
digits 11.-v.; it is inserted on the first and second phalanges, and spreads out into a
thin layer almost covering these bones.

(26) Hatensor minimi digiti appears to be totally absent, unless it is a part of extensor
communis that is absent; the tendon of that division of the “communis” which
supplies digit v. is separated from the others at the wrist.

(27) Extensor carpi ulnaris is the outermost of the extensors; its origin extends
down the arm to a point beyond the middle; the tendon, which is very strong and
round rather than flat, is inserted on to the outer side of the base of the last
metacarpal.

(28) Supinator radii brevis~—This muscle is distinctly double as in the Orang; the
posterior boundary is marked by the exit of the nerve; the entire muscle is inserted
on to the radius for more than one third of its length; the insertion of the deeper
layer extends nearly an inch below that of the upper layer.

(29) Extensor ossis metacarpi pollicis.—The origin of this muscle is from the radius
and ulna, and from the interosseous ligament; the muscle divides early, and the
two tendons pass down in close contact, and are inserted on to radial carpal and base of
metacarpal respectively.

(30) Havtensor secundi internodii pollicis —This muscle arises from the inner side of
the ulna, from the interosseous membrane, and from the septa between itself and the
extensor tndicis and other adjacent muscles; its long tendon is inserted on to the second
phalanx of the thumb.

(31) Eatensor indicis—This muscle is almost exactly the same size as the last, and
their origins are close together, being partly from the septum dividing them; the
extensor indicis also arises from the upper surface of the ulna, below the origin of the
extensor carpi ulnaris; it is attached to the index only by a strong tendon which joins
that of the branch of the eatensor communis supplying that digit.

(82) Eatensor primi internodii pollicis—This muscle is superficial to the extensor
ossis metacarpi pollicis, from which, however, it is hardly separable ; it arises, with that
muscle, from the radius, from the interosseous membrane, and from the septa between

ANATOMY OF THE ANTHROPOID APES. 189

itself and adjacent muscles; its tendon commences at the wrist, and is therefore rather
shorter than that of the extensor ossis metacarpi pollicis, but almost exactly of the same
thickness; the two tendons run in very close contact; it is inserted on to the base of
the metacarpal of the thumb on the inner side.

(83) Pronator radii teres.—The origin of this muscle is much more distinctly double
than in the Orang ;
two and a half inches in length.

(34) Palmaris longus.—This muscle is very slender, and ends in a tendon about half-
way down the arm ; towards its origin from the flexor condyle the muscle is not at all
distinct. Its tendon is continuous with a vertical tendinous sheet, from which fibres
of the flexor carpi radialis and of the flexor sublimis arise; at the wrist it is con-
tinuous with the very dense and stout palmar fascia.

(85) Flexor carpi radialis.—It arises from the flexor condyle, from the septa between
itself and the pronator radii teres, flexor sublimis, and palmaris longus; the muscle
becomes free about three inches in front of the wrist; its tendon is visible upon the
under surface of the muscle, first of all about halfway down the arm; the muscular
fibres cease only just at the wrist. It is inserted on to the base of the second

its insertion on to the radius is chiefly tendinous, and measures

metacarpal.

(36) Hlexor carpi ulnaris—The muscle arises from the flexor condyle and from a
considerable portion of the ulna, and from the septa between itself and adjacent muscles,
as well as from the fascia covering the arm ; it becomes tendinous only on the ventral
side, and is inserted principally on to the ulnar distal carpal (pisiform); it measures
half an inch in diameter at its insertion.

(37) Flexor sublimis (perforatus) digitorwm—The four muscular bellies which
together make up this muscle are quite separate from each other halfway down the
arm; the undivided muscle arises from the flexor condyle, from the septa between
itself and adjacent muscles, and from the radius ; the part which arises from the radius
mainly belongs to the flexor of the third digit, which has the stoutest tendon, com-
mencing earlier than in the others upon the under surface of the muscle; the actual
tendon itself of the fourth digit is the longest ; that supplying the little finger is very
much more slender than the rest; the tendons supplying digits 11., IV., and v. lie
superficially to the remaining tendon; a muscular slip measuring three and a halt
inches in length arises from a tendinous intersection upon that part of the /levor
sublimis belonging to the little finger, and is inserted on to the deep flexor tendons of
digit 1v. by a short flat tendon of its own.

(38) Flexor longus pollicis.—This muscle is very distinct from the next to be
described ; it arises chiefly from the radius and from the interosseous membrane, but
also from the septum between itself and the flexor carpi radialis; its tendon com-
mences on the under surface of the muscle about halfway down the forearm; the
tendon is free from muscle on a line with the base of the thumb metacarpal; at this

190 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

point it gives off a very slender tendon running to the thumb ; the main tendon of the
muscle supplies the index.

(39) Flexor profundus ( perforans) digitorum is a muscle rather larger than the last;
it arises chiefly from the ulna and from the interosseous membrane ; the three tendons
become separate at the wrist after perforating the tendons of the flewor sublimis ; they
are attached to the terminal digits of the fingers (Nos. 11., Iv., and v.).

(40) There are four Lumbricales: the first arises from the deep flexor tendon of the
index; it gives off a slip to the tendon of the /levor sublinvis belonging to this digit ;
the second muscle arises wholly from the tendon (deep flexor) of digit m.; the third
muscle arises from this tendon and from the next one, that of the fourth digit; the
fourth muscle arises from the two last tendons of the series; each lumbricalis is
inserted on to the extensor tendon of the digit to which it belongs.

§ 7. Lhe Muscular Anatomy of the Hind Limb.

(1) Gluteus maximus arises from the fascia lata, from the ilium itself, and from the
coecyx ; it is inserted partly on to the fascia covering the thigh and partly by a strong
tendon on to the femur just opposite the end of the insertion of the quadratus femoris,
continuously with the gluteus maximus; but arising from the tuber ischii, in common
with the diceps and other muscles which take their origin therein, is a fleshy mass
which is inserted on to the femur continuously with the tendon of the gluteus, and
which also partly fuses with the rectus externus and the femoral head of the biceps. I
cannot find any sharp demarcation between the fibres of this muscle and those of the
gluteus maximus at their insertion, though the posterior part, which has a mainly
muscular insertion, evidently corresponds to the ischio-femoral of the Orang (see p. 211).
The length of the line along which the conjoined muscles are inserted is three and a
half inches.

2) Gluteus medius arises chiefly from the ilium as far down as on a line with the
anterior boundary of the origin of the glutwus minimus ; it also arises from the fascia
lata. The fibres of the muscle rapidly converge towards its insertion ; some way in
front of the insertion a strong tendinous band is developed within the muscle; the
under and upper surfaces of the muscle only become tendinous a short way in front of
its insertion. At itsinsertion (on to the great trochanter) it comes into close connection
with the tendon of the pyriformis.

(3) Gluteus minimus.—This muscle has a fleshy origin of an inch and three quarters
in length from the border of the greater sciatic notch. It becomes partly tendinous on
the upper surface some way in front of its insertion; its insertion is so perfectly
continuous with that of the scansorius that it is impossible to say where one begins
and the other leaves off. The line of insertion of the two muscles together measures
an inch and three quarters.

(4) The scansorius is well developed; it arises chiefly from nearly the whole of the

ANATOMY OF THE ANTHROPOID APES. 191

anterior border between the anterior spine and the glenoid cavity, but also—though to
a very slight extent—from the fascia lata and from the septum between itself and the
gluteus medius. Its insertion behind that of glweus minimus has been already
mentioned. The muscle begins to be tendinous on the upper surface rather more than
an inch away from the actual insertion, at a less distance on the ventral surface; the
actual tendon of insertion measures about half an inch in length.

(5) Pyriformis runs in close proximity to the glutwus medius; it ends in a somewhat
rounded tendon, which is attached to the femur, on to the great trochanter, between
the insertions of the glutwi medius and minimus.

(6) Obturator internus, together with the gemelli (of which the external is the
larger), is inserted into the upper part of the fossa behind the great trochanter.

(7) Obturator externus has an entirely fleshy insertion just behind the lesser
trochanter.

(8) Quadratus femoris is fleshy throughout.

(9) Biceps femoris.—This muscle has two heads; the ischial head arises from the
tuber ischii in common with the semimembranosus and semitendinosus; it forms a
round fleshy belly, which is tendinous on the lower surface near its origin, and becomes
tendinous on the outer surface about halfway along; it ends in a flat tendon, which
widens out, becoming continuous with the fascia covering this part of the leg, and is
also inserted on to the fibula; this upper part inserted on to the fibula remains thicker
than the rest. The femoral head arises from the femur and from the septum between
itself and the vastus externus along a line measuring two inches and three quarters ;
its fibres are partly inserted on to the tendon of the long head, and partly become
continuous with the fascia covering the thigh.

(10) Semimembranosus.—This leaves the head of origin common to itself, the biceps,
and the semitendinosus as a flat tendon one inch and three quarters in length, and one
quarter of an inch in breadth; it forms a fleshy belly, slightly tendinous on the inner
surface at both ends; it is inserted by a short strong tendon on to the inner side of the
tibia below the hgament uniting the tibia and the femur.

(11) Semitendinosus—This muscle is rather larger than the last; it leaves the
common muscular origin, which it shares with the last two muscles, almost simul-
taneously with them, 7.¢., the common head becomes trifurcate. Its tendon of insertion
measures rather over an inch and a half in length; it becomes wider at the actual
insertion, which is close to and below that of the gracilis.

(12) Iliaco-psoas.—The two separate muscles are inserted by a common tendon on
to the lesser trochanter.

(13) Sartorius—This muscle is roundish at its origin, but becomes flat and strap-
shaped before the middle of the thigh; it is more slender than the gracilis. Its
diameter some way in front of the insertion is three fifths of an inch. At its insertion
it becomes much wider. ‘The insertion is half tendinous and half muscular, the

192 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

fo

tendinous insertion being anterior and lying just to the outside of that of the gracilis.
The line of insertion measures one inch and a half.

(14) Rectus femoris —This muscle is partly tendinous and partly muscular at its
origin from the ilium below the scansorius and just in front of the glenoid cavity ;
about halfway down the thigh it fuses with the external vastus muscle.

(15) Vastus.—The vastus externus, vastus internus, and crurc@us were all present ;
but they were so closely connected the one with the other that it is best to regard
them as one muscle arising from the greater part of the surface of the femur, and
inserted on to the patella and its ligament.

(16) Gracilis—The gracilis arises from the symphysis pubis by a thin tendon; the
muscle is flat, and gradually narrows towards its insertion, which is by a tendon rather
more than an inch in length and about half an inch in breadth; the insertion lies
between that of the sartorius and semitendinosus.

(17) Pectineus—This muscle arises in front of the adductor longus, and is very
slightly overlapped by it; its origin is chiefly fleshy, but there is a tendinous surface
of small extent on the posterior surface of the muscle, which is overlapped by the
adductor. The insertion measures three quarters of an inch in length, and lies
immediately behind the lesser trochanter; for almost one quarter of its extent it is
entirely tendinous, and the whole of the upper surface glistens.

(18) Adductor longus arises from the pubis, immediately behind the origin of the
pectineus, which it slightly overlaps, and from the septa between itself aud the
adjacent muscles; at a distance of four and a half inches from its origin it fuses with
the adductor magnus, and is inserted in common with that muscle.

(19) Adductor brevis.—This muscle arises below the adductor magnus ; it consists of
two parts: one arises from the pubis, below the adductor longus, and from the symphysis,
the other arises further back from the ischium and from the septa between itself and
the adductor magnus and obturator internus; the two join just before the insertion ;
its line of insertion measures an inch and a half in length; it commences about on
a level with the middle of the lesser trochanter, and ends on a level with the end
of the insertion of the glutwus maximus; the insertion is partly muscular and partly
tendinous.

(20) Adductor magnus.—This muscle is divided into two distinct parts, which are
perfectly separate both at origin and insertion ; one part is inserted in common with the
adductor longus, and it is a question whether these two together should not be regarded
as the equivalent of the adductor longus. The outer part of the muscle arises from
the ischium just behind the gracilis and in contact with it in front; behind it is in
contact with the origins of the semimembranosus and the other muscles which arise from
the tuber ischii ; together with the gracilis it completely covers over the deeper part of
the adductor magnus. The muscle is flattish at first, but afterwards becomes thicker ;
its origin is fleshy above, but in the middle line below it has a tendon, which soon dies

ANATOMY OF THE ANTHROPOID APES. 193

away. It is inserted on to the inner condyle of the femur. The second part of the
adductor magnus arises below the first head and the gracilis from the symphysis pubis
and from the ischium ; the line of origin extends from close to the anterior extremity
of the symphysis back to a point upon the ischium corresponding to about the middle
of the origin of the first half of the adductor magnus; it arises not only from the bone
but from the septa between itself and the posterior head of the adductor brevis and
obturator internus ; at about an inch from its insertion it is joined by the adductor
longus. 'The insertion of the muscle upon the femur extends from the end of the first
third of the bone up to its very extremity; on the inner side, where the insertion is
common to it and the adductor longus, the muscle is strongly tendinous; on the outer
side the attachment is muscular and formed by coarsely fasciculate fibres.

(21) Tibialis anticus—This muscle is formed of two distinct parts, which, however,
originate in common; the two tendons pass close together below the ligament
connecting the astragalus with the first metatarsal; the hinder of the two tendons,
which is the shortest, is inserted on to the cuneiform bone; the anterior tendon is
inserted, in common with the astragalo-metatarsal ligament, on to the proximal end of
the first metacarpal.

(22) Eatensor proprius hallucis—This muscle arises from the fibula, from the
interosseous ligament, and from the septa between itself and the adjacent muscles; it
becomes entirely tendinous just at the astragalo-metatarsal ligament, underneath which
it passes, and, running along the upperside of the hallux, is inserted on to the distal
phalanx.

(23) Eatensor longus digitorum pedis.—The muscle arises from the fibula, from the
interosseous membrane, from the fascia covering the leg, and from the septa between
itself and the tidialis anticus, extensor proprius hallucis, and peroneus tertius. The
muscle becomes free about two inches in front of the ankle. ‘The tendons of insertion
are visible upon the upper surface of the muscle about an inch before they become free
from muscular fibres. In the middle of the ankle the muscle splits into two: one
division is entirely tendinous, the other is invested on the lower side with muscle for a
distance of about an inch beyond the division. The front tendon supplies the second and
third digits; it divides rather beyond the middle of the metacarpal into two tendons,
of which that supplying digit 1. is rather less than half the diameter of that supplying
digit 11. The second division of the muscle again divides, just at the commencement
of the metatarsal region, into two equally sized tendons, which are each as large as the
tendon of digit 11.; they supply digits tv. and y.; in every case the tendons are inserted
on to the terminal of the phalanx of the digit.

(24) Extensor brevis digitorum consists of four muscular slips, which supply the
thumb and the next three digits; their tendons join the long extensors of the digits
just at the commencement of the first phalanx ; the four muscles arise in common from
the os calcis, but they become almost immediately distinguishable into the four slips ;

VOL. XIII.—PART Vv. No. 3.—ebruary, 1893. 25

194 MR. F, E. BEDDARD—CONTRIBUTIONS TO THE

those supplying the three middle digits lie between the metatarsals, the first of them
lying between the second and third metatarsals ; their tendons arise at the commence-
ment of about the last third of the metatarsals.

(25) Peroneus longus.—This is the outer of the two peroneal muscles (there are
only two); it arises from the fibula and from the septa between itself and adjacent
muscles.

(26) Peroneus brevis arises from the fibula and from the septa between itself and
adjacent muscles ; its tendon, which is superficial to that of the last muscle, is inserted
on to the outer side of the head of the first metatarsal ; some way before its insertion
it gives off a very fine tendon, which passes along the extensor surface of the last meta-
tarsal; at the end of this bone the tendon passes over to the outer side and joins the
tendon of the long extensor of this digit.

(27) Gastrocnemius.—This has the usual two heads, which join at about the middle
of the calf. The outer head is entirely tendinous at its origin from the external
tuberosity; for half an inch or so from the point of origin it is fused with the fleshy
head of the plantaris; the tendon spreads out over the outer surface of the muscular
belly, and gradually dies away. The inner head is half fleshy and half tendinous at its
origin, which is more extensive than that of the outer head, extending as it does over
the condyle ventral of and behind the tuberosity as well as from the tuberosity itself ;
the relative positions of the fleshy and tendinous parts of the head of origin are precisely
those of the fleshy plantaris and the tendinous head of the outer half of the gastro-
cnemius; the fleshy parts are both internal. The tendon spreads over the upper surface
of the muscle and reaches back rather further than in the case of the outer head; the
conjoined muscles become tendinous on the inner surface some distance in front of their
insertion (largely muscular) on to the calcaneum.

(28) The solews may be regarded as a third head of the gastrocnemius, which is,
however, distinct from it nearly up to the point of their common insertion. ‘The muscle
arises from the lower surface of the head of the fibula; its origin is fleshy above;
below there is a layer of tendon, which extends back, covering nearly the whole of the
ventral surface of the muscle, to about the middle of the calf of the leg; further back
still it is the upper surface which becomes tendinous.

(29) Plantaris.—This is a slender muscle, arising, as already mentioned, in common
with the outer head of the gastrocnemius ; the muscle passes into a slender tendon ;
four inches from its crigin this tendon passes between the gastrocnemius and the soleus,
though in closer connection with the-former.

(30) Popliteus arises from the external condyle of the femur by a strong thick tendon ;
this passes just beneath the external ligament of the knee; the muscle is inserted on
to the head and shaft of the tibia along a line measuring three inches in length. The

tendon of origin of the muscle ends in the interior of the muscle, being ensheathed in
muscular fibres.

ANATOMY OF THE ANTHROPOID APES, 195

(31) Flexor longus digitorum pedis.—This muscle arises from the anterior two-thirds
of the shaft of the tibia and from the septa between itself and adjacent muscles. Its
tendon commences a little way in front of the ankle-joint, but it is covered on the
palmar surface with muscle continuously ; the fibres belonging to itself do not end
until the commencement of the /wméricales. It divides into three tendons, which
supply digits 11., IV., V.

(32) Mexor brevis arises from the calcaneum below the abductor hallucis ; at about
the middle of the sole of the foot it gives off a delicate muscular slip which ends in a
very fine tendon joining the flevor perforatus of the fourth digit; beyond this it divides
into two tendons, of which the outer is the stronger of the two; this becomes the
perforatus tendon of digit 11. At the point where the muscle passes into tendon it is
reinforced by a muscular slip from the flexor longus digitorum; it is difficult, therefore,
to say whether the tendon belongs to one muscle or the other. ‘The inner tendon of
the muscle becomes the perforatus tendon of the index, the perforatus tendon being
furnished, not by the fleaor profundus, as in the other digits, but by the flexor longus
digitorum pedis.

33) Flexor profundus digitorum.—The muscle arises from the upper two-thirds of
the shaft of the fibula and from the septa between itself and adjacent muscles ; the
tendon becomes quite free from muscular fibres at the ankle. It gives off a strong
branch to the thumb, which runs to the distal phalanx of the same; half an inch
beyond this it is joined to the tendon of the flewor longus digitorum pedis, just in front
of the origin of the branch of the latter going to the last digit ; beyond this it divides
into two tendons, approximately equal in thickness, which are the perforating tendons
of digits mm. and Iv.

(84) Lumbricales.—There are three lumbricales. The innermost of these arises
from the branch of the tendon of the flexor longus digitorum belonging to the index,
and from the inner of the two tendons of the deep flexor ; the outermost of the three
arises partly from the outer surface of the outer of the two tendons of the deep flexor
and partly from the tendon of the flexor longus digitorum supplying the fifth digit.

(35) Zibialis posticus.—This muscle is the deepest of the three flexor muscles ; it
arises from the tibia, from the fibula, from the interosseous membrane, and from the
septa between itself and the two remaining flexors; its tendon of insertion is as large
as those of either of the two other flexors; it is inserted partly on to the ligaments of
the sole of the foot lying deep of the long flexor tendon and partly on to the tibia.

(36) Abductor hallucis arises from the caleaneum and ligaments of the sole of the
foot, and is inserted on to the base of the first phalanx of the hallux by a fleshy
insertion.

(37) Abductor minimi digiti arises principally from the calcaneum, but also from the
plantar fascia of the foot ; it is inserted by a short tendon on to the base of the first

phalanz.
2F2

196 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

(38) Flexor brevis hallucis arises from the cuboid bone by a short flat tendon ; it is
inserted by a single insertion, in common with the adductor hallucis, which is fleshy.

(39) Adductor hallucis.—This is a strong fleshy muscle which arises partly from the
sheath of the peroneus longus, partly from the plantar fascia, and partly from the heads
of the metatarsals; the muscle measures one inch in diameter at the middle; it is
inserted in common with the fleshy /lewor brevis on to the base of the first metacarpal.

(40) A very small slender muscle arises by a tendon measuring half an inch in
length from the sheath of the peronwus longus; the delicate muscular belly in which
it ends measures 3 mm. in diameter and about an inch in length; it passes across
the transversus pedis in close connection with it, but I could not determine its insertion ;
I found this muscle only in one foot.

(41) Transversus pedis is of some size; it arises from the metatarsal ligament of the
second, third, and fourth digits; it is inserted on to the thumb in common with the
adductor hallucis.

§ 8. Comparison of Musculature with that of the Common Chimpanzee.

In the following table the principal differences between Zroglodytes calvus and
Troglodytes niger are shown (according to Sutton’s account of the myology of the
latter 1) :—

T. calvus. T. niger.
Pect. minor ............ Insertion : coracoid. Insertion: capsule of shoulder-joint.
Biceps cruris ......... Ischial head present. Ischial head absent.
STL. daonacoocnoseasese From head of fibula only. From upper third of posterior surface
of fibula only.
Flex. prof. digit. ...... Attached by a vinculum to No such vinculum® (?).
Flex. long. digit.
Flex. long. digit....... Supplies digits 1., Iv., v. Supplies digits 11., v.
Lumbricales............ Three. Four.
Flex. long. poll. ...... Well-developed; supplies index Absent or feebly developed; supplies
and pollex. only pollex.
Eat. min. digit. ...... Absent. Present.

Macalister * has noted that in the Chimpanzee the dorso-epitrochlear ended in a
fascia in the middle third of the arm; it is thus more rudimentary than in 7roglodytes
caluus. The flexores profundus and pollicis were found by Macalister to be fused into
a single muscle; but the double condition shown in the Ape described in the present

* Journ, Anat. & Phys. vol. xviii. p. 66 (1884).
* This vinculum, however, is stated by Macalister to occur. Sutton does not say it is absent; he does not
refer to it.

* “On some points in the Myology of the Chimpanzee and others of the Primates,” Ann. & Mag. Nat. Hist.
vil. (1871) p. 341.

ANATOMY OF THE ANTHROPOID APES. 197

paper has been recorded by Wilder. The eatensor indicis, which I found to supply
the index only, may in the Common Chimpanzee send a tendon to the middle finger
also. I have referred above to Mr. Sutton’s statement that the fibular head of the
soleus is the only head present, but that it is more extensive than in the present species.
Sir G. Humphry found in a Common Chimpanzee the fibular origin restricted, as in
the case of “Sally,” to the head of the fibula; but in the specimen dissected by him the
tibial head also was present. Dr. Chapman! found no plantaris in the Chimpanzee 2
dissected by him, and no transversus pedis. It is clear from the above very brief and
incomplete notes upon recorded investigations into the myology of the Common
Chimpanzee that it is not easy at present to say exactly what is the normal arrangement
of the muscular structure of that animal; we are evidently not yet in a position to
discriminate between what are variations and what are characteristic arrangements.
If this is the case with the Common Chimpanzee, which has been dissected by so many
anatomists, it is obviously much more the case with the Bald-headed Chimpanzee, of
which, at present, only a single specimen has been dissected. I do not think it, there-
fore, worth while to attempt an exhaustive comparison of its muscles with those of
other Anthropoids ; the value of such a comparison would be very far indeed from
being commensurate with the labour of collecting various papers and abstracting the
necessary data. ‘There is no reason to suppose that when other examples of Troglodytes
calvuus have been dissected they will prove to be identical in every point with the
individual studied by myself. I must therefore leave to further workers the task of
constructing a muscle formula of this Ape for comparison with other Anthropoids.

§ 9. The Palate.

The accompanying drawing (Plate X XV. fig. 2) illustrates the palatal ruge of this
Chimpanzee. It will be observed that the folds upon the hard palate, although fairly
well marked, are irregular in their arrangement and incomplete compared with what
they are in the lower Apes. This holds good for all the Anthropoid Apes so far as
they are known, and for Man. Sir Richard Owen says *, on the other hand, “In the
higher Quadrumana the palate is smooth or unridged as in Man.” The palate is
certainly smoother in Man than in the Chimpanzee, but there are ridges which, how-
ever, are much fewer than in the Ape.

The drawing (Plate XXV. fig. 2) precludes the necessity of an elaborate description

f the palatal ridges of T’roglodytes calvus, which, moreover, possibly have some range
of variation as they have in Man ‘; it is, however (in my opinion), important to illus-

‘ «On the Structure of the Chimpanzee,” Proc. Acad. Nat. Sci. Philadelphia, 1879, p. 52.

* Its absence is also asserted by Bischoff and Briihl, Hartmann, however (‘ Der Gorilla, p. 52), says that
it is normally present.

® Comp. Anat. vol. iii. p. 396.

“ H. Allen, “ The Palatal Rugse in Man,” Proc. Acad. Nat. Sci. Philad. 1888, p. 254.

198 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

trate a point of this sort in an animal which cannot be readily examined by a person
who is treating of the subject of the palatal ruge.

§ 10. The Brain.

I have compared the brain of the Bald-headed Chimpanzee with the brains of two
Common Chimpanzees; the latter, however, naturally presented some differences from
each other, which rendered the task of comparison more difficult.

The brain of “Sally” weighed, after removal of the pia mater and after an immersion
of four months in spirit, 82 0z.; it had been allowed to dry for about an hour and a
half before weighing; it was then damp but not wet.

The two other Chimpanzees’ brains were weighed under the same conditions 1, and
were found to weigh respectively 65 oz. and 62 oz.

These two Chimpanzees were of about the same size, and not much more than half
the size of “Sally.” ‘They were both a little larger (about 4 or 5 inches taller) than the
animal examined by Dr. Symington’, the brain of which weighed under pretty much
the same conditions 84 oz. I cannot account for the very great difference ; it prevents
me from attempting to draw any conclusions as to the weight of the brain in Zvroglo-
dytes calvus.

The brain of Zroglodytes calvus is deeper in proportion to its length than either of
the two brains of Troglodytes niger with which I compared it.

The lengths of the brain as compared with their height are as follows :—

Length. Height.

mm. mInm.
LU OOIOIES GUIS 5 6 os 6 2 5 08 65
9 OWA) 6 8 60 5 OD) 58

eh (GER IB) GS go AOR 628

The measurements are taken by placing vertical plates beside the brain, and are
therefore only true as regards the proportions.

The actual measurements of the brain of Troglodytes caluus are as follows:—
Length 103 mm., breadth 80 mm., height 62 mm.

In viewing the cerebrum from the upper surface (Plate XXIII. fig. 3), the most
striking difference between the two species was the condition of the parieto-occipital
fissure (the “ Simian fissure,” as it has been called). In Man this fissure is of small

* One of them has been eight months, the other ten months in spirit.

* “On the Viscera of a Female Chimpanzee,” Proc. Roy. Phys. Soc. Edinb. vol. x. p. 300.

* In this Chimpanzee the temporo-sphenoidal lobes projected downwards in a much more marked degree
than in the other. The vertical diameter of the brain was naturally increased thereby; 5 or 6 mm. must be
allowed for this difference.

ANATOMY OF THE ANTHROPOID APES. 139

extent; the length is generally (according to Quain’s ‘ Anatomy’) about an inch. In
the Chimpanzee, on the other hand, these fissures on each side are very long; in both
of the specimens which I examined they extended to within one-sixth to half an inch
of the furrow lying between the cerebrum and cerebellum; and as figured by Hart-
mann ! they pass in an almost straight course across the brain, being traceable up to
the median furrow of the brain. In the brain of “Sally,” on the other hand, these
fissures, although recognizable laterally, were connected with the median furrow (on
Plate XXIII. fig. 3, P.o,f.) by an irregular bent fissure on one side only. A closer
comparison of the two brains showed an interesting reason for this difference. In the
Common Chimpanzee the parieto-occipital fissure is deep, and its posterior wall is said
to be markedly convoluted. This was so, at any rate, with one of the two individuals
whose brains are among my stores at the Gardens. If that part of the brain of
Troglodytes calvus lying between the letters P.o.f.? and P.o.f. in fig. 5 of Plate XXIIL.,
the median longitudinal furrow of the brain, were infolded, we should get a very close
resemblance to the brain of Troglodytes niger.

The Sylvian fissure of Troglodytes calvus presents, at any rate, one very interesting
point. It has been more than once pointed out that the posterior and longer of thé
two branches of the Sylvian fissure is more upright in the Chimpanzee than in Man.
This was undoubtedly the case with the two brains examined by myself, the angle of
inclination being pretty much the same in both. In the brain of Zroglodytes calvus,
however (Plate XXIII. fig. 2 i’.s.), the posterior limb of the Sylvian fissure was much
more upright than in the Common Chimpanzee, resembling therefore that of the Gorilla,
the Gibbon, and the Orang, though not so upright as in the latter.

The fissure in the brain of Z’roglodytes calvus, which appears to correspond to the
anterior branch of the Sylvian fissure as figured by Gratiolet?, is quite as large as in the
Common Chimpanzee, but more horizontal in direction. I take it, however, that this is
not the true anterior branch of the Sylvian fissure. For this long fissure, which lies in
front of (below) the true anterior branch of the Sylvian fissure, is not continuous with
the posterior branch of the Sylvian fissure, as can be seen by raising the temporo-
sphenoidal lobe. Comparing the brains of the Chimpanzees with that of the Orang
and the Gibbon, the true anterior branch of the Sylvian fissure (/s.@.) is seen to be
very short. In Zroglodytes calvus it is directed more upwards than in 7. niger.

The jissure of Rolando certainly varies in position in the Common Chimpanzee.
In one of the two specimens which I have before me this fissure is not much behind
the transverse axis of the brain; this is the case also with the brain of Troglodytes
calvus. But in another Common Chimpanzee the point of the V formed by the two
converging furrows is distinctly (more than half an inch) behind the transverse axis.

* ¢Anthropoid Apes,’ fig. 37, p. 192.
* «Mémoire sur les plis cérébraux de ’homme et des primates,’ pl. vi. figs. 2, 6.

200 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

It is evident, therefore, that much stress cannot be laid upon this alleged difference
between Man and the Chimpanzee. The following measurements will show these
differences more clearly :—

Total length of cerebrum Length between anterior

measured from above a end of cerebrum and
vertical plate at each end point of V in fissure
and covering the cerebrum. of Rolando.
mm. mm.
Troglodytesical user. arise mie: 100 54
x (OUEPONN'S Fe Beas ce 98 52
SM MMLETCEACS Mica Ard Aaah 4506 BT

The fissure of Rolando in the brain of the Bald-headed Chimpanzee has the same
two anterior connections that it has in the common species; its shape is that of the
letter W; the innermost convexity is, however, much more marked than in either of
the two Chimpanzees’ brains with which I have compared it. As to the other fissures,
they show such variation in individuals that I do not think a detailed description of
one brain will serve any useful purpose. I therefore direct attention to the accom-
panying drawings (Plate XXIII.) of the brain, which will be of more assistance to
investigation in the future than any description.

The under surface of the brain shows one noteworthy difference between the two
species of Chimpanzee. In the Common Chimpanzee the frontal lobes are, as in the
lower Apes, keeled below in the middle ventral line, looking as if they had been
artificially pressed together by the thumb and forefinger. There was hardly any sign
of this in Troglodytes calvus, and in this particular the brain is more like that of the
Orang.

The interval between the temporo-sphenoidal lobes was very much less in Troglo-
dytes calvus than in the Common Chimpanzee, not very much more than one half.
This is partly due to the position of the lobes in question, and is correlated of course
with the more vertical direction of the Sylvian fissure already referred to. The
apex of the temporo-sphenoidal lobes is decidedly less blunt than in the Common
Chimpanzee.

The posterior aspect of the brain is also characteristic. The lateral masses of the
cerebellum (Plate XXVIII. figs. 1, 2) come into contact behind, overlapping the
median tract, which they largely conceal. This is precisely what occurs in the Orang,
but not in Troglodytes niger. In that Chimpanzee (Plate XXVIII. fig. 2) the median
tract of the cerebellum is not concealed by any overgrowth of the cerebellar hemi-
spheres. The peculiar form of this region of the cerebellum, as compared with that
of the Common Chimpanzee, will be best appreciated by a comparison of figs. 1 and 2
of Plate XXVIII., which represent the posterior aspect of the brain of “ Sally” and
of one of the two Common Chimpanzees belonging to the Prosector’s stores.

ANATOMY OF THE ANTHROPOII APES. 201

§ 11. Conclusions.

In the preceding pages I have incidentally attempted to criticize the various species
of Chimpanzee that have at one time or another been proposed ; but the main object has
been to endeavour to show that the Chimpanzee which lived in the Society's Gardens
from 1883-91 is Du Chaillu’s Troglodytes calvus, is not Duvernoy’s T. tschego or any
other variety, and is a perfectly distinct species of Chimpanzee, which has, however,
hardly a claim to represent a distinct generic type. Troglodytes calvus differs from
LT. niger in well-marked external characters, in less well-marked skull characters, and
apparently in its muscular anatomy and brain. I am not, however, desirous of empha-
sizing too much the myological differences, since we obviously do not know the range
of variation in 7’. calvus, or, for the matter of that, in 7. niger. The skull of 7. calvus
may be said to show an exaggeration of the characters proper to the genus Troglodytes,
from which I exclude the Gorilla.

III. On THE ORANG REPUTED TO BE SIMIA MORIO.

Although perhaps most persons now consider that there is only one species of Orang
Outang, more than one name has been given to supposed different forms. The Sumatran
Orang, for example, has been regarded as distinct from the Bornean ape, and the
smaller Bornean Orang has been distinguished from the larger animal (Sinmia satyrus)
under the name of Simia morio. The small Orang presented to the Society on 15th
April, 1891, which died on September 22nd, was believed to be a representative of the
latter species.

The best figures known to me of the large Orang illustrate a paper by Dr. Hermes
published some fifteen years ago'. In those plates the young and adult Orangs are
shown in several ways. Other figures which I have consulted are Gervais’s °, Chenu’s *,
Wallace’s 4, and Chapman’s®, and that of Flower and Lydekker ®. I do not trouble to
indicate older illustrations, such as those given by Temminck, for most of them show
signs of inaccurate drawing or reproduction, and it would be unprofitable to build any
conclusion upon them. Of the coloured figures referred to, it seems to me that
Dr. Hermes’s are far away the best, and they are the only ones besides that of
Dr. Chapman which give a good lateral view of the head. If the coloured drawing of
the head of ‘‘ George,” which I now exhibit (Plate XXIV.), be compared with Hermes’s

1 Zeitschr. f. Ethnologie, Bd. viii. 1876, pls. xv. & xvi.
* Mammitéres, vol. i. pl. i.

° Encycl. d’Hist. Nat., Quadrumanes, p. 39, fig. 42.

* «The Malay Archipelago,’ vol. i. p. 64.

° «The Structure of the Orang,” Proc. Acad. Nat. Sci. Philad. 1879, pl. xi.

® ¢ Mammals Living and Extinct, p. 733. The cut is from a drawing by Wolf.
VOL. XIU.—PART V. No. 4.—February, 1895. 26

202 MR. F. E. DEDDARD—CONTRIBUTIONS TO THE

illustration of the side view of the head of a young Orang, a number of important
differences at once appear. In the first place, the shape of the head is quite different.
The larger Orang shows the “ brachycephaly ” which is characteristic of the Orang, as
compared with the Chimpanzee or Gorilla, to a very much greater degree than does
the head of the supposed lesser Orang. Indeed, the head of this animal is not unlike
that of a Chimpanzee; its shortness, however, as compared with its breadth becomes
plainer after an inspection of the accompanying drawings (Plates X XI. and XXV. fig. 1),
which represent the vertex of the Orang and the Chimpanzee. But even here the dis-
crepancy is not very great, the measurement being :—

Length of cranium. Breadth of cranium.
Chimpanzee (Troglodytes caluus) . . . 140 mm. 116 mm.
Orans|(S¢mias 07701) eee eS 2mm: 113 mm.

The difference in the height of the forehead is also very marked. It is well known
that this is a character of age, the older Orangs having less lofty foreheads than the
young ; but ‘“‘ George” was probably not more than four or five years old at the time
of his death, inasmuch as none of the permanent teeth had put in an appearance.

Again, the relative baldness of the forehead of the two Apes furnishes a remarkable
difference. In the Orang examined by myself the baldness was more pronounced than
in the animal figured by Dr. Hermes. It will be noticed that the hair on the temples
stops short on a level with the ear, whereas it extends much further forward in the
animal illustrated in Hermes’s figure, and also in the Orang’s head figured by Abel !.
It is true that in Chenu’s figure the hair stops at the same point as it does in the Orang
which forms the subject of the present communication; but the forehead of the animal
figured by Chenu is much higher, and in other respects it agrees with the typical
Orang.

The length of the hair also is greater in the young of the larger Orang. This is
shown in Hermes’s figure and in the woodcut given by Mr. Wallace. But the length
of the hair upon the head seems to be a question of age. It is short in the adult Orang
figured by Hermes.

I do not, however, propose on these characters to establish a species, whether called
Simia morio or by some other name. My object is rather to take the opportunity
afforded me of contributing fresh data to the gradually accumulating material, which
will ultimately permit of a definite opinion upon the question. In the meantime
I submit that the differences between the animal of which a drawing is exhibited
(Plate XXIV.) and the typical Orang figured by Dr. Hermes are quite as marked jas
the differences between, say, a Tartar and an Aryan.

* “Some Account of the Ourang Outang ete.,” Asiat. Research. vol. xv. pl. i.

ANATOMY OF THE ANTHROPOID APES. 203

In this case the differences are not believed to amount to a difference of species; but
if a Mongolian were to come under scientific observation for the first time, the peculiar
characters of the face would be undoubtedly sketched and published for purposes of
reference. I think, therefore, that it is worth while to publish the drawing of the
Orang “‘ George,’ even if it were definitely known to belong to the same species as the
Orang figured by Dr. Hermes, which is far from being proved. At the very least the
two drawings exhibit the extreme range of variation of one species. While describing
the head I may call attention to the scanty eyebrows and to the slight development
of beard upon the chin.

§ 1. The Hand.

The back of the hand is illustrated in the accompanying drawing (Plate XXVI.
fig. 1), and the palm of the hand in another drawing which I also exhibit (Plate XX VII.
fig. 1). Iam not acquainted with any good illustrations of this member or of the foot
in the Orang.

The back of the hand is hairy down to the distal extremity of the second phalanx in
all the fingers except the second (index). But there is a remarkable patch covered
with very short stubbly hair, caused, no doubt, by the friction produced as the animal
walks resting partly upon the backs of the hands. ‘This patch runs obliquely across the
back of the hand, as shown in the drawing, and is about 4 an inch to ? in breadth.
The hairs upon the fingers are short and stubbly; on the index they are nearly
absent, a narrow line only on the outer side of that finger being hairy.

The grooves on the palm of the hand may be compared with those of the Chimpanzee.
The right hand has in both cases been selected for illustration. They are distinctly
more human in the Orang, the greater resemblance to man being chiefly due to the
fewness of the cross lines. The two lines also which run to the roots of the third and
fourth fingers from the transverse line dividing the palm are seen in the human hand,
but are barely traceable in that of the Chimpanzee. It will be noted that there is
only one continuous cross line (a, 6), and that the line e of the Chimpanzee hand is
wanting. The palm of the hand is illustrated in Alix’s paper referred to in my descrip-
tion of the Chimpanzee’s head. There is also a figure of the palmar as well as the
upper surface of the manus of the Sumatran Orang (regarded by Fischer and Anderson !
as a distinct species and synonymous with Geoffroy’s 7. bicolor) in a paper by Abel ?,
but there is no indication of the grooves nor of the patches covered with short hair on
the back of the hand ; it is very probable that these characters are at least accentuated
during life in captivity, which necessarily gives less opportunities for climbing and
more for walking upon a hard surface.

> Cat. of Mamm. Indian Museum, vol. i.
2 + Some Account of the Ourang Outang etc.,” Asiat. Research. vol. xv. (1825) p. 489.

26 2

204 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

§ 2. The Foot.

The upper and plantar surfaces of the foot (Plate XXVI. fig. 2 and Plate XXVII.
fig. 2) will now be described. The great toe, as has often been pointed out, has no nail.
The covering of hair extends, as in the case of the hand, as far as the end of the second
phalanx; but the hair is considerably longer on the foot than on the hand; the differ-
ence in this respect of hand and foot is well illustrated im the accompanying drawings
(Plate XXVI. figs. 1,2). The long hair, however, stops short at the first phalanx; the
second is covered with short stubbly hair only.

The lines on the sole of the foot differ from those of the Chimpanzee. The second
transverse line (6 in Plate XX VI. fig. 2) is absent ; the line @ goes right across the sole.
Other differences will be apparent on comparing the two figures.

§ 3. The Muscular System.

The muscular anatomy of the greater Orang has been described by several ana-
tomists, including Duvernoy', Bischoff?, Chapman?, Lange‘, and Westling®. It
might therefore seem unnecessary for me to burden zoological literature with further
observations upon a topic which is very far from being new. Nevertheless, I venture
to submit the following account of the myology of the animal, sheltering myself
from criticism behind the opinion of the foremost authority upon the Anthropoids—
Prof. Hartmann.

Prof. Hartmann says :—“ The muscular system of Anthropoid Apes is very interesting.
... . The amount of material which has been collected up to this time is, unfortu-
nately, too scanty to enable us to draw satisfactory conclusions in all cases. We are
often unable to decide whether the conditions presented to us in the case of Anthropoids

are normal or exceptional. .... The assertions on the subject which have been
published to the world and accepted as authoritative have already been shown to be
to some extent untrustworthy. .... Briihl justly remarks that in no department of

anatomy more than in that which treats of the muscles is it more essential that we
should not decide whether a form is normal or exceptional until it has been repeatedly
examined” (‘Anthropoid Apes,’ p. 150).

1 « Tes caractéres anatomiques ete.,” Arch. Mus. t. 8 (1855).

* « Beitriige zur Anatomie des Hylobates leuciscus und zu einer vergleichenden Anatomie der Muskeln des
Affen und des Menschen,” Abh. Bayer. Akad. Bd. x. (1870).

8 “ On the Structure of the Orang Outang,” Proc. Acad. Nat. Sc. Philad. 1880, p. 160.

4 SB. Akad. Wien, Bd. Ixxix.

> « Beitriige zur Kenntniss des peripherischen Neryensystem,” Bihang K. Svensk. Vet.-Akad. Bd. ix.

no. 8.

ANATOMY OF THE ANTHROPOID APES. 205

§ 4. The Muscular Anatomy of the Fore Limb.

(1) Pectoralis major (Plate XXVIII. fig. 4, Pect.1).—This muscle is divided into two
distinct portions, separated by a distance of an inch and a half at their origin, but
gradually converging towards their common insertion. The posterior half of the
muscle is rather more than an inch and a half wide at its origin; its fibres arise from
the sternal part of the 3rd, 4th, and 5th ribs and from the edge of the sternum
adjacent; at its origin the muscle is flat and thin, but gradually gets thicker as well
as narrower; at about an inch before its insertion it again widens out, and is formed of
a wide flat tendon and a narrow muscular strip. The anterior half of the muscle is
thick and narrow, being approximately of the same diameter throughout, except at the
insertion, where it becomes wider and thinner; it arises by a head, measuring rather
less than one-third of an inch in diameter, just below the omohyoid of its side from the
sternum; it has no connection whatever with the clavicle. It becomes fused with the
second part of the pectoral just before their common insertion and lies above it; the
line of insertion of the pectoral upon the forearm measures two inches and a half,
and commences from the head of the humerus just above the biceps ; it is tendinous
on the lower surface but muscular above.

(2) Pectoralis minor (Plate XXVIII. fig. 4, Pect.2).—This muscle is composed of
two perfectly distinct parts: the first arises from the third and fourth ribs at the
junction of the bony ribs with their cartilaginous portions; it is partly overlapped at
its origin by the second part of the muscle. It measures one inch and a quarter in
greatest breadth, and rapidly narrows to the tendon of insertion, which commences (on
the inferior surface of the muscle) an inch and a half from the actual insertion.

(3) The second half of the Pectoralis minor has a broad thin origin from ribs four,
five, and six; but the muscle is already very narrow before it becomes free from the
attachment at its origin. It is inserted by a long and narrow tendon just above and in
common with the anterior extremity of the tendinous insertion of the pectoralis major
on to the head of the humerus.

The sterno-clavicular ligament is attached to the coracoid process in common with
the tendon of the anterior part of the pectoralis minor; its entire length is two
inches and three-fifths; it arises by a few fibres from the anterior end of the sternum
in front and to the outside of the origin of the pectoralis major. It is figured by
Miss Westling.

(4) Coraco-brachialis arises in common with the biceps from the coracoid process ;
the apparent attachment of the muscle is along a line two inches and three-fifths in
length; it is, however, really fixed to the humerus by two separate attachments,
between which intervenes a thick tendinous ridge attached at both its ends to the
humerus, but free in the middle; upon these are inserted some of the fibres of the
coraco-brachialis; the anterior end of the tendon is attached to the flat broad tendon

206 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

of the latissimus dorsi; the posterior attachment of the muscle is to the humerus direct
along a line measuring as nearly as possible an inch, ending at the middle of the
humerus.

(5) The biceps is composed of two very distinct portions: the coracoidal head arises
from the coracoid process in common with the coraco-brachialis ; the two are fused for
a length of two anda half inches. This muscle is rather thinner than the humeral
half; it fuses with the latter one inch and seven-tenths in front of their common
insertion. The muscle is fleshy except just at its origin and for a short way beyond,
where it is covered by a glistening tendinous layer; its connection with the humeral
half of the muscle is as follows—rather more than half an inch before the fusion of the
two the coracoid head becomes tendinous on one side; this tendinous part is inserted on
to the fleshy part of the humeral head on the side of the muscle which faces the bone ;
the fleshy part of the muscle is inserted on to the conjoined tendon of the two heads.

The humeral head arises in the usual way by a strong tendon which gradually passes
into muscle; it begins again to be tendinous some little way in front of its fusion with
the coracoid head.

(6) Latissimus dorsi is a large, flat muscle extending in its origin as far back as
the crest of the ilium; it also arises from lumbar fascia and from posterior ribs; the
muscle rapidly narrows towards its insertion, and at a point about on a level with where
the teres major becomes free from its attachment to the scapula divides into two
portions, one of which is very small. It has a nearly circular section, and measures in
diameter only one-fifth of an inch; its total length is three and a half inches from the
point where it leaves the rest of the latzssemus dorsi to where it joins the teres major,
in common with which it isinserted. The remaining portion of the muscle passes into
a broad flat tendon one inch and one-third in length; the insertion of this tendon is
J-shaped, the recurved bit being anterior and joining the insertion of the teres major ;
the straight part of the tendon is inserted on the tricipital grooves just below the
insertion of the pectoralis major: it is nearly coextensive with the insertion of that
muscle. I have already mentioned that the coraco-brachialis is partly inserted on to
the tendon of the latissimus dorsi. Just at the point where latissimus dorsi passes
into tendon it gives off a dorso-epitrochlear slip, which is a round and fleshy muscle
about one-third of an inch in diameter; at a distance of two inches from the elbow
the muscle passes into a tendinous strip to which are attached some of the fibres of
the triceps and of the brachialis anticus; it is therefore vertical, being apparently
attached to the humerus; a little later it becomes free as an extremely fine tendon
which lies lightly stretched like a violin-string between the shaft of the humerus and
its flexor condyle.

(7) Trapezius—The origin of this muscle from the spines of the vertebra extends
back as far as the commencement of the origin of the latissimus dorsi; anteriorly the
origin reaches the head, but owing to the removal of the brain I could not make out

ANATOMY OF THE ANTHROPOID APES. 207

the exact manner of its origin. It is inserted on to the external third of the clavicle
and on to the greater part of the scapular spine.

(8) Rhomboideus major.—This muscle arises below the last from the spines of the
cervical and first dorsal vertebra; its line of origin measures three and a half inches;
it arises by fleshy fibres, among which there is, here and there, a slight admixture of
tendon. It is inserted on to the posterior border of the scapula immediately above the
insertion of the serratus for a length of one and nine-tenths of an inch about two-thirds
of the length of the border of the bone.

(9) Rhomboideus minor.—This muscle, unlike rhomboideus major, has a distinctly
tendinous origin of about one quarter of an inch in length; the muscle arises just
behind the rhomboideus major, but is not overlapped by it, except for about a quarter
of an inci: anteriorly; the origin of the muscle is about one and a quarter inch in
length. Its insertion is not very distinct from that of the last muscle; it occupies the
rest of the posterior border of the scapula; it overlaps the insertion of the last muscle
before they become joined. .

(10) Rhomboideus occtpitalis——This is a slender flat muscle, measuring nearly one
inch across at the origin from the occipital and gradually diminishing towards its
insertion. For the last inch and a half or so of its course it runs parallel to, and in
close contact with, the rhomboideus major. But it is quite distinct from it, being not
flat but cylindrical, and ending in a longish and very narrow tendon of insertion on to
the extreme upper angle of the scapula.

(11) Levator angult scapula.—The insertion of this muscle is on to the outer angle
of the scapula just above the termination of the line of insertion of the serratus
magnus.

(12) Omohyotd has a fleshy origin from a tubercle on the inferior border of the
scapula one inch from the glenoid fossa.

(13) Serratus magnus arises from all twelve ribs and by twelve more or less marked
digitations; the posterior four or five are less marked than those in front; it is inserted
to the whole of the posterior border of the scapula along a line below the insertion of
the rhomboideus, extending as far as the teres major in front.

(14) Deltoid.—This muscle is very large; it arises from the last inch or so of the
clavicle, from the acromion, and from the spine and posterior border of the scapula,
mainly by a thin tendon which forms a fascia covering the underlying supra-spinatus ;
the latter origin forms an almost distinct head. The muscle is fleshy down to just
before its insertion; the insertion is mainly on a rough triangular area, the deltoid
impression, which measures nearly two inches in length. A few fibres from the deltoid
are continued on to the drachialis anticus.

(15) The teres major arises from an area of the scapula near the upper inner border
(axillary), from the axillary border itself, and from the septum between itself and the
infra-spinatus, which is really, as has already been pointed out, partly the origin of

a

208 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

the deltoid. The area of origin of the teres major measures rather more than one inch
and a quarter in length. The muscle is flat and strap-shaped and of some thickness ;
it is inserted on to the humerus by a flat tendon, some of which, as has been already
explained, belongs to the latissimus dorsi; the line of insertion, which is just opposite
to that of the greater part of the insertion of the datissimus dorsi, measures about an
inch and a half.

(16) eres minor.—This muscle has an entirely fleshy origin partly from the axillary
border of the scapula, partly from the septum between itself and the infra-spinatus ;
the line of origin measures one inch and three-fifths; it is inserted on to the head of
the humerus by an almost entirely fleshy insertion below, and separated by an interval
from, the insertion of the infra-spinatus.

(17) Infra-spinatus occupies the whole of the infra-spinous fossa, to the greater part
of which, however, it is not attached ; it arises from the spine of the scapula up to the
head, from the posterior border, from the axillary border, and from the fascia covering
it; it also arises from the septa between itself and the two teres muscles. Its tendinous
insertion is continuous with, and cannot be separated from, the insertion of supra-
spinatus and the ligament uniting the scapula with the humerus.

(18} Supra-spinatus occupies the whole of the supra-spinous fossa.

(19) Subscapularis covers and arises from nearly the whole of the subscapular fossa ;
the insertion on to the humerus measures one inch and a quarter in length, the lower
part of the insertion below the tuberosity being fleshy.

(20) Triceps.—This muscle has (excluding the dorso-epitrochlear) three heads of
origin. The middle or long head arises from the border of the glenoid cavity, and
from the inferior border of the scapula for an inch behind this ; this origin is chiefly
muscular, though tendinous where it is attached to the glenoid border ; its inferior
surface is covered by six or seven narrow tendinous bands which extend for a very short
way down the muscle. The outer head arises from the humerus commencing about
half an inch below the insertion of the teres minor: the origin of the inner head
commences a little below that of the outer head.

(21) Brachialis anticus—This muscle is large and fleshy; it arises from a large
portion of the shaft of the humerus on both sides of the insertion of the deltoid; on
the inner side of the humerus the origin extends a little way above the origin of the
internal humeral head of the triceps; it passes under a tendinous arch left in the
origin of the external humeral head of the triceps. The origin of the muscle is also
from the septum between itself and the external head of the ¢riceps; on the outer side
of the insertion of the deltoid the origin of the muscle does not extend forwards much
beyond the termination of the deltoid insertion, from this point downwards the origin
of the muscle occupies the whole of the inferior surface of the humerus, coming into
contact, on each side, with the origins of the triceps and supinator longus. ‘Towards
the distal end of the attachment some of the fibres arise from a septum between itself

ANATOMY OF THE ANTHROPOID APES. 209

and the triceps; this septum is continuous with the dorso-epitrochlear ; posteriorly it
becomes a narrow ligament, inserted on to the flexor condyle of the humerus; an inch
and a half from the distal extremity of the humerus the muscle becomes free, and,
curving over the joint, is inserted on to the ulna, being continuous also with the
ligament binding the humerus and ulna. On the outer face the insertion is muscular ;
on the inner face it is tendinous anteriorly, and also where it joins the aforesaid
ligament.

(22) Anconeus is present, and looks like a continuation of the triceps.

(23) Supinator radii longus.—This is a large muscle; it arises just in front of the
extensor carpi radialis longior, from the upper part of the external ridge of the
humerus; anteriorly to this it is continuous with the fibres of the coraco-brachialis, of
which it appears to be a continuation ; the origin from the bone measures an inch and
three-quarters in length. It is inserted by a flat broad tendon, two inches long, upon
a prominent ridge on the external border of the radius.

(24) Eatensor carpi radialis longior.—This muscle arises from the lower part of the
external condylar ridge of the humerus, just below the last muscle. Three inches
from its origin it becomes a tendon, which passes over the wrist in close connection
with the tendon of the next muscle, and is inserted on to the base of the metacarpal
ot the index; just before its insertion it is reinforced by a thin tendon arising from the
extensor carpi radialis brevior.

(25) Extensor carpi radialis brevior arises from the extensor condyle of the humerus
by a tendon common to it and the other extensors; also from the fascia covering the
supinator brevis, and from the septum between itself and the extensor communis ; it
becomes entirely tendinous three inches from the wrist. At about an inch from the
wrist it gives off a thin tendon which, running obliquely forwards, joins the tendon of
the extensor carpi radialis longior just before its insertion. ‘The main tendon of the
present muscle, which is slightly broader than that of the “longior,” is inserted into
the base of the metacarpal of the middle digit.

(26) Extensor communis digitorum.—This muscle arises from the extensor condyle,
and lies between the extensor carpi radialis brevior and the extensor minimi digit, trom
the septa between which muscles and itself it also arises; about halfway down the
forearm it divides into four muscles, from each of which a tendon arises supplying
digits 1.-v.; that supplying digit 1v. gives off a very thin tendon, which joins the deep
extensor of digit 111.; they are inserted over the first and second phalanges.

(27) Extensor minimi digiti lies between the eatensor communis digitorum and the
eatensor carpi ulnaris; just before the wrist it divides into two, which join the tendons
of the extensor communis, which go to the fourth and fifth digits.

(28) Extensor carpi ulnaris arises from the extensor condyle and from a considerable
length of the ulna; the short thick tendon is inserted on to the outer side of the base
of the last metacarpal.

VOL, XI11.—pPart v. No. 5, February, 1893. 24H

210 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

(29) Supinator radii brevis—This muscle appears to be divided into two parts: one
superficial and one deep. ‘The upper part of the muscle is covered by a fascia, from
which some of the extensors partly arise; it arises from the ligament of the elbow-
joint and from the ulna; it passes right round the radius on to the flexor side, and is
inserted on to that bone as far down as the beginning of the insertion of the pronator
vadii teres; the lower layer of the muscle is inserted on to the extensor side of the
radius.

(30) Extensor ossis metacarpi pollicis.

r

This muscle is considerably the largest of the
deep extensors; it takes origin from the radius, from the ulna, from the interosseous
ligament, and from the septum between itself and the ewtensor secundi internodit
pollicis; the radial origin commences two and a half inches below the bend of the
radius, and a little in front of, and below, the attachment of the deep layer of the
supinator radié brevis; the tendon of this muscle is short (two inches in length), and is
inserted on to the radial carpal.

(31) Eatensor secundi internodii pollicis.—This is a slender muscle, arising from the

ulna just below the ulnar origin of the last muscle ; it also arises from the interosseous
ligament; the tendon, which is long and thin, passes above the tendon of the extensores
carpi radialis to be inserted on to the proximal end of the last phalanx of the thumb.

(32) Eatensor indicis arises from the ulna, from the interosseous ligament, and from
the septum between itself and the eatensor carpi ulnaris; it divides into two just
before the wrist, and it is inserted on to the first phalanx of the index, and on to the
corresponding phalanx of the third digit.

(33) There is an extensor primi internodii pollicis, the tendon of which passes in
close contact with the tendon of the eatensor ossis metacarpi; it is inserted on to the
inner side of the base of the thumb metacarpal.

(34) Pronator radii teres—This muscle arises from the flexor condyle of the humerus,
beside the palmaris longus and above the other flexors; it also arises from the radius
and from the septa between itself and adjacent muscles; on the radial side, where it is
free, it is largely covered by glistening tendon ; on the opposite surface it is nowhere
free, some fibres taking origin from the flewor carpi radialis just at the end of the
muscle; its insertion on to the radius, which is fleshy, measures one inch and two-
fifths in length, and commences immediately behind the insertion of the swpinator
brevis.

(35) Flevor carpi radialis arises from the flexor condyle of the humerus and from
the septa between itself and the pronator radii teres and the flexor sublimis ; it becomes
free immediately below the insertion of the pronator radii teres ; its tendon of insertion
does not commence until just before the wrist; it is inserted on to the base of the
second metacarpal.

(36) Palmaris longus.—This muscle is the most external and superficial of the flexors,
as well as the smallest; it is hardly more than one quarter of an inch in width at the

ANATOMY OF THE ANTHROPOID APES, 211

widest part; it becomes tendinous about halfway down the arm; the tendon is inserted
on to the palmar fascia just at the wrist.

(37) Flexor carpi ulnaris—This muscle arises from the flexor condyle of the
humerus, from the fascia covering the forearm, from the septum between itself and the
flexor sublimis, and from the first two-thirds of the ulna; its tendon of insertion (on to
pisiform) is very short and strap-shaped.

(38) Flexor sublimis ( perforatus) digitorum arises from the flexor condyle between
the flexor ulnaris and flexor carpi radialis, from the septum between itself and these
muscles and also the flewor profundus, and from a part of the radius behind the
insertion of the pronator radii teres; it separates into four tendons in front of the wrist,
each of which is inserted on to the second phalanx of digits 11.—-v., being perforated
by the tendon of the flexor profundus. All four tendons are approximately equal
in size.

(39) Flexor profundus (perforatus) digitorum.—This muscle appears to represent
both the muscle so called and the flexor pollicis longus of human anatomy, since it
arises both from the radius and ulna; it also arises from the interosseous ligament,
and from the septa between itself and adjacent muscles. ‘The muscle itself divides
into four some way in front of the wrist; of these divisions those belonging to the
tendons of digits 11. and Im. are the most prominent, each with a glistening tendinous*
surface beneath. The tendons of the two outer digits arise earlier than the other two,
from the common muscular mass; the four tendons are associated in pairs, those of the
two outer digits forming one pair, those of the two inner digits another pair. Each
tendon is towards its insertion not obscurely grooved upon the under surface ; more
distally still the two halves of each tendon are easily separable; each is inserted on to
the terminal phalanx. A small muscular slip arises from the fourth muscle and passes
into a long fine tendon, which joins the tendon of the fourth digit some way beyond
the origin of the /umbricales.

(40) The dumbricales are four in number; each passes from the deep flexor to the
extensor tendon on the dorsal side of the first phalanx of its digit.

(41) The pronator quadratus is about an inch and a half in length; it passes across
from the radius to the ulna at the wrist end of these bones.

§5. The Muscular Anatomy of the Hind Limb.

(1) Gluteus maximus.—This muscle has an extensive tendinous origin from the
anterior and posterior border of the ilium, and an entirely fleshy origin from the
coccyx ; the tendon of insertion rapidly narrows towards the actual insertion.

(la) Ischio-femoral'.—This muscle arises in common with the three next muscles
from the tuber ischii; its fleshy insertion is on to the femur just above the origin of

1 T use the name given by MM. Gratiolet and Alix (doc. cit. on p. 180 of this memoir).
2H 2

212 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

the femoral head of the diceps, and measures two and a half inches, being continuous
with that of the gluteus maximus.

(2) Glutwus medius is largely covered by the last; it arises from the greater part of
the fossa of the ilium, and from the tendon of the glutwus maximus (= fascia lata) ;
the tendon of insertion on to the head of the femur is dorsad of that of the glutwus
minimus.

(3) Glutwus minimus.—This is of course the smallest of the three glutei; its
greatest diameter is three quarters of an inch; it arises from the border of the greater
sciatic notch ; it is inserted partly on to the ligamentous capsule of the head of the
femur, partly on to the femur itself between the insertions of the glutwus medius
and the next muscle.

(4) Iam not quite certain what name to give to a triangular fleshy muscle taking
origin from the ilium, and inserted on to the femur in front of and below gluteus
minimus.

(5) Pyriformis.—This muscle comes through the great sciatic notch; it adheres
closely to the gluteus medius, and is inserted by a tendon on to the femur just behind
the insertion of that muscle.

(6) Obturator internus has, as usual, accompanying it two gemell?, and is inserted in
common with the next.

(7) Obturator externus arises before the last, and is inserted in common with it on
to the fossa behind the great trochanter.

(8) Quadratus femoris.—This muscle is entirely fleshy; it arises from the tuber
ischii below the muscle next to be described ; it measures about half an inch across at
the widest part.

(9) Biceps femoris.—This muscle has two heads: one of them arises from the hip-
bones, the other from the femur; the first origin is from the tuber ischii, in common
with the last muscle and the two next to be described ; the muscle gets wider and
thinner towards its insertion, which is partly on to the fascia covering the knee-joint,
and partly (in common with the femoral head) on to the head of the fibula; the femoral
head of the diceps arises from the femur along a line measuring two and a half inches,
almost exactly co-extensive with the insertion of the quadratus femoris, from the
tendon of insertion of which some of its fibres take origin ; its insertion is continuous
with that of the long head, and is on to the fibula in common with the very small portion
of the humeral head which is so attached ; some of its fibres seem to run down and
become continuous with flexor, which muscle, at any rate partly, arises from the tendon
of insertion of the dzceps.

(10) Semimembranosus arises by a flat, mainly tendinous, head in common with and
below the origins of the biceps and semitendinosus ; it is inserted by a short and stout
tendon on to the inner side of the tibia.

(11) Semitendinosus.—This muscle is nearly of the same size as the last ; its origin

ANATOMY OF THE ANTHROPOID APES. 213

is in common with the last muscle; it becomes free from the semimembranosus three
quarters of an inch from the biceps and one inch and a half from the commencement
of the common origin. The muscle is entirely fleshy until an inch and a half before
its insertion by tendon; this tendinous insertion is thin and flat, and becomes consider-
ably wider at the actual connection of its fibres with the bone; the insertion is higher
up the leg than that of the gracilis, but is situated to the inside of it.

(12) Sartorius.—This muscle is exceedingly slender ; it arises from the anterior end
of the ilium, and passes obliquely over the thigh to be inserted by a broad flat tendon
on to the tibia above insertions of. the gracilis and semitendinosus.

(15) Rectus femoris.—The origin of this muscle is entirely tendinous below, and half
tendinous, half muscular above ; it arises from the ilium just in front of the glenoid
cavity, and from the fascia covering itself and the vastus externus; towards its insertion
it becomes glistening on the under surface ; the insertion is by a flat short tendon on
to the patella in common with a part of the tendon of the conjoined vasti.

(14) Vastus.—I cannot separate the vastus externus from the internus, or either of
them from the crureus; they all form together one muscle, which arises from a large
portion of the surface of the femur below the head, and by a tendon from the outer side
of the head continuous with the insertion of the gluteus minimus. The conjoined
muscle is inserted by a wide tendon partly on to the patella and partly on to the
ligaments of the knee-cap.

(15) Gracilis.—This muscle arises from the symphysis pubis by an origin measuring
nearly an inch and a half in length, and lying superficial to that of all the other
muscles arising here; the muscle becomes gradually narrower towards the tendon of
insertion, which measures rather more than an inch in length; this tendon is at first
narrow, but widens out to a diameter of about three quarters of an inch at its actual
insertion, which is outside, and for the most part below, that of the semitendinosus.

(16) Pectineus.—The origin of this muscle is in front of, and in contact with, that of
the adductor longus; its insertion on to the femur is posterior to and above the lesser
trochanter, and below, as well as partly posterior to, the insertion of the adductor
brevis.

(17) Adductor longus.—The origin of this muscle is, as just stated, from the pubis,
immediately behind that of the pectineus ; its insertion is on to the linea aspera of
the femur, ventrad of the insertion of the adductor magnus; some of the fibres of the
vastus arise from its tendon of insertion.

(18) Adductor brevis.—The origin of the adductor brevis is behind that of the
adductor longus, and slightly overlapped by it; it lies in front of, and is hardly distin-
guishable from, that of the adductor magnus; it is inserted by a very thin and flat
tendon between the insertion of the pectineus and the glut@us maximus; the length
of the insertion is three quarters of an inch; it commences just below the trochanter.

(19) Adductor magnus is, of course, much the largest of the three adductors; its

214 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

origin extends right round from the anterior part of the symphysis pubis to the ischial
tuberosity ; it divides near its insertion into two bands of muscle, of which the anterior
is the wider; this (the anterior) is inserted along a line measuring an inch and a
quarter, and lying between the insertion of the adductor longus and the origin of the
femoral head of the déiceps ; the second part is inserted close to the end of the femur.

(20) Tidbialis anticus.—This muscle arises from the head of the tibia, and from its
shaft down to about the middle ; it also arises from the septum between itself and the
extensor communis and from the fascia covering the leg. It is inserted by a stout tendon
measuring one inch and a half in length to the radial tarsal.

(21) Extensor proprius hallucis is a slender delicate muscle arising from the fibula
and from the interosseous membrane; the area of origin of the muscle is an inch and
a half or so in length; the tendon in which the muscle ends is slender; its expanded
flattened extremity is inserted on to the base of,the phalanx of the hallux.

(22) Hatensor longus digitorum pedis.—This muscle arises from the tibia and for
rather more than half of the fibula, from the interosseous membrane, and from the
septa between itself and the tibialis anticus and peroneus ; at the ankle-joint the muscle
gives rise to three tendons, which are inserted on to the last phalanx of each of the
last three toes.

(23) Hatensor brevis digitorum is represented by the fleshy mass covering the foot ;
this is really separable into three muscular slips, which supply digits 11.-1v.; the muscle
supplying digit 11. is partly inserted on to the tendon of the long extensor; the two
extensors supplying digit 1v. do not join until long after they have both become
tendinous.

(24) Peronwus longus.—'The muscle arises from the head of the fibula and from its
shaft down to about the middle, from the septum between itself and the flexor on one
side and the eatensor longus digitorum on the other; the tendon crosses the foot
between the tarsal bones and the metatarsals, and deep of the intrinsic muscles of the
foot: it is inserted on to the metatarsal of the hallux.

(25) Peronwus brevis.—This muscle arises from the lower part of the shaft of the
fibula and from the septum between itself and the peronewus longus and from the fascia
covering the leg ; its tendon is inserted on to the outside of the last metacarpal.

These are the only two peroneal muscles.

(26) Gastrocnemius.—The gastrocnemius arises by three distinct heads. The outer
head arises in common with the flexor from the outer head of the femur; it becomes
separate from the flevor two and a quarter inches from their common origin. The
inner head arises from the femur just behind the internal condyle, and just below, and
in front of, the insertion of the hinder part of the adductor magnus. The third head,
which corresponds to the solews of other animals, including Man, arises by a stout
tendon on the under surface of the head of the fibula. ‘The three heads join rather
further than halfway down the leg. ‘The common tendon is inserted on to the os

ANATOMY OF THE ANTHROPOID APES. 215

calcis; the muscular fibres continue to within a quarter of an inch of the actual
insertion.

The plantaris muscle appears to be totally absent.

(27) Popliteus——The muscle runs obliquely across under the surface of the knee
from the external condyle of the femur to the tibia. Its insertion on to the latter
measures an inch and a quarter in length.

(28) Flexor longus ( perforatus) digitorum pedis.—The origin of this muscle extends
from just below the head of the tibia and the insertion of the popliteus to more than
halfway down the bone. It is connected also with the origins of the two remaining
long flexors. It divides beyond the ankle-joint into three tendons, which go to
digits IL., I1., and v.

(29) Flexor brevis is a broad strap-shaped muscle arising from the calcaneum ;
beyond the trifurcation of the flexor longus tendon it divides into three tendons, one
of which joins the superficial (perforated) tendons of digit 1v., the other two supply
digits 11. and 111.; these branches become the perforated tendons of those digits, being
quite indistinguishable in their characters from the -perforated tendons of adjacent
digits. The tendon supplying the second digit does not become fused to the super-
ficial long tendon, as it does in the case of that supplying the fourth digit; what
happens is that the tendon of the flewor accessorius takes the place of the superficial
tendon, which the tendon of the real superficial tendon perforates ; the former runs to
the last phalanx of the digit as if it were a branch of the flexor profundus. ‘The same
thing occurs with the tendon supplying the third digit, only that here the perforating
tendon is really the deep flexor.

(30) Flexor profundus digitorum.—The origin of this muscle is from the outer con-
dyle of the femur, from the fibula to beyond its middle point, and from the septa
between itself and adjacent muscles; it divides at the middle of the sole of the foot
into two tendons, which supply digits m1. and Iv.

(31) There are four lumbricales, as in the hand: that supplying the index arises
only from the flexor perforatus of that digit; the others are attached partly to the
deep and partly to the superficial flexors, binding them together in a complicated
fashion, of which there is only a trace in the hand.

(32) Tibialis posticus.—This muscle lies deep of the other long flexors, and is
smaller than either of them ; its tendon is inserted on to the wrist. ©

(33) Abductor hallucis is a fleshy muscle arising from the caleaneum ; it ends in a
tendinous expansion inserted on to the phalanx of the great toe.

(84) Abductor minimi digiti is a fusiform muscle, also arising from the caleaneum ;
it divides into two muscles just before the head of the metatarsal; these soon become
long tendons, which are inserted on to the distal extremity of the metatarsal of the
little finger and on the proximal end of the first phalanx ; the Jatter tendon is the
larger.

216 MR. F. E. BEDDARD—CONTRIBUTIONS TO THE

(35) Flexor brevis hallucis—This muscle divides near its insertion into two bellies,
one of which is inserted in common with the abductor, the other with the adductor.

(36) Adductor hallucis.—This is a large fleshy muscle arising from the plantar
fascia and from the heads of metacarpals 1. and I1l.; it is inserted in common with
the last muscle.

(37) Lransversus pedis.—This muscle is much more limited in extent than in the
Chimpanzee ; it arises along a narrow line of the plantar fascia, measuring four-fifths
of an inch in length on a line with the second metatarsal ; it is inserted entirely on to
the end of the metatarsal of the great toe, and not at all on to the phalanx.

Prof. Huxley, in his ‘ Manual of the Anatomy of Vertebrate Animals,’ mentioned
some of the principal muscles of the Anthropomorpha. My dissection of the Orang
does not, however, bear out all his statements, though no doubt there are variations.

_Thus the transversus pedis is not absent, though it is smaller than in the Chimpanzee.
The adductor brevis is not absent, as Bischoff stated 1, but is present, as Miss Westling
has pointed -out+,—_Dr. Chapman! speaks of the scansorius as “ gluteus minimus.”
It has been shown in the preceding pages that both these muscles are present in the
Orang dissected by myself; it is possible, therefore, that in the Orang dissected by
Dr. Chapman the glutwus minimus was really absent; the probability of the absence
of this muscle in the Orang is increased by the fact that Dr. Chapman identified it in
the Chimpanzee. Dr. Chapman found, as I did, no trace of a plantaris; vat Sandifort
(quoted by Dr. Chapman) asserted the presence of this muscle. Dr. Hartmann, how-
ever, failed to find it; its absence, therefore, must be regarded as typical. The same
observer has described a double origin to the dorso-epitrochlear, part of it arising from
the clavicle ; I found nothing of the kind.

§ 6. The Palate.

The ridges upon the hard palate of the Orang have been figured by Gegenbaur ? ;
but this figure shows certain differences from what I have observed in the individual
described in the present paper. ‘The number of these ridges appears to be about the
same ; but they are much more regular in the palate described and figured by Gegenbaur.
This is particularly so with the last four ridges. In both specimens (pl. v.) the palatal
ridges are more numerous than in the Chimpanzee, indicating, so far as this character
can be made use of, the greater proximity to Man of the Chimpanzee. On the other
hand, the great irregularity of the palatal ridges in the Orang is such as is met with in
the Anthropoids generally. A peculiar feature about the palate of the Orang which
I examined is its deep black pigmentation, shown in the figure (Plate XXYV. fig. 3).
After it had been allowed to macerate for some days, in order to prepare the skull for

* These references will be found on p. 204 of the present memoir.
* « Die Gaumenfalten des Menschen,” Morph. Jahrb, Bd. iv. p. 573.

ANATOMY OF THE ANTHROPOID APES. 217

examination, the black colour could be rubbed off, staining the fingers. I am uncertain
whether to regard it as pathological; if normal, it would strengthen the reasons for
regarding this animal as a distinct species of Orang.

IV. EXPLANATION OF THE PLATES.

PLATE XX.
Head of Troglodytes calvus, drawn after death in August 1891: full face, natural size.

PLATE XXI.

Troglodytes calvus, head of the same animal, viewed vertically from above, natural
size.

PLATE XXII. oe

A hand and foot of the same, palmar surface, natural size : Fig. 1. Hand; Fig. 2.
Foot.—The letters refer to the lines on the palm of the hand and on the
sole of the foot, which are described in the text.

PLATE XXIII.

Brain of the same, natural size.

Fig. 1. Lateral view of right side.

Fig. 2. Lateral view of left side.

Fig. 3. Brain, viewed from above.

Fig. 4. Brain, viewed from the underside.

Fs, Sylvian fissure; /.s.a. Anterior branch of Sylvian fissure ; P.o.f. Parieto-
occipital fissure; (Af. Fissure of Rolando.

PLATE XXIV.

Head of young Orang “ George,” natural size.

AYER; 2XOXOVi.
Fig. 1. Head of the same, viewed vertically from above.
Fig. 2. Palate of Troglodytes calvus (‘ Sally ”), to illustrate the ridges.
Fig. 3. Palate of Simia satyrus (?) (“ George”’).

PLATE XXVI.
Upper surface of hand and foot of Orang (“ George”): Fig. 1. Hand; Fig. 2. Foot.

VOL. XII.—Part v. No. 6.—February, 1893. aes

218 CONTRIBUTIONS TO THE ANATOMY OF THE ANTHROPOID APES.

PLATE XXVII.
Palmar surface of hand and foot of Orang (“‘ George”): Fig. 1. Hand; Fig. 2. Foot.

—The letters refer to the lines on the palm and sole, which are described in
the text.

PLATE XXVIII.
Fig. 1. Brain of Troglodytes calvus (‘ Sally”), from behind. c¢. Cerebellum.
Fig. 2. Brain of Troglodytes niger, from behind. ¢. Cerebellum. 2. Median part of
cerebellum not concealed by lateral lobes.
Fig. 8. Ear of Troglodytes calvus (“Sally”). H. Helix; T. Tragus; F. Fossa.
Fig. 4. Pectoral muscles of Orang (“George”). Cl. Clavicle; 1-5. Ribs; Pect. 1.
Pectoralis primus; Pect. 2. Pectoralis secundus.

TROGL

\ rT

Minter Bro

aah Tian is
Cues

ane
Ait)
dda at

ihe

vie
UN. het
itt es)
cates

re ye ayy
tera

J. Smut del.et lith.

Mintern Bros. imp.

HAD? OF ViROCLODY LES seAL Vil Se
FROM ABOVE.

ah

[eae
rt ae

ieee

veniNs
iy

Bier Lo0b, Soe Vol AU, GE. LEE.

J. Seat del et lith Mintern Bro. imp
HAND & FOOT OF TROGLODYT

2 Cl. AU GE, XA

J. Smit, del.et lth. Bots Minterm Bros, imp.

BRAIN OR TROGLODY TES CALY US.

he

fey
at

we

J. Smit delet lith.

HEAD OF ORANG.

Mintern Bros. imp

Mintern Bros, imp.

ith

it del. ex

CL

M

HI
7 tos

PALATE

D

pie AP
AM Veveden
Nis ‘

ie

-
5

fo)

d. Smit del.et. lth.

hee

J. Smut del.et lth.

rd

fat)

Aros) tra
i ene

CL, Ze

J, Smit del.et ith

ANATOMY OF

VIII. On the British Palwogene Bryozoa. By J. W. Grucory, 5.Se., 2.Z.8.,
British Museum (Nat. Hist.).

Received May 17th, 1892, read June 14th, 1892.

[Puares XXIX.-X XXII. ]

Contents.
Page | Page
eeintroductionseermee arenas 219 | VY. Miscellaneous Records ...... 262
iieRerminologypemeren erence 220 | VI. Stratigraphical Distribution... 263
[enee Classiticationyermre ries 221 | VII. Affinities of the Fauna...... 264
IV. Systematic Synopsis ........ 225 | ValitieeeBibliooraphyarrrerceiric rrr 266
IV.a. The Systematic Position of the IX. Explanation of the Plates.... 276

Adeonedlidae i \s.5255..- 241

1. Introduction.

P ROBABLY few groups of British Neozoic fossils have been so much neglected as the
British Lower Cainozoic Bryozoa. While those of the Crag were carefully monographed
by Busk in 1850 and those of the Cretaceous, Jurassic, and Paleozoic rocks have been
described in numerous memoirs, but little has been done on the Paleogene fauna. In
Morris’s ‘ Catalogue of British Fossils’ published in 1845 only one species is mentioned,
and it was not till 1850 that some were described and figured by Lonsdale in Dixon's
‘Geology of Sussex’; he described four species, of which only one was regarded as new.
In 1866 the next contribution was made by Busk [No. 7] in a paper entitled
“ Description of three Species of Polyzoa from the London Clay of Highgate, in the
collection of N. T. Wetherell, Esq., F.G.S.” This paper, short though it be, is the best
piece of work that has been done on the British Eocene Bryozoa. Since then Mr. G. R.
Vine [4, p. 673] has published a list of the recorded species and has subsequently
described two collections, both of which are now in the British Museum. With these
additions the list numbered twenty-oue, but of these only four are here retained, as the
remainder are either based on identifications that I have been unable to verify or on
indeterminable fragments.

The neglect of this group has no doubt been mainly due to the comparative rarity of
specimens: collectors who have devoted a good deal of time to our Lower Tertiaries
have only met with a few fragments and have not felt much interest in them. . Even in
the principal Museums the British Paleogene Bryozoa are very sparsely represented,
with the single exception of the British Museum, which contains all the material from
many large collections; the collection there now includes all existing types and
figured specimens, with the exception of one specimen figured in this communication.
The principal part of the Bryozoa collection in the British Museum consists of the

vol. xu1.—Part vi. No. 1.—June, 1893. 2k

220 MR. J. W. GREGORY ON THE

« F. E. Edwards Collection,” including Lonsdale’s types. Busk’s types and many other
specimens were obtained with the “ Wetherell Collection,” while additions from the
London Clay of Fareham and Sheppey were made by the acquisition of the collections
of Mr. G. R. Vine and Mr. A. Bell.

The present paper therefore consists mainly of a description of the British Museum
Collection, and for permission to undertake this I have to thank Dr. H. Woodward,
F.R.S. I have of course examined all other available material, including that in
the Reed Collection at York, at the Woodwardian Museum, Cambridge, and in the
Museum of Practical Geology. I must also thank Mr. EK. H. M. Platnauer and Mr. EK.
T. Newton for valuable assistance when examining the collections under their care ;
and I am especially indebted to Mr. H. Woods for the kind loan of the Cambridge
specimens.

Furthermore I must express my best thanks to my colleague Mr. R. Kirkpatrick, of
the Zoological Department, for his ever-ready assistance ; owing to his kindness, I have
enjoyed every opportunity for the examination of the collections of recent Bryozoa, and
especially the type specimens, to which constant reference has been necessary ; he has
also repeatedly discussed the difficulties that have been met with, and his knowledge
of the recent Bryozoa and their literature has always been placed most generously at
my service.

II. Terminology.

Most of the terms employed have a well-established meaning, and consequently do
not require to be here referred to; but the apertures and pores of the Cheilostomata
are so important in diagnosis, and have been so differently employed therein, that it
seems advisable to define them. At the same time a few alterations in terminology are
suggested, as it is hoped thereby to secure greater precision in the description of the
fauna.

Orifice. The opening of the mouth of the polypide: it corresponds in size and shape to the

operculum. In fossil Membraniporide, &c., it cannot be determined.

Aperture. The opening occupied by the membranous area which surrounds the orifice. The
aperture may be primary and either correspond to the orifice as in Lepralia or may be a
large space in the middle of which the orifice opened, as in Membranipora. Or it may be
secondary, formed by the peristome rising up into a tube and concealing the original primary
aperture ; the form of the latter may, however, be always told from the operculum.

Sinus. A notch on the lower side of the aperture, as in Schizoporella.

Trypa. A pore which perforates the front wall of the zocecium; it occurs only in the Micro-
porellidee : it is generally assumed to correspond to the sinus.

Other names have been previously given to this, but there seem to be valid objections to
them all. Jullien has called it the “‘ fenestrula;”’ but this term is already in use for the
interspaces in the zoarium of the Fenestellide. D’Orbigny included it among the “ special
pores,” and as such it is often referred to, though this also includes different structures.

The term “ zocecial pore” is hardly definite enough ; the terms “true pore,” ‘accessory
opening” (D’Orbigny), “ central pore” (Busk), are subject to the same objection.

BRITISH PALAOGENE BRYOZOA. 221

Peristomial Pore (“ Sublabial pore” of Busk). A pore below the aperture which simply leads into
the peristomial chamber.

Punctures. A series of pores left between the anastomosing spines of the front wall of Cribrilina, &c.

Areole. Pits or tubular depressions occurring in linear series around the margins of zoccia, ¢. g.
in Notamia wetherelli.

Macule. A term suggested for the small irregular cavities in the walls of the zoccia: they
correspond to the main part of the “ pores (origelles” of Jullien [No. 3, p. 607], but since
Pergens [No. 7] has thrown such discredit on Jullien’s views on these structures it seems
hardly advisable to circulate this term. The name is derived from “ macule,” the meshes of
a net, as, according to Pergens, they originate simply by non-calcification of part of the
wall. When seen on the front wall of a zocecium they resemble small pits or depressions.

* Opesiule. A term applied by Jullien to the secondary small apertures, of which a pair usually
occur on the front walls of the zocecia of Micropora, &c.

Ill. Classification.

Probably no one who has tried to determine to which of the twenty to thirty families
of Cheilostomata some form new to him must be referred will complain of an attempt to
arrange these families into groups. Among the Kuechinoidea, for example, there are
twenty-five families distributed amongst five orders, some of which are divided into sub-
orders. Butamong the Cheilostomata we have as many or more families, without any
definite larger groups, except the ill-fated ones proposed by Dr. Jullien [No. 4] and the
antiquated ones of Mr, Busk'. The inconveniences of this are manifold ; the diagnosis
of each family has to be of inconvenient length, and the task of discovering the exact
systematic position of any species is a matter of much difficulty.

Neither the Rev. T. H. Hincks nor Mr. Waters offer much encouragement to an
attempt at any serious alteration, as the former points out emphatically that all classifi-
cations at present must be tentative and the latter discourages what he calls “an attack
along the whole line.” But then all classifications are probably more or less tentative
and temporary, and, so far as I am able to judge, some grouping of the families is an
essential preliminary to an attempt to revise the families in detail and dissipate the chaos
in which at present the fossil Bryozoa are involved.

Though there is of course much uncertainty as to the exact taxonomic value of
several characters, there does seem to bea pretty general agreement as to the most
important structures. The development of the front wall seems about the leading
feature, as so many of the other characters, ¢. g. the aperture, the position and deve-
lopment of avicularia, &c., are correlated with this. The use made by Jullien of the
front wall has perhaps prejudiced some workers against this structure; but Jullien has
based his classification on modifications that most workers regard as of very slight value,
while his method of nomenclature is quite his own. As M. Dollfus has pointed out in
an admirable criticism, Dr. Jullien simply does not accept the principle of priority.

1 Busk of course based his divisions on zoarial characters, and these, though somewhat improved by

Dr. Ortmann [No. 1, pp- 3, 4], are now quite inadequate.

222 MR. J. W. GREGORY ON THE

Three groups of the Cheilostomata may be conveniently based on the character of
the front wall. In one, including the Membraniporidan series, this structure is absent
or only imperfectly developed: the name of “ Athyriata” (from «and éupedc, an oblong
shield) is therefore suggested for it. Ina second group the wall is well developed, but
there is an additional communication between the exterior and the polypide by means
of a pore (trypa) on the front wall or by a sinus on the lower margin of the orifice.
The exact homology of these two structures has never, so far as I am aware, been
clearly demonstrated, but it has been generally accepted, for example, by Hincks,
Waters, and Macgillivray. For this group the name of ‘‘ Schizothyriata ” is proposed.
Finally, there is the group in which the calcification of the front wall is complete ; it
may therefore be called the “ Holothyriata.”

In addition to these there is a series of forms whose affinities seen very doubtful.
With one or two exceptions they are rarely or never found fossil, and my opportunities
of studying them have been but limited. They may be divided into two divisions, one
of which may be a natural group. ‘This includes the Mitiidwe, Chlidoniide, and
Eucratiide ; the terminal or subterminal apertures and simple tubular or pyriform zocecia
of these families suggest that they are among the most primitive of living Cheilostomata.
They are here left grouped together, and Busk’s name, the Stolonata, is accepted. For
the other division Smitt’s name of “ Cellularina” is adopted ; but this is certainly not a
natural group. Thus some, such as the Cellulariidz, Bicellariide, and Epistomiide
(Notamiidee of Hincks), seem clearly allied by their large membranous areas and aperture
to the Membraniporidan group; the Catenariide may include representatives of both
the Holothyriata (e. g. Catenicella utriculus, Macgill.) and the Schizothyriata. Among
the latter there may be divisions corresponding to both of the great families ; thus
Catenicella amphora, Busk, is analogous to the Microporellide, and C. pulchella (Maple-
stone) to the Schizoporellide. It is, however, not improbable that the Catenicellide
branched off independently from the main Cheilostomatous stem at a very early
period.

Without more detailed information upon the anatomical structure of the polypides
of the families in this “ carpet-bag” group it seems unadvisable to attempt to place
them definitely. In the Catenariide we have both holostomatous and schizostomatous
(e. g. Claviporella) genera, but until we know more of the anatomy of the polypides it
seems very uncertaiu as to whether this character possesses the same significance as in
those higher Cheilostomata where the skeleton is of a specialized and complex type.
Amongst these the hard parts certainly seem to offer reliable classificatory characters.

Through each of the three suborders an evolutionary series can be traced. Thus
among the Athyriata the Membranoporide seem to be the most primitive, and this
family passes up into the Cribrilinide and Hiantoporide in the manner suggested by
Mr. Hincks [No. 2, pp. 199-200, and No. 5 [pt. 3], pp. 471-472 and 479-480] and
My. Kirkpatrick [No. 2, pp. 616-617].

BRITISH PALAOGENE BRYOZOA. 223

There seems to be a similar evolutionary series in the Holothyriata, where the main
branch develops from the simple Cyclicoporine through the Lepralinew to the more
specialized Smittiide; also in the Schizothyriata from the simple Schizoporella or
Schismoporella to such a form as Adeonella pectinata.

The division of both the Reteporide and Celleporide into the schizostomatous and
holostomatous groups appears to be generally regarded as inevitable. The dis-
memberment of the Selenariide is more likely to be criticized, but it is not a new idea.
It was first done by Prof. Smitt in 1873 [No. 3], and Mr. Hincks [No. 7, p. 125] has
given it the sanction of his high authority by the remark in describing Cupularia
umbellata, Detr., that “this form clearly belongs to the Steganoporellidan series and
must be transferred to it.”

The survival of the family Selenariidee seems to me to well illustrate the necessity
for a grouping of the families; so long as these have been allowed to remain in
independence, such an olla-podrida of species of different families agreeing only in
zoarial form has been able to hang together. The moment we introduce a more
scientific system, define suborders, and try to indicate the affinities of the families, such
a group as the Selenariide falls to pieces.

These remarks are not intended as a formal defence of the classification. Its publi-
cation will be justified only if it is found to aid in bringing the Cheilostomata, and
especially the fossil forms, into better order than they are in at present.

Synopsis of the Classification followed.
Order CHEILOSTOMATA.
I. Suborder Stroronara.
Forms with simple tubular zocecia and terminal or subterminal apertures.
Family 1. Anrzipm. For diagnosis see Macgillivray, No. 3, p. 195.

2. HucRATIIDA. i as pe LoGs
3. CHLIDONIIDA. Ee 3 op Jb ILO,

: II. Suborder CELLULARINA.,
A group of forms with simple zocecia and tufted phytoid zoaria, and probably including repre-
sentatives of the three following suborders.
Family 4. Cettunarip®. For diagnosis see Macgillivray, No. 3, p. 199.

5. BIcELLARIIDA. A D sp [do ZUR,
6. Ertstomip# (Notamiide). es Hincks, No. 2, p. 98.

7. CAaTENICELLID2. = Macgillivray, No. 3, p. 197.
8. BrraxaRIID@. . ee BE)

III. Suborder Atuyriatva.,
Cheilostomata with the front wall uncalcified or incompletely calcified.
Family 9. Farciminariipm. For diagnosis see Macgillivray, No. 3, p. 204.

10. FrLusrripam. tS 35 53 p-. 208.

224 MR. J. W. GREGORY ON THE

Family 11. Mempranirorip®. Athyriata with the front wall mainly membranous and occupied
by an opesial aperture ; this does not correspond to the operculum. The opesium is sur~
rounded by a raised margin. External ocecia.

Subfamily 1. Membraniporine. Membraniporide with open opesia and without, or with but a
small, extra-opesial front wall

Subfamily 2. Electrinine. Membraniporide with the normal zoccia tubular and with a
terminal opesium.

Subfamily 3. Lunulitine. Membraniporide with patelliform zoaria, and with vibracularia
systematically arranged.

Family 12. Crrsrimmipa, Athyriata with a front wall formed by the overarching and branching
of one or more spines. External ocecia.

Subfamily 1. Cribrilinine. Cribrilinidee with the front wall formed by the overarching and
fusion of numerous circumareal spines ; the interspaces remain as grooves or pores.

Subfamily 2. Hiantoporine. Cribrilinidze with the front wall formed of one large spine arising.
from the margin.

Subfamily 3. Steginoporine. Cribrilinide with the front wall formed by the overarching of
spines arising from the peristome.

Family 13. Microroripa. Athyriata with a calcified front wall. Zocecia surrounded by raised:
margins. Nofinternal diaphragms. External ocecia.

Subfamily 1. Microporine. Zoccia all normal or onychocellaria (large vicarious avicularia)
irregularly distributed.

Subfamily 2, Sedenarine. Microporide with patelliform zoaria and vibracularia systematic-
ally arranged.

Family 14. STEGANOPORELLID”, Athyriata without external ocecia and with the zocecia divided
into two chambers by a calcareous diaphragm.

Family 15. Cznzarupa. Athyriata with internal ocecia which open by a pore above the aperture.
The zocecia are surrounded by raised margins; the aperture is situated within the
depressed front wall.

IV. Suborder ScHIZOTHYRIATA.

Cheilostomata which are schizostomatous or trypiate (i. e. provided with a trypa; see p. 220)
or both.
Family 16. Scuizoporniirp®. Schizothyriata not provided with a trypa.
Subfamily 1. Schizoporelline. Schizoporellidee with simple primary aperture and external
occa.
Subfamily 2. Schizoreteporinee.' Schizoporellidze with the zocecia obliquely placed on a
unilaminar, reticulate or ramose, erect zoarium.
Subfamily 3. Schismoporine.’ Schizoporellidee with urceolate zooscia growing in dense
masses ; aperture terminal or subterminal.
Subfamily 4. Biporinee. Schizoporellidee with a patelliform unilaminate zoarium, with
vibracularia systematically arranged.

1 §chizoretepora, n. gen., for which at present the subfamily diagnosis also serves as the diagnosis, is the type
genus ; it includes the schizostomatous Reteporas, of which S. (R.) tessellata (Hincks) [No. 2, p. 358, pl. xix.
figs. 9-12], is a convenient type.

2 Schismopora, Macgillivray [No. 4, p. 29], is the type genus, and S. costata the type species: it does not

BRITISH PALAOGENE BRYOZOA. 225

Family 17. AproneLLip#. Schizothyriata with a schizostomatous primary aperture and a variable
secondary aperture. Goncecia and no external marsupia.
Family 18. Mrcroporriiip®. Schizothyriata provided with a trypa.
Subfamily 1. Microporelline. Holostomatous Microporellide with external marsupia.
A 2. Schismoporeilinee. Microporellidze which are both schizostomatous and trypiate.
. 3. Adeonine. Microporellide which are holostomatous and have goneecia, but no
external marsupia.

V. Suborder HotorHyRtatTa.

Holostomatous Cheilostomata which have the front wall wholly calcified.
Family 19. Leprariipm. Holothyriata with a simple primary aperture.
Subfamily 1. Lepraliine. Lepraliidie with external ocecia.
Alliance 1. Cycliopora. Lepralinz with simple zoewcia having orbicular apertures
which are often surrounded by raised rims.
Alliance 2. Lepralia. Lepraliine with the aperture usually horseshoe-shaped and never
truly orbicular.
Subfamily 2. Teichoporine. Lepraliidee with goneecia and no external occia.
3. Reteporine. Lepralide with the zocecia obliquely placed on a unilaminar,
reticulate or ramose, erect zoarium.

PP)

Family 20. Crrierorip#. Holothyriata with barrel-shaped or urceolate zocecia, usually growing
in heaped masses; aperture terminal or subterminal.

Family 21. Smirriip#. Holothyriata with a raised secondary orifice ; the primary orifice is often
denticulate.

Order CYCLOSTOMATA.

Family 1. IpMonrrp#. | Family 2. H»rnroporip®.

IV. Systematic Synopsis.

Class ECTOPROCTA.
Subclas GYMNOLAMATA.
Order CHEILOSTOMATA.
Suborder STOLONATA.
Family EUCRATIID.
Genus Noramia, Fleming, 1828 (non Busk, Hincks, &c.).
Diagnosis. Zoariam erect and phytoid; zocecia biserial, joined back to back; the

apertures of each series respectively open in the same direction. Aperture large, on the
front of the cell. Neither vibracula, avicularia, nor ocecia. [Fleming, No. 1, p. 541.]

appear to have been formally diagnosed ; but the list of six species with the figures of their opercula published
in the ‘ Prodromus’ leaves no doubt as to its nature. {Macgillivray, No. 1, dec. xvii. pp. 168, figs. 1-6. See also

Macgillivray, No. 2, pt. v. pl. ii.]

226 MR. J. W. GREGORY ON THE

Species 1. NoraMIA WETHERELLI (Busk), 1866.

Syn. Dittosaria wetherelli, G. Busk, 1866, Geol. Mag. in. p. 301; G. R. Vine, 1889, Proc. Yorks.
Geol. & Polyt. Soc. xi. pp. 158-159, pl. v. fig. 1.

Records. W. Whitaker, No. 1, p. 594; G. R. Vine, No. 1, p. 675.

Diagnosis. Zoarium in small phytoid tufts; imperfectly known. Branching
dichotomous.

Zoccia elongate, pyriform. Aperture median and symmetrical, oval, the longer
axis in the direction of the length of the zoarium. The aperture opens on the upper
border and occupies about a quarter of the front of the zocecium. ‘The surface is
ornamented with a double series of areole; the innermost series forms an ellipse
passing close round the upperside of the aperture and crossing the front wall at about
the middle ; the outermost series runs close along the hinder margin. The number
varies from 8 to 16 in the inner series, and from 20 to 26 in the outer.

Distribution. London Clay, Highgate (Brit. Mus.).

Dimensions. The zocecia of the specimen figured measure a trifle over ‘5 mm. in
length.

Figures. Pl. XXTX. figs. 1a, 6. Part of a zoarium, X 37 diam. Brit. Mus.

Affinities of the Species. This species differs from Notamia loricata (Linn.) in that
in the recent species the aperture occupies half the front of the zocecium and is
obliquely placed ; it also has no regular series of areole. ‘The same characters serve to
distinguish it from Notamia americana (Lamx.)'. A nearer ally is the Notamia prima
(Reuss)”, which differs from it by the smallness of the mouth and the absence of areole.

Remarks. This species was founded by Busk on a specimen in the Wetherell Collec-
tion which cannot now be recognized, but other specimens labelled by Busk occur and
enjoy almost as much authority as the actual figured specimens. Busk made it the
type of a new genus, Dittosaria, which has been ignored or overlooked by nearly all
subsequent writers. He recognized that it was a close ally of Notamia (Gemellaria),
but distinguished it by its mode of branching; he restricted the old genus to those
which at every fork retain a continuation of the main stem in addition to the two
branches. But this is not even a specific character, as is shown by the following
quotation from Mr. Hincks’s [No. 2, p. 20] description of Motamia (G.) loricata:—* The
branches are given off from each side of the uppermost pair in a stem close to the top,
and at times the stem ascends between them and a triplet is formed in place of the
more usual bifurcation.” The only other point of difference is that the mouth in this
species is not “ slightly oblique” as it should be to conform to Mr. Hincks’s diagnosis
of the genus. But this is hardly of generic value, and Busk certainly regarded the
other as the main character. The genus differs from Pasythea, Lamx., by the absence
of the two notches at the lower corners of the aperture.

* Loricaria americana, Lamouroux, No. 2, p. 7, pl. lxv. fig. 9.
* Gemellaria prima, Reuss, No. 7, p. 170, pl. vii. figs. 6, 7.

BRITISH PALZOGENE BRYOZOA. 227

Nore oy tHE Use or tor Name Gemellaria.The name Gemellaria was first invented for a genus of Bryozoa
by J. C. Savigny somewhere about the year 1810; it was not, however, published till 1826 [Audouin,
No. 1, p. 242], and then only in the French form of Gémellaire ; so far as I am aware, it was first used in a
Latinized form in 1830 by Blainville [No. 2, p. 425], who did not himself accept it. Before the publication
of Gemellaria or Gémellaire the genus had been described in 1821 by Lamouroux [No. 2, p. 7], who named
it Loricaria. Audouin, who completed Savigny’s work when the latter was disabled by ill-health, of course
treated “ Gémellaire” as a manuscript name and accepted Loricaria. Most subsequent authorities, however,
have accepted Gremellaria and date it from 1805, 1809, or 1811. Johnston [No. 2, p- 293, footnote] seems
to have entertained doubts as to the accuracy of this proceeding, but accepted it on the idea that copies
of some work of Savigny’s had been placed in the principal libraries: he obviously could get no reliable
information regarding it.

Mr. Hincks accepts the genus and quotes as its author “ Savigny, 1811.” The only reference he gives in
his Bibliography [ No. 2, p. 588] to Savigny is “ Iconographie des Zoophytes de YEgypte,” from the ‘Description de
Egypte.’ Miss Jelly | No. 1, p. 284] quotes the same work, and so does Macgillivray [No. 3, p. 223), who,
however, adds “not seen by me.” I regret to have been unable to find any such work; there is none such in
the Natural History Museum copy of the ‘ Description de l’Egypte,’ nor is any referred to in “ A Bibliographical
Account and Collation of ‘ La Description de l’Egypte’ ” (London Institution : private circulation, 1838, vo,
76 pp.). None of the ordinary bibliographical works of reference give any information regarding it.
I therefore cannot help concluding that the authorities who have quoted this mysterious “ Iconographie ”
really refer to Audouin’s “ Explication sommaire des planches de Zoophytes de ’Egypte.... . ” That the
date of this is 1826 and not 1811 admits of no doubt: the work was only entrusted to Audouin for completion
in 1825, and monographs issued in 1521 are quoted. oricaria has therefore the prior claim to adoption, but
unfortunately it had been previously used among fishes. Fleming [No.1, p. 541], therefore, in 1828 renamed it
Notamia. N. loricata he clearly regarded as the type, for the only other species he associated with it
(NV. bursaria) he made the type of a new genus, Epistomia. Lamouroux did not include this latter species
in his Loricaria, but in the Sertularian Dynanema [No. 1, p. 79]. Fleming, it must be remembered, only
proposed Notamia asa change of name owing to the preoccupation of Zoricaria. The name Notamia cannot
therefore be separated from its type species and applied to one which both Lamouroux and Fleming assigned to
another genus. There is therefore no option but to follow Fleming and substitute Notamia for Gemellaria
and regard the species bursaria as the type of Epzstomia.

The only alternative is to accept Blainville’s name Gemicellaria [No. 1, p. 425], proposed in 1830, but there
does not seem any sufficient reason for a departure from the ordinary rule of nomenclature.

Suborder ATHYRIATA.

Family MEMBRANIPORIDA.

Subfamily MEMBRANIPORIN &.

Genus Mrempranipora, Blainville, 1834.
[Blainville, No. 2, p. 447.]
Diagnosis.1 Membraniporide in which the opesial aperture is generally of a simple
form and the lamina is absent or but slightly developed.

* It will be seen from this diagnosis that in deference to recognized opinion Amphiblestrum is accepted ; it
appears to be an artificial but very convenient group.

VOL. XIi1.—ParRT vi. No. 2.—June, 1893. 2.

228 MR. J. W. GREGORY ON THE

Species 1. MremBranipora EOCENA (Busk), 1866.
Syn. Biflustra eocena, G. Busk, 1866, Geol. Mag. iti. p. 300, pl. xu. fig. 2; W. Whitaker, 1872,
Mem. Geol. Surv. iv. pt. 1, p. 594; G. R. Vine, 1886, Rep. Brit. Assoc. 1885, p. 673.
Biflustra (Membranipora) eocena, G. R. Vine, 1889, Proc. Yorks. Geol. & Polyt. Soc, xi.
p- 160, pl. v. fig. 4.
Flustra crassa, Desm., J. Morris, 1843, Cat. Brit. Foss. p. 37; Huxley & Etheridge, 1865,
Cat. Foss. M. P. G. p. 332; Huxley & Newton, 1878, Cat. Tert. & Post-Tert. Foss.
M. P. G. p. 14.

Diagnosis. Zoariwm large, expanded, foliaceous. Bilaminar, the internal face ribbed
by long and prominent angular ridges.

Zowcia quadrangular, arranged in long, oblique lines. The opesia are elliptic and
fairly regular, with a strong, slightly raised rim ; this is surrounded by a flat area, on
the part of which that covers the continuation of the zocecium are two distinct rounded
avicularia. The width of the surrounding area and the prominence of the rim vary
somewhat in different parts of the zoarium, but within a restricted area are quite
uniform.

Avicularia: usually a pair on the front wall below the aperture.

Figures. Pl. XXIX. fig. 2. Part of zoarium, from a specimen from the London
Clay, Highgate; Brit. Mus. No. 49729; x 16 diam. Fig. 5. Another specimen
showing back view, xX 21.

Distribution. Thanet Sand, Pegwell Bay (M. P. G.). London Clay, Southampton.
London Clay, Highgate. Edwards Coll. Brit. Mus. 49729.

? Bracklesham Beds, Bracklesham.

Remarks. This species was founded by Busk, who gave four figures of it; these well
show the general form of the zoarium, the thickened longitudinally ribbed back, the
form of the opesia, and the large front wall below the aperture. These are the main
specific characters. Busk’s type was in the Wetherell Collection, but it cannot now be
found. Though the figures do not show the pair of avicularia, there can be no doubt
of the species, for the Wetherell Collection contains many specimens from Highgate
labelled by Busk and Wetherell. The specimen from which the accompanying figures
have been drawn is from Southampton. A small specimen in the Edwards Collection
from Bracklesham appears to belong to this species, but as it only shows the back view
of the inner lamina it is impossible to be certain. ‘The Thanet Sand specimens are so
much worn that one cannot be sure of the identification.

The species belongs to the group of Membranipora of which . savarti (Aud.) [No. 1,
p. 240, pl. x. fig. 10; see also the figures by Smitt, No. 3, p. 20, pl. iv. figs. 92-5]
is a convenient type; from this, however, it differs in the absence of the crenu-
late margin and the two tubercles sometimes present in that species; the area of the
front wall is much larger than in Audouin’s species, and the back is longitudinally
ribbed instead of having the flat surface marked off into regular rectangles as shown by
Smitt. The plain prominent rim and large front wall also separate this species from

BRITISH PALZOGENE BRYOZOA. 229

M. lacroixi (Aud.) [No. 1, p. 240, pl. x. fig. 9]. I. eocena is more nearly allied to Mem-
branipora appendiculata (Reuss), of which a good figure has been given by Mr. A. W.
Waters [No. 12, pl. ii. fig. 3], but from this it differs in that Reuss’s species has a single
large avicularium on the lower side of the aperture and not quite in the median line;
the opesia is also somewhat too large. /. macrostoma (Reuss) is another ally ; but
this has the rim that borders the opesia closer to the margin of the zocecia, so that the
flat depressed marginal space is absent.

Species 2. MEMBRANIPORA BUSKI, 0. sp.
Syn. Membranipora lacroizi, G. Busk (non Aud.), 1866, Geol. Mag. vol. iii. pl. xi. figs. La & d;
(fide Vine), J. W. Judd, 1883, Geol. Mag. dec. 2, vol. x. p. 527; G. R. Vine, 1889,
Proc. Yorks. Geol. & Polyt. Soc. vol. xi. pt. 2, pp. 159-160, pl. v. fig. 2 (copied from Busk),
fig. 3 (original) ; H. W. Bristow, 1889, Geol. I. Wight, ed. 2, p. 284.
Membranipora reticulum, Vine (non Linn.), ibid. vol. xii, pt. 1, pp. 59, 60.
Diagnosis. Zoarium encrusting or foliaceous. The back is flat and not ribbed.
Zoecia arranged in long series. Opesia very large: no lamina or front wall, the
raised rims of adjoining zocecia being in contact. The general form is oblong, the
length being not much greater than the width, except at the bifurcations of a row,
where the two zocecia are long and narrow. ‘The raised rims are thick and plain.
Avicularia irregularly scattered, small, generally in the lower right-hand corner of
the zocecia.
Oacia not always present, narrow, globose.
Distribution. Headon Beds, Colwell Bay, I. of Wight ; London Clay, Highgate.
Type. Brit. Mus. No. B 4629.
Figures. Pl. XXIX. fig. 11. Part of a zoarium with an ocecium; X 55 diam. Brit.
Mus. No. B 4625. Fig. 12. Part of a specimen (Mus. Pract. Geol.) with ocecia, x 55 diam.
Affinities. This species in its general characters very closely approaches M. lacroizi,
Aud. [No. 1, p. 240, pl. x. fig. 9], and as such the London Clay specimen has been
figured by Busk. With this identification I agreed until seeing the specimens in the
Museum of Practical Geology: these were collected by Mr. Chapman and are clearly
the same as those which he has kindly presented to the British Museum. ‘They, however,
show the ocecia, and thus clearly separate the species from J. lacroiai, from which,
according to Mr. Hincks’s diagnosis [ No. 2, p. 130], these structures are absent.

Species 3. MEMBRANIPORA CRASSOMURALIS, 0. sp.

Diagnosis. Zoarium irregular, encrusting.

Zowcia oval, irregularly distributed. Each zocecium surrounded by a thick prominent
rim. ‘The interspaces between these rims are very narrow. When encrusting ribbed
bivalves the zocecia are more regularly arranged, running along the ribs or pressed into
the furrows. Opesia usually occupying the whole of the area, but in some a thin

narrow lamina occurs.
My iy

230 MR. J. W. GREGORY ON THE

Owcia triangular: surrounded by a rim like that around the zocecia.

Avicularia sparsely and irregularly scattered over the zoarium: occupying the
small triangular areas between the zocecial margins.

The raised rim is usually plain, but may bear a single minute tubercle, the base of a
small spine.

Distribution. Barton Beds, Barton. Bracklesham Beds, Bracklesham.

Type. Brit. Mus. No. 49741.

Figures. Pl. XXIX. fig. 10 a. From Barton. Several zocecia, showing the ocecia,
avicularia, bases of spines, and lamina. Fig. 10. Another specimen, growing on a
strongly ribbed Pecten.

Remarks. This species appears to be most closely related to that figured by
Reuss [No. 14, p. 179, pl. ix. figs. 1, 2] as Membranipora elliptica (Hag.) from the
Leithakalk (Helvetian) of Hisenstadt. But the London Clay species appears to be
certainly distinct from that represented in von Hagenow’s original figure [No. 1, p. 268,
pl. iv. fig. 6], in which the rims surround the area instead of the zoccia and thus
are separated by a wide space, both in the centre and youngest part of the zoarium ;
there are neither laminee nor ocecia. Hagenow remarks on the “ vertieften Zwischen-
riumen” with ring-shaped pores. But as to the identity of IZ. crassomuralis with
the Eisenstadt species I do not care to express a definite opinion witheut seeing
Reuss’s type. Pergens [No. 1, pp. 15, 16] seems to have entertained the same
doubts as to the correctness of Reuss’s identification, for though he quotes MV. elliptica
from the Austro-Hungarian Miocenes, he does not include Reuss’s reterence in his
synonymy.

This species belongs to the MW. lacroizi group, but it differs in the following characters :
(1) it has triangular ocecia, whereas these structures are said by Hincks [No. 2, p. 130]
to be absent in the recent species; (2) the rim is not crenulate ; (3) the avicularia are
fewer, and there is never more than one spine on the rim.

From Membranipora eocena (Busk) it differs in the absence of any space below the
area and outside the rim, and also of the two small lateral avicularia; the zocecia are
also arranged more irregularly.

Membranipora temporaria, Waters [No. 6, p. 288, pl. vii. fig. 16], from the Murray
River cliffs, is an allied species, but differs in the presence of two small lateral avicularia
and a larger “ infra-area.”

Another species with which this new one must be compared is Membranipora
lowopora (Reuss) [No. 2, p. 166, pl. vii. fig. 11: for later figures see No. 14, pp. 39-
40, pl. ix. figs. 4, 5; the author’s original figure in No. 1, p. 97, pl. xi. fig. 24,
has been subsequently repudiated by him], but this has larger front walls, on which the
avicularia are placed, instead of in the angles.

Reuss [No. 13, p. 101, pl. xxiv. figs. 4 & 5c] has himself also figured the typical
Cretaceous MW. elliptica from the Unter Planer of Saxony, and one of his figures shows

BRITISH PALAHOGENE BRYOZOA. 231

pores at the ends of some of the zoccia in the positions occupied by the ocecia in
M. crassomuralis ; Reuss, however, regards them, no doubt correctly, as avicularia.
In the same work Reuss [ib. pl. xxiv. fig. 5, pp. 100-101] has figured a variety of
M. subtilimargo which resembles MV. crassomuralis more than does the typical form ;
but the absence of ocecia and lamine clearly distinguishes it.

Species 4. MEMBRANIPORA TENUIMURALIS, 0. sp.

Syn. Membranipora lacroixi, Busk, 1866, Geol. Mag. iii. pl. xii. figs. 1b & le; W. Whitaker,
1872, Mem. Geol. Surv. iv. pt. 1, p. 594.

Diagnosis. Zoarium encrusting (or ? sometimes free), spreading as a thin gauze-like
layer.

Zoecia ivregularly distributed. Form irregular, oval, quadrangular, hexagonal or
polygonal: closely crowded. The opesia are very large, almost as large as the zocecia:
coincident with the area. There are small triangular depressions between the margins
of the opesia of the different zocecia. Walls thin, sometimes crenulate. There is often
a pair of tubercles on the margins of the zocecia, and these may fuse to a single large
tubercle on the infra-area.

Avicularia: usually a pair of small ones in the infra-area covering the continuation
of the zocecia.

Ovwcia, none.

Distribution. London Clay, Highgate. Clarendon Hill, Fareham, Portsmouth.

Type. Wetherell Coll. Brit. Mus. No. 49736.

Figures. Pl. XXIX. fig. 5. London Clay, Highgate. Brit. Mus. No. 49736 (one of
Busk’s type specimens).—Figs. 6 & 7. Other specimens from same locality.

Affinities. This is also a species of the puzzling Jacroizi group. Its nearest ally is
probably WM. tuberculata (Bosc), which it resembles in its tuberculation [No. 1, t. iil.
p. 143. Bose gives as a reference the Plustra dentata of O. F. Miller, Zool. Dan. iii.
pp. 24, 25, pl. xev. figs. 1, 2, but this is quite different]. But it differs from this in
the greater thickness of the walls in IM. tuberculata and in the presence in that
species of a small front wall; in the new species, moreover, the zocecia are more
regularly hexagonal in form and are more elongated ; there is also a small depressed
area in the corners between the rims margining the opesia. From the recent WZ. mem-
branacea (Linn.) it differs in the regularly alternate arrangement and rectangular
shape of the zocecia in that species; WM. tenuimuralis also lacks the hollow marginal
spines so characteristic of the recent species.

From M. lacroixi (Aud.) it differs in the presence of avicularia, and of the pair of
tubercles or knobs; the form of the zowcia is angular instead of oval, and the margins
of the opesia are rarely crenulate. The comparison with J. lacroiat is especially
necessary as Dr. Pergens makes MV. Jawa, Reuss [No. 11, p. 252, pl. xxxvi. fig. 14], a
synonym of this species; and M. Jawa appears to be the closest ally of the London

232 MR. J. W. GREGORY ON THE

Clay Bryozoan. M. laxa is a somewhat doubtful species ; it has not been referred to by
Mr. Waters [No. 12] in his recent revision. Reuss’s figure may only represent a
specimen in which the whole of the front wall is broken away and only the lateral walls
are left; but if that is the case it is certainly not IM. lacroixi, and in view of Pergens’s
conclusion it would not be safe to act on this view. Reuss’s figure shows more regularly
hexagonal zoccia; the margins appear to be separated entirely by a narrow groove,
and there are no tubercles. Hence it seems safest to make a new species for this
London Clay form rather than to assert the existence of so doubtful a species as the
North Italian Bartonian 7. laxa in the Lower Eocene of the London Basin.

Species 5. MEMBRANIPORA VIRGULIFORMIS, 0. sp.

Diagnosis. Zoarium of elongate, cylindrical, solid shoots, somewhat resembling those
of Cellaria.

Zoccia in regular longitudinal series, elongate, rectangular. Opesia large, oval, sur-
rounded by a thick raised and plain non-crenulate rim. A large depressed front wall
below the area, often with a pair of triangular depressions.

Oacia, none.

Avicularia single, prominent, lateral, on the upper left-hand margin of the zocecia.

Distribution. London Clay, Highgate.

Type. Brit. Mus. No. 49658. Edwards Coll.

Figure. Pl. XXIX. fig. 8. Part of zoarium, X 20 diam.

Affinities. In its mode of growth this species resembles J. sigillata (Poutt.)
[No. 1, p. 110; see also Smitt, No. 3, p. 8, pl. il. figs. 64-68], but the zocecia in that
species are more irregular in form and distribution, while their general form is lozenge-
shaped instead of rectangular. It also recalls to mind M. monostachys, Busk |No. 2,
p. 31, pl. ii. fig. 2], but from this it differs by the somewhat pyriform shape of the
zocecia and the more curved instead of flattened front wall of that species.

Among the Lower Tertiary species, this most closely resembles Membranipora macro-
stoma (Reuss) | Cellaria macrostoma, Reuss, No. 1, p. 64, pl. viii. figs. 5,6; Biflustra
macrostoma, Reuss, No. 11, pp. 274, 275, -pl. xxxiil. figs. 12, 13], but in that the sub-
areal portion of the front wall is regularly rounded and has not the pair of triangular
depressions seen in the new species.

Species 6. MEMBRANIPORA DISJUNCTA, n. sp.

Diagnosis. Zoarium forming a large encrusting surface; the zocecia are arranged in
disconnected rows, which are radially disposed; there are several centres of radiation
in each zoarium.

Zoecia elliptical ; opesia large, surrounded by a prominent rim; the mouth opens
at one end of the opesium ; the rest is occupied by a thin calcareous lamina.

Avicularia and owcia unknown.

BRITISH PALAOGENE BRYOZOA. 233

Distribution. London Clay, Highgate.

Type. Brit. Mus. No. 69205. Wetherell Coll. Encrusting Hippochrenes ampla.

Figures. Pl. XXIX. figs. 9a, 6. Fig. 9a, part of zoarium, magnified 4 diam.,
showing radial growths; fig. 96, x 12 diam.

Affinities. The mode of growth in loose disconnected rows resembles that often
assumed by IV. catenularia (Jameson) [No. 1, p. 561, name only| (Pyripora of Mac-
gillivray) | No. 1, pt. xi. p. 24], but the much greater size of the opesia in this species

is quite distinctive.

Genus Lunutites, Lamarck, 1816.
[Lamarck, No. 1, i. p. 194.]

Diagnosis. A genus of Membraniporide with a unilaminate, conical, or cup-shaped
zoarium. The zocecia are arranged in radial rows; radial rows of vibracularia either
separate the zocecia or occur alternately.

Type species. JL. radiata, Lamk. [No. 1, p. 195].

Species 1. LUNULITES TRANSIENS }, n. sp.

Syn. Lunulites urceolata, Lonsdale, 1850, in Dixon’s Geol. Suss. pp. 159, 160, pl. i. fig. 8; 1878,
do. ed. 2, pp. 201, 202, pl. i. fig. 8.
Lnunulites ? radiata, Lonsdale, 1850, in Dixon’s Geol. Suss. ed. 1, pl.i. fig. 8; 1878, do. ed. 2.

Diagnosis. Zoarium of medium size, depressed, circular, thin, cup-shaped ; convex
margin curved.

Zoecia. Opesia with the aperture large, orbicular, elongate ; a small lamina at the
lower end. ‘The lateral margins are steep ; the inner margin slopes more gently. A
pair of small tubercles occur on some of the margins between the two zocecia.

Vibracularia large, aperture clithridiate ; the radial series are connected by a groove ;
they increase in size towards the periphery, and gradually pass into normal zocecia (whence
the specific name). On the concave side the ridges are irregularly distributed and are
separated by deep grooves; there are numerous large pores ; on the narrower parts of
the ridges there may be only a single line of pores.

Dimensions. Diameter -5 mm.; height 1°25 mm. Taken from a small complete
specimen. In some fragments the number of zoccia is from 18-20; number of
zocecia in a radial series 10.

Distribution. Upper Eocene, Barton Beds, Barton. Middle Eocene, Bracklesham
Beds, Bracklesham, Bramshaw, Brook, Whitecliff Bay.

Type. Brit. Mus. No. 49724. From Barton. Edwards Coll.

Figures. Pl. XXIX. fig. 13. Part of zoarium showing back, x 24 diam. Fig. 14.
Several normal zocecia, x 24 diam.—Pl. XXX. fig. 1. Another specimen, showing the

* Referring to the gradual passage from vibracularia to zocecia.

234 MR. J. W. GREGORY ON THE

ancestrula. Fig. 2. Part of zoarium from Bracklesham (B 4339), showing the front wall
partly broken away. Fig. 3. Part of a worn specimen from Bracklesham, resembling
L. wrceolata.

Affinities. This species belongs to the L. radiata, Lamk., group, which the Marquis
de Gregorio [No. 1, p. 248] has recently proposed to make into a new subgenus, Demi-
clausa; this, however, is against all rules, as L. radiata is clearly the type species of
the genus. If, therefore, the separation is to be made, it is the other group that must
be renamed and removed. Demiclausa is an absolute synonym of Lunulites.

This species was figured by Lonsdale as Lunulites urceolata, Lamk., but from the
latter it widely differs in the fact that the vibracularia are connected by depressions
into long radial lines; in L. wrceolata they are disconnected.

From Lunulites radiata, Lamk., this differs by the gradual transition from the vibra-
cularia to the normal zocecia, and by the presence of a lamina and tubercles on the rim
of the opesia. The species agrees most closely with L. subplana, Reuss [No. 3, p. 264,
pl. xi. fig. 108], but the apertures in that species are not clithridiate, nor does there
seem to be a gradual transition from vibracularia to zocecia. It clearly differs from
LInunulites quadrata, Reuss [Cellepore quadrata, Reuss, INO py OO ple xaueti oma (ewan
the explanation of the better figure given in Reuss, No. 11, pl. xxviii. fig. 18, the species
is called Lepralia tetragona |, by the form of the aperture and the absence of the raised
rim immediately around it. The original figure gives a suggestion of a similar passage
from vibracularia to normal zocecia.

In the main character of this species it resembles Lunulites goldfussi, Hag. |No. 2

= 5

padO2> (pl: xl. ig: 5], but that differs by the irregular distribution of the vibracularia.

Genus BISELENARIA, Nov. nom.

Syn. Diplotaxis, Reuss, 1867, non Kirby, 1837, Ueber Bry. deut. Unteroligocins, Sitz. k. Ak. Wiss.
Wien, Bd. lv. Abth. i. p. 231.

Diagnosis. A Membraniporid with a bilaminate zoarium, which is small and circular
and discoid in form; typically the form is bun-shaped. ‘The zocecia of the upper layer
have regular Membraniporidan apertures, with numerous normal vibracularia irregularly
scattered, or one to each zocecium. The zocecia of the lower surface are much modified ;
the aperture is contracted by the great thickening of the peristome ; in the zocecia near
the centre the aperture is sometimes completely closed or persists as a long narrow slit ;
the vibracularia are similarly modified ; some of the peripheral zocecia more nearly
resemble those of the upper layer.

Type species. Biselenaria placentula (Reuss), op. cit.

Remarks cn the Genus and its Affinities. —Reuss practically founded his genus Diplo-
taxis simply on the one character of its bilaminate zoarium; the species included in
it are forms of much interest, and there seems to be no reason to question the validity

BRITISH PALZOGENE BRYOZOA. 235

of the genus, though it has been overlooked or merely mentioned by subsequent authors.
Unfortunately, however, the name was preoccupied among Coleoptera by Kirby in 1857,
and as it is still in use for that group the Bryozoan genus must be renamed. ‘The
nature of the zocecia of the lower surface is somewhat puzzling; four explanations
of their nature may be offered:—First: the zoarium may be fixed, probably in mud;
in that case the peripheral zocecia would be normal; but as they became more central
by the growth of the colony they would gradually become aborted and their apertures
closed; the distribution of the under zoccia supports this view. Second: the zoarium
may be free and the modified zocecia of the lower surface may all be swimming vibracula
instead of normal zocecia; in that case the thickening of the peristome would be due
to the necessity for greater muscular attachments. ‘Third: the zoarium may be fixed
by radical fibres or tubes given off from the modified zocecia. And fourth: the zoarium
may be free and the peculiar lower zocecia may be goneecia, as the thickened and con-
tracted apertures resemble those of elements in other genera, such as Teichopora, which
appear to be clearly goneecia. So long as the genus remains known only by extinct
species it may be impossible to decide between these views, but I am inclined to accept
the first, though there are points that make for the second.

The genus differs from the rest of the group by its bilaminate nature and the structure
of the inferior zoccia. It is possible that it ought to be subdivided, one branch including
the type species and all the rest of those in which there is a vibracularium to every
zocecium.

Species 1. BiseLENARIA oFFA !, n. sp.

Diagnosis. Zoariwm: a small circular disk, thickest in the middle and tapering
towards the periphery.

Zoeca irregular in form and distribution ; a group of small ones occurs in the centre ;
the largest are in a circle at a little distance from the margin. The opesia are large
and elliptical, surrounded by a thickened margin ; some of the opesia are slightly trigonal.
The vibracularia are very irregular in distribution ; they resemble the normal zocecia
in general form, but the rim is thicker in proportion to their size.

The zocecia of the lower side vary from being identical with those of the upper side
to being quite closed; all intermediate forms occur, but a spathulate form with the
aperture remaining as a slit or small pore is the commonest. Some of the vibracularia
have the very typical auriculate appearance.

Distribution. Barton Beds, Barton.

Type. Brit. Mus. No. 49759. Edwards Coll.

Figures. Pl. XXX. fig. 4. Zoarium of type specimen: upper surface. Fig. 4a. Part
of another specimen: under surface. Fig. 5. Upper surface of another zoarium.

' Offa, a bun.
VOL. XIII.—Part v1. No. 3.—June, 1893. 2M

236 MR. J. W. GREGORY ON THE

Affinities. This species differs from the type species, Biselenaria placentula (Reuss),
in several important respects; the most striking is that in the type there is a vibra-
cularium to every zocecium, situated just at the apex. This is practically the main
character used in the separation of Cupularia and Selenaria; as in this case it is there-
fore generic, it might be thought that the two species ought to be separated into two
genera, one including B. placentula, corresponding to Cupularia, and one including
B. offa, corresponding to Selenaria. The two species, however, agree so closely
that it would appear to be unnecessary to make a new genus upon this character alone.
In merely specific points, the concavo-convex form of B. placentula, its more irregular

opesia, and the larger size and smaller number of its inferior zocecia all distinguish it
from B. offa.

Family CRIBRILINIDA.
Genus Criprinina, Gray, 1848.
Diagnosis. Hincks, No. 2, p. 184.

Species |. CRIBRILINA VINEI, 0. sp.
p ’ Pp

Syn. Membraniporella nitida, Jobnst. var. eocena, G. R. Vine, Notes on Brit. Eoc. Polyzoa, 1889,
Proc. Yorks. Geol. & Polyt. Soc. vol. xi. pt. ii. pp. 161-2, pl. v. fig. 6.

Diagnosis. Zoarium encrusting.

Zoucia large, quincuncially arranged ; globose. Orifice large, orbicular ; elongated
transversely. Margin of the orifice raised, thin and plain.

The front walls of the zocecia are traversed by 9 or 10 pairs of furrows; the upper
5 or 6 pairs of these are horizontal; the lowest 3 or 4 pairs in a radial fan. There
are two or three pores in each furrow. The furrows do not reach the middle line of the
front wall, and upon this there is a varying number of fairly large pores.

Avicularia large: a pair on each side of the orifice.

Owcia large: very globose, often covering the lower part of the adjoining zocecium,
Perforated by numerous, fairly large pores.

Distribution. London Clay, Sheppey.

Type. Brit. Mus. No. B 4514. Vine Coll.

Figures. Pl. XXX. fig. 8. Part of the zoarium, X 55 diam.

Affinities. This species was regarded by Mr. Vine as only a variety of the recent
Membraniporeila nitida; he remarked the presence of a series of small pores in the
furrows, and that Mr. Hincks did not mention them in his diagnosis of that species.
But the existence of these pores is the generic character that separates Cribrilina
from Membraniporella, and into the former genus this species must necessarily go.
From the species to which Mr. Vine referred it, it differs also in the presence of the pores
on the ocecia, in that the lower furrows are radial instead of them all being horizontal,

BRITISH PALAOGENE BRYOZOA. 237

and in other features. From the common and widely distributed C. radiata [Moll, No. 1,
p. 63, pl. iv. fig. 17] this species differs by its larger orifice and by the furrows being
more numerous and differently arranged. Among recent species it most closely resembles
C. philomela, Busk [No. 8, pp. 132-3, pl. xvii. fig. 6, pl. xxii. fig. 7], to which it is allied
by the large size of the orifice and the big globose occia; it differs, however, in the
ocecia being plain in the recent species, and also in having more pores on the furrows
and none in the middle line.

Probably the nearest ally to this species is Cribrilina manzonii | Lepralia manzonit,
Reuss, No. 14, p. 171, pl. i. fig. 6], from Médling, near Vienna, which agrees with it in
the large size of the orifice and the arrangement of the furrows: Reuss does not figure
any ocecia, and consequently this important character cannot be used for comparison ;
but the absence of the pair of large lateral avicularia and the greater number both of
pores and furrows in C. manzonii are sufficient to distinguish the two.

The species belongs to Cribrilina, even restricted as narrowly as is done by

Dr. Jullien [No. 3, 604].

Family MICROPORIDZ.

Genus MIcROPORA.

Diagnosis. Hincks, No. 8, pt. i. p. 161.

Species 1. MicROPORA CRIBRIFORMIS, 0. sp.

Syn. Membranipora holostoma, Busk, var. perforata, G. R. Vine, 1891, Proc. Yorks. Geol. & Polyt.
Soe, vol. xi. p. 60.

Diagnosis. Zoarium encrusting.

Zoecia oval, sometimes tapering below. ‘The lower part of the front wall is very
tumid and rises above the raised margin. The aperture is small; the upper margin is
regularly curved, the lower margin sinuous. The front wall is crowded with macula,
which are very irregular in form and numbers. ‘There is usually a pair of narrow slit-
like opesiule situated at the extreme margin of the occia, just below the corners of
the aperture.

Distribution. Barton Beds, Barton.

Type. Brit. Mus. No. B 4583.

Figures. Pl. XXX. fig. 6. Part of zoarium. In one of the zoecia the front wall has
been broken away and shows the absence of internal partitions.

Affinities. This species is very clearly marked by the sinuous lower border of the
aperture and the cribriform aspect of the whole front wall. Both characters, as well
as the form of the zocecia and other less important points, separate it from /. holostoma
(Busk) [No. 6, p. 36, pl. iii. fig. 11], from the Crag.

Probably the most nearly allied species is MM. gracilis (Miinst.) [Cellepora gracilis,

2m 2

238 MR. J. W. GREGORY ON THE

(Su)

Miinster, in Goldfuss, No. 1, i. p. 102, pl. xxxvi. fig. 15], of which Reuss [No. 11,
p- 291, pl. xxix. fig. 13] has given a good figure; from this it is distinguished by the
form of the orifice, the absence of a ridge on the lower side of the aperture, and the
much greater coarseness of the macule. Waters [No. 12, p. 15] includes the Crosara
species as a synonym of J. coriacea (Esper). ‘The same characters separate it from
M. iniinsteri (Reuss) [ No. 6, p. 30, pl. x. fig. 2], which is very nearly allied to I. gracilis.

As in the new species some of the zocecia and the opesiule are replaced by large
pores, while in others these are no larger than some of the macule, it is evident
that Mr. Hincks is fully justified in refusing to regard the presence of these opesiule
as an essential character of the genus.

Genus ONYCHOCELLA, Jullien, 1881.

Diagnosis. Microporidee with large vicarious avicularia scattered over the zocecia
[Jullien, No. 1, p. 277].

Species 1. ONYCHOCELLA MAGNOAPERTA, 0. sp.

Diagnosis. Zoarium encrusting, forming a large compact crust.

Zoecia usually hexagonal, occasionally becoming rounded at the edges and oval
where they are less crowded. Apertures slightly clithridiate, very large, occupying
nearly the whole front of the cell; the aperture is restricted by a small lamina at the
lower side of the zoecium. The margins of the zocecia are raised, plain, and non-
crenulate.

Avicularia: large vicarious cells, long and tapering; irregularly scattered over the
zoarium.

Distribution. Brockenhurst Beds (Mid. Headon), Brockenhurst.

Type. Brit. Mus. No. 49738. Edwards Coll.

Figures. Pl. XXX. fig. 7. Part of zoarium, X 7 diam., including one of the large
tapering vicarious avicularia. 4

Remarks. The subdivision of the great genus Membranipora to which Jullien [No. 1,
p- 277] gave the name Onychocella appears to be based on more reliable characters
than most of the genera which that author has proposed, and it seems to be now gener-
ally accepted [see Waters, No. 12, pp. 8, 9]. The nature of the avicularian cells of
this new species shows that it belongs to this group. Its nearest ally is O. angu-
losa (Reuss) [No. 1, p. 98, pl. xi. fig. 10], from which it differs in the much
smaller size of the aperture in that species. If, as Waters suggests, Rhagasostoma
hexagonum, Kosch. [No. 1. p. 30, pl. v. figs. 5-7], is only a synonym of O. angulosa, it
will be unnecessary to compare them further; but if, as appears probable, it is a

distinct species, the structure of the aperture wil) clearly distinguish it from the
Brockenhurst form.

BRITISH PALAAOGENE BRYOZOA. 239

O. maqnoaperta is closely allied to some Upper Cretaceous species; of these 0. cyclo-
stoma (Goldf.) | Hschara cyclostoma, Goldfuss, ‘Petrefacta Germaniz,’ Th.1i. 1826, p. 23,
pl. viii. fig. 9] appears to be about the nearest ; the evidence for referring it to Onycho-
cella is given by von Hagenow’s figures [No. 2, p. 75, pl. ix. figs. 7, 8, pl. xii. fig. 3]:
from this, which is biflustrine in habit, it may be distinguished by its clithridiate
aperture; the avicularian cells agree in general character. From QO. koninckiana (Hag.)

Cellepora (Discopora) koninckiana, Hag. ib. p. 95, pl. xi. figs. 10, 11] it differs in the
ovate shape of the avicularian cells, which in the Maastricht species are lanceolate.
O. santonensis, D’Orb. [Eschara santonensis, D’Orbigny, No. 2, p. 109, pl. 673. fig. 4},
agrees with it in the large size of the aperture and the shape of the avicularia; but the

ocecia in that species are pyriform, the lamina larger, and the lower side of the mouth
straight. 0. drya, D’Orb. | Eschara drya, ib. p. 168, pl. 677. figs. 7-9], has also a large
aperture, but this is much wider and not clithridiate; the zocecia are also different in
shape. D’Orbigny has figured amongst his Escharas a large series of species which
must be referred to Onychocella, though many of them may be reduced to synonyms.

rom most of them, such as 0. allica (D’Orb.), 0. archosia (D’Orb.), O. charonia
(D’Orb.), O. clito (D’Orb.), and O. cressida (D’Orb.), the new species may be distin-
guished by its large aperture.

The occurrence of the genus Onychocella in Cretaceous rocks has been frequently
pointed out; the British Museum Collection contains a specimen from the Calcaire
& polypiers (Bathonian) of Ranville, that must be referred to this genus.

Suborder SCHIZOTHYRIATA.

Family SCHIZOPORELLIDA. (Myriozoidw of Hincks.)
Genus ScHIZOPORELLA, Hincks.

Diagnosis. See Hincks, No. 2, p. 237.

Species 1. ScHIZOPORELLA MAGNOAPERTA, 0. sp.

Diagnosis. Zoarium, a foliaceous expansion.

Zoecia somewhat irregularly arranged, though with a tendency towards quincuncial.
In shape they are pyriform, well rounded above, tapering below. The front wall is
tumid, forming a raised triangular area. A raised lip around the orifice, which is oval ;
the sinus is median, small but distinct. The zocecia are separated by a depressed flat
margin, around which is a row of large deep areole.

Avicularia one on each zocecium, beside and below the orifice; they have raised,
elliptic borders.

Oucia q

Distribution. Barton Beds, Barton.

240 MR. J. W. GREGORY ON THE

Type. Brit. Mus. No. 49733. Edwards Coll.
Figures. Pl. XXX. fig. 9. Part of a zoarium from London Clay, Sheppey; Brit.
Mus. No. B 4514, x 2 diam.

Affinities. This species belongs to the group of which the common Schizoporella
umcornis, Johnst., is a good representative; it agrees with the latter in its umbo, sub-
orbicular mouth, and small sinus. From that species, however, it clearly differs in the
much larger size of the aperture and the pyriform shape of the zocecia in the new species,
in which also the umbo is lower down, and there is one lateral avicularium instead of
the pair usually present in S. wnicornis; the areole are also limited to a single series.
The large size of the aperture at once distinguishes this from most of the Continental
Miocene and Lower Cainozoic species, such as §. goniostoma | Cellepora goniostoma,
Reuss, No. 1, p. 87, pl. x. fig. 18; for better figures see Reuss, No. 14, p. 176, pl. ii.
fig. 6, pl. iii. fig. 3] and S. rugulosa [Reuss, No. 14, p. 169, pl. iii. fig. 2]. 8S. dunkeri
[Reuss, No. 1, p. 90, pl. x. fig. 27] agrees in some respects, ¢. g. the single lateral
avicularium, the large mouth, and blunt umbo; it is probably the nearest ally of this
species. Reuss’s species may be distinguished by its higher umbo, marginal avicularia,
and shorter and more rectangular zocecia. Among recent species it agrees closely
with S. simplex D’Orb. [Eschara simplex, D’Orbigny, No. 1, p. 13, pl. v. figs. 5-8],
from which it differs in the pyriform shape of the zocecia.

In the general form of the zoccia this species agrees strikingly with Microporella
membranacea (Reuss) | Hschara membranacea, Reuss, No. 6, p. 32, pl. v. fig. 6], from
Oberburg ; the possession of a sinus instead of a trypa, of course, distinguishes it from
that species.

Species 2. SCHIZOPORELLA MAGNOINCISA, n. sp.

Diagnosis. Zoarium foliaceous.

Zoecia narrow and elongated ; peristome raised and almost subtubular. Aperture
large and with a very large sinus; the angles of the peristome above the sinus vary
considerably in prominence, but never meet. One line of areole. Front wall smooth
and evenly convex.

Avicularia: one on each zocecium, just below the aperture; lateral in position ;
mandible pointing upwards to the angle between the aperture and sinus.

Oecia (none 2).

Iistribution. Loudon Clay, Copenhagen Fields.

Type. Brit. Mus. No. B 4515. Fragment enclosed in a septarian nodule.

Figure. Pl. XXX. fig. 10, x 30 diam.

Affinities. The large size of the sinus of this species would necessitate its inclusion
in Gemellipora if that genus of Smitt’s [No. 3, p. 35] be accepted. Its nearest ally

BRITISH PALHOGENE BRYOZOA. 241

appears to be Schizoporella gonversi (Reuss) [No.14, p. 159, pl. vii. fig. 7], from
Rauchstallbrunn, but in that species the zocecia are shorter and broader, the areol,
fewer, and there is a pair of avicularia above the aperture.

The large size of the sinus allies this species to Schizoporella beyrichi, Stol., but it
differs in that the zocecia are elongate and rectangular instead of hexagonal, they are
not quincuncially arranged, and the zoarium is not Cellarian (Cellaria beyrichi, Stoliczka,
No. 1, p. 83, pl. 1. fig. 10).

Schizoporella insignis, Hincks [No. 4, pt. 5, p. 134, pl. v. fig. 10], differs in the
quincuncial arrangement of the zocecia, the central umbo, and the raised line at a little
distance from the margins of the zocecia. ‘The shape of the zocecia and the absence of
the tubercles above the aperture distinguish this new species from S. pauper (Reuss)
[| Lepralia pauper, Reuss, No. 14, p. 164, pl. v. fig. 4], which has a large sinus. The
last two species with which it is necessary to compare this are S. variabilis (Reuss)
[Hemeschara variabilis, Reuss, No. 12, p. 508, pl. i. figs. 1-5] and S. unicornis (Johnst.)
[Lepralia unicornis, Johnston, No. 2, p. 320], which both belong to the same group.
From the former the London Clay species is mainly to be distinguished by the size of
the sinus. The latter differs by its umbo, the absence of macul, and the smaller
aperture ; the zocecia, however, agree in general form.

IV. a. The Systematic Position of the Adeonellide.

The genus Adeona was established by Lamouroux [No. 1, pp. 478-482, pl. xix. fig. 2 |
in 1816 for some Bryozoa with short jointed stems and reticulate zoaria; he took
an Australian species, 4A. grisea, as his type: this species has also been made by
Macgillivray the type of a genus Dictyopora, which is therefore necessarily a synonym.
Enlarged figures of the zocecia have been given by Kirchenpauer [No. 1, pl. i. fig. 8,
pl. ii. fig. 10] and Macgillivray | No. 1, pl. 66], and these show that it possesses a trypa
or zocecial pore and a simple holostomatous orifice, and must therefore be referred to
the Microporellide. But this genus and its allies have long given much trouble to
systematists and the classification is still unsettled. Busk’s ‘Challenger’ Report must
certainly be held responsible for much of the confusion, as he there founded a genus
Adeonella based wholly on zoarial characters; in consequence he included in it a
miscellaneous series of species that must be divided among the several genera. Thus
his Adeonella distoma has a trypa and is one of the Microporellide, while others, such
as A. polymorpha, have no such pore and must belong to a different genus and family.

The subject has been attacked by Messrs. Hincks, Waters, and Macgillivray, and
each of these has advocated very different conclusions. Mr. Hincks [No. 8, pt. 1.
pp- 150-158, especially 155 & 157] has discussed the matterat length with the following
results: he maintains (1) that as Adeonella is based only on zoarial characters it is not
distinct from Adeona ; (2) the latter genus he places among the Microporellide, distin-
guished from Microporella by the substitution of goncecia for external ocecia ; (3) as he

242 MR. J. W. GREGORY ON THE

regards Adeonellopsis as based only on the possession of a peristomial pore he declines
to accept it.

Mr. Waters’s conclusions | No. 6, p. 294, and No. ro, pp. 3, 32, 33] are very different ;
he abandons Adeona as a synonym of Microporella, and speaks of the type species as
Microporella grisea, form Adeona; in his last essay he accepts Adeonella for torms
without a trypa but with a peristomial pore, the latter a character of very doubtful
value.

Mr. Macgillivray’s conclusions [No. 2, pt. ix. p. 134] seem to me more, though not
entirely, satisfactory. He accepts Adeonel/a in much the same sense as Mr. Waters ; but
he fully grasps the significance of the absence of the trypa and removes the genus
to the Mucronellinew (or Smittide). He agrees with Mr. Hincks and differs from
Mr. Waters in separating Adeona from Microporella owing to the absence of external
ocecia in the former; finally, he founds the genus Adeonellopsis for forms resembling
Adeonella, but without a peristomial pore.

Before proceeding to discuss these views I must again express my thanks to my
colleague Mr. Kirkpatrick for allowing me constant access to the recent species, and
especially to Busk’s type specimens, and also for the opportunity of frequent discussion
of all the points involved.

The first point to be decided is what are the true affinities of Adeonella. The first
species described by Busk was A. polymorpha, and this he seems to have regarded as his
type; Mr. Waters certainly includes it in the genus as restricted by him. Mr. Hincks
[No. 4, pt. xiii. pp. 294-296] has quoted Busk’s remark [No. 8, p. 183] that “as
regards the general zocecial characters there is no difference whatever between Adeona
and Adeonella.” ‘This remark seems to me quite inexplicable. Adeonella poly-
morpha has no trypa, which seems to be generally regarded as implying a difference in
family. A. polymorpha is therefore not one of the Microporellide at all, and cannot be
synonynous with Adeona, which has a trypa*. Macgillivray has clearly recognized
this, and has removed Adeonella to his Escharide. But this seems to me to be going
rather too far; in Adeonella polymorpha and all the species which seem to be con-
generic with it, the primary orifice is always schizostomatous, and therefore the
genus cannot enter the holostomatous group: its true affinities appear to me to belong
to the Schizothyriata allied to the Schizoporellide; the secondary orifice appears to
distinguish it from both the Schizoporellide and the Microporellide; the presence of
goneecia instead of external ocecia still further separates it from the Schizoporellide,
but allies it to its old associates of the Adeoninew. Its true position therefore appears
to be as a distinct family intermediate between the Schizoporellide and the Micro-
porellidz, with one link attaching it to each.

* Macgillivray’s figures of Adeona (Dictyopora) cellulosa show an occasional absence of the trypa [Macgil-
livray, No. 1, dec. y. pl. 47. fig. 1 a, 6]. A dissection of a specimen with the same feature shows that it is
duc simply to the trypa being overgrown and concealed by the avicularium.

BRITISH PALZOGENE BRYOZOA. 243

In regard to Adeonellopsis, it seems to me absolutely necessary to accept that or Busk’s
[No. 8, p. 178] Reptadeonella, as it is going rather far to place such a species as
“ Lepralia” violacea, Johust., in Adeona or Microporella. Reptadeonelia is prior by two
years, but it was based only on zoarial characters and was never properly diagnosed,
and I therefore prefer to accept Macgillivray’s better defined genus [No. 2, pt. ix.
p. 134, and No. 3, p. 210].

Reuss has described another species which it is necessary to consider in connection
with the Microporellide, as it possesses both a trypa and an oral sinus. The species
which shows this feature is of such interest in connection with the evolution of this
group that it is advisable here to diagnose the new genus necessary for its reception.

ScCHISMOPORELLA}, n. g.

Diagnosis. Zoarium lepralian or escharine.

Zoecia elongate, oblong (in known species). Aperture orbicular, with a large sinus.
The front wall has a zocecial pore.

Owcia external, globose.

‘Type species. Schismoporella schizogaster (Reuss)*, 1847, Helvetian, Austria.

This genus may at first throw doubt upon the assumed homology of the sinus and
trypa, the latter being regarded as the more specialized. The occurrence of Schismo-
porella may, however, be explained by a repeated formation of a sinus after the zooecial
pore has travelied well away from the peristome: or else the division of the trypa
into two or more pores has very frequently taken place, and there seems no impro-
bability in one of these parts persisting as a sinus. If neither of these explanations is
correct, then Schismoporella is probably a primitive form uniting characters now divided
between two families.

The following synopsis summarizes the classification of this group that is here

proposed :—

( Schizoporellide : schizostomatous. External occia.
Adeonellide : primary aperture schizostomatous. Goncecia.

Schizo- ( Adeonine : with ¢ Zoarium fenestrate, &c. Adeona.
thyriata. ‘ goneecia. { ,, foliaceous or encrusting. Adeonellopsis.
| Microporelide: | Microporellinee : = Microporella, Tessarodoma, &c.
\ trypa present. external marsupia.
| Schismoporelline : = Schismoporella.

| 5 O
\. with trypa and sinus.

1 From oxiopa, a slit, and épos, a pore.
2 Cellepora schizogaster, Reuss, 1847, No. 1, p. 84, pl. x. fig. 9; Molla schizogaster, D’Orbigny, No. 2,
p. 888 ; Lepraha schizogaster, Reuss, No. 14, p. 161, pl. iii. fig. 10.

VOL. XIll.—PART vi. No. 4.—June, 1893. 2N

244 MR. J. W. GREGORY ON THE

It may be objected that the genera of the Adeonine are based on zoarial
characters; but these are of such a marked description, and lead to such modifications
and dimorphism of some of the zocecia, that they seem certainly of generic value. In
regard to Adeona and Adeonellopsis there is the further difference of the presence of a
peristomial pore in the latter.

It seems also advisable to rediagnose Adeonella in accordance with this scheme, and
consider what species should be included in it. But at present the diagnosis of the
genus is the same as that of the family, as I am at present aware of only the one genus;
the diagnosis is therefore: “ Schizothyriata with a schizostomatous primary aperture
and a secondary orifice variable in form. Goncecia present, but no external marsupia.”
Before giving a list of the species I had better refer to the question of the value of the
peristomial pore, as if Busk [ No. 8, p. 167] and Ridley [No. 1, p. 47, pl. vi. fig. 6] are
right in considering it of generic importance, then Adeonella must be subdivided. Since
Mr. Kirkpatrick [No. 1, pp. 77, 78, pl. viii. fig. 5] has shown that this structure in
“« Gigantopora”’ lyncoides, Ridley [No. 1, p. 47, pl. vi. fig. 6], is only formed by the
avicularia, little value has been attached to it. Messrs. Hincks [No. 6, pp. 268, 269]
and Waters [No. g, p. 192] also dismiss it as valueless, as the bridge is not always
present in different zocecia of the same zoarium of Schizoporella biturrita, Hincks (or
S. tuberosa, Reuss).

List of Species of Adeonella.

‘Type. Adeonella polymorpha, Busk, No. 8, p. 183, pl.

Adeonella, ctr. polymorpha, Gioli, No. 1, pp. 261, 262, pl. xiv. fig. 8.

platalea, Busk, No. 8, p. 184.

intricaria, Busk, No. 8, p. 185.

regularis, Busk, No. 8, p. 186.

—— atlantica, Busk, No. 8, p. 186.

pectinata, Busk, No. 8, p. 189.

(This species has a large lyrula within the secondary orifice and hiding the primary aperture.
Busk has not figured the operculum, but its shape shows that the primary aperture is
schizostomatous. It ought, perhaps, to be separated as a subgenus.)

polystomella (Reuss), No. 1, p. 70, pl. viii. figs. 27, 28.

pallasi (Heller), No. 1, p. 115, pl. iii. figs. 1, 2? = A. polystomella.

dispar (Macgill.). For references see Jelly, No. 1, p. 259 (agrees with Adeonella, but has a
sinus also in the secondary orifice).

sulcata (M.-Edw.), Eschara sulcata, M.-Kdwards, No. 1, pp. 47-49, pl. v. fig. 2, non
Flustra sulcata, Lamouroux, No. 3, p. 609, pl. 92. figs. 3, 4.

Juegensis (Busk), No. 3, p. 90.

BRITISH PALAOGENE BRYOZOA. 245

List of Species that have been referred to Adeonella.

Adeonella distoma (Busk), No. 5, p. 127, pl. xviii. fig. 1 = Adeonellopsis distoma.

, var. imperforata (Busk), No. 8, p. 188 = Adeonellopsis.

(This form has a trypa, but it is covered over by the avicularian cell; by the kindness of
Mr. Kirkpatrick I have been enabled to dissect off an avicularium, and thus demonstrate
the presence of a trypa. The form is probably entitled to specific distinction, and I there-

fore record it as Adeonellopsis imperforata.)
Eschara pulchra, Stoliczka, No. 2, pp. 87, 88, pl. ii. fig. 10 = Adeonellopsis.

coscinophora, Reuss, No. 1, p. 67. = _
mucronata (Macgill.), No. 1, dec. v. p. 43. © = a3 (? coscinophora, Reuss).
Cellepora heckeli, Reuss, No. 1, p. 85. = heckeli (Reuss).
Lepralia violacea, Johnston, No. 2, p. 825. = Bx (? heckeli, Reuss).
Eschara lichenoides, Lamk., No. 1, p. 176. = ue
Microporella fissa, Hincks, No. 4, pt. ii. p. 381. = 2
Adeonella japonica, Ortmann, No. 1, p. 54. = # CpGecmy lees: m8.
; figure the opercula,
sparassis, Ortmann, No. 1, p. 54. = a Ag f
—— tuberculata, Ortmann, No. 1, pp. 53, 54. = a gas oe peserenee
\. may be incorrect.
Porina subsulcata, Smitt, No. 3, p. 29, pl. vi. figs. 136-140.= 5

Eschara syringopora, Reuss, No. 1, p. 68, pl. viii. fig. 23. ? = Teichopora.
ornatissima, Stoliczka, No. 1, pl. ii. fig. 7. Probably a Schismopora with a peristomial pore.

—— ciliata, Pallas, Elenchus, p. 38. = Microporella.
Flustra sulcata, Lamx., No. 3, p. 609, pl. 92. figs. 3,4. = ?
Cellepora imbricata, Lonsdale, No. 1, pp. 507, 508. = Adeonellopsis.

Family MICROPORELLIDZ:.

Subfamily ADEONIN&.
Genus ADEONELLOPSIS, Macgillivray.

Species 1. ADEONELLOPSIS WETHERELLI, 0. sp.
Syn. Flustra, sp., Wetherell, 1837, Trans. Geol. Soc. (2) iv. pl. ix. fig. 22.
Microporella violacea, var. fissa, var. b, Vine, 1889, Proc. Yorks. Geol. & Polyt. Soc. xi.
p- 162, pl. v. fig. 7 d.
Microporella violacea, var. fissa, var. a, Vine, 1891, ibid. xii. p. 61.

Diagnosis. Zoarium erect: branching dichotomous; bilaminar and either flat or
cylindrical branches.

Zoecia tumid: usually pyriform; irregular in form; elongate and ovate or sub-
hexagonal. Lower zocecia immersed. ‘The orifice is at the summit of a large raised
head, the peristome being somewhat tubular; the orifice is oval, lunate, or semi-

2N 2

246 MR. J. W. GREGORY ON THE

circular in shape. The front wall contains an elongate, depressed areola, the floor of
which is cribriform, being perforated by from 4 to 8 pores. A line of punctures runs
around the margin of the zocecia.

Avicularia large, pointing obliquely upwards: situated close below the peristome.

Gonecia sparsely scattered, low; aperture smaller than in the normal zoccia.

Distribution. London Clay: Fareham (abundant); Highgate; Haverstock Hill ;
Sydenham ; White Conduit House.

Figures. Pl. XXX. fig. 12. Part of a zoarium from the London Clay, Haverstock
Hill, x 3diam. Fig. 12 6. Several zocecia from the upper part of the same specimen.
Fig. 12¢. Zocecia from lower in the same specimen. Fig. 13. Zocecia from base of
another specimen.—Pl. XX XI. fig. 1. Another specimen.

Type. Brit. Mus. No. 49756, Edwards Coll.; Highgate. Wetherell’s figured speci-
men is B. M. No. B 4445.

Affinities. Wetherell found a minute fragment of this species in a well at Hampstead,
and gave a good but small figure of it; this, however, seems to have escaped subsequent
notice. Mr. Vine first described the species, and he regarded it asa variety of the well-
known recent species Adeonellopsis (Reptadeonella, Microporella, &c.) violacea
(Johnst.); from this, however, it differs very markedly in the nature of the avicularia,
the cribriform area, the subtubular peristome, &c. The species to which it is most
closely allied is Adeonellopsis distoma (Busk) ; from this the main difference is in the
avicularian orifice, which is much smaller in proportion to the size of the peristomial
orifice, and it is placed below the latter and not included within the rim, which
includes both the avicularium and orifice. In the London Clay species the avicularia
are always directed very obliquely upwards.

Busk has suggested that Reuss’s Kschara coscinophora is synonymous with
A. distoma; but agreeing with Mr. Waters [No. 6, p. 283, and No. 13, p. 162], who
records the latter from the Italian Upper Kocenes, I prefer to keep them distinct.
The London Clay species agree more with A. distoma than <A. coscinophora. The
specimens of the latter which agree most with our species are those from the Middle
Oligocene of Séllingen in Prussia, figured by Reuss [No. 7, p. 186, pl. xi. figs. 1-4]:
his figure 1 allows of a careful comparison of equivalent zocecia. The differences
between the species are that in A. wetherelli the avicularia are oblique or transverse
and much larger, the cribriform plate is larger and has more regular pores, and the
secondary aperture is more raised.

1 There seems some confusion as to the localities and horizons of Mr. Vine’s types of this species and the
next; the specimen figured as var. b (i.e. fig. 76) is recorded as from the Bracklesham Beds of the Isle of
Wieht ; the slide is, however, correctly labelled from the London Clay.

BRITISH PALHOGENE BRYOZOA. 247

Species 2. ADEONELLOPSIS INCISA, 1. sp.

Syn. Microporella violacea, var. fissa, var. a, Vine, 1889, Proc. Yorks. Geol. & Polyt. Soc. Xi.
p- 162, pl. v. figs. 7, 7 a.

Diagnosis. Zoarium erect, bilaminar, and forming thick, short, subcylindrical shoots.

Zoucia elongate, lanceolate, quincuncially arranged. The orifice is oval or suborbi-
cular ; it opens on the sloping upper surface of the high tumid head, which also bears
a large peristomial pore. A pair of large avicularia occur immediately below the
orifice. The zocecia are sharply defined by lines of depression marked by rows of
areole. The trypa is a median narrow slit.

Gonecia

Distribution. London Clay : Haverstock Hill (! Bracklesham Beds, fide Vine).

Type. Brit. Mus. No. 49661.

Figure. Pl. XXX. fig. 11. Part of Mr. Vine’s type.

Affinities. This species in its slit-like trypa closely resembles Adeonellopsis perforata
(Reuss) [Eschara perforata, Reuss, No. 11, p. 231, pl. xxxiii. fig. 5], but the latter
has no peristomial pore. It differs from A. wetherelli, Greg., by the pair of avicularia

forming a peristomial pore, instead of having one median avicularium; the trypa is
also different. The specimen figured by Reuss [No. 7, pl. xi. fig. 6] as Hschara diplo
stoma, Phil., also belongs to this genus, but differs in the form of the trypa and of the
orifice. ‘The two other forms (figs. 5 and 7) associated with it by Reuss seem different,
and that represented in fig. 7 is probably a second species of Schismoporella.

Suborder HOLOTHYRIATA.

Family LEPRALITD&.
Subfamily LuEPRALIIN A.
Genus Lepratia, Hincks, 1880 (non Johnst. &c.).

Diagnosis. Hincks, No. 2, p. 297.

Species 1. LEPRALIA LONSDALEI.
Syn. Eschara brongniarti, pars, Lonsdale (non M.-Edw.), 1850, Dixon, Geol. Suss. pp. 161, 162,
Plone Oey.
Diagnosis. Zoarium thick, encrusting.
Zoecia small, ovate; very irregularly distributed. Form irregular, varying from
somewhat elongate to short and round. Aperture lepralian, very large : lower margin
straight or curved outwards ; the lateral constriction is, however, very slight. Surface

248 MR. J. W. GREGORY ON THE

granular. Zocecia separated by deep depressions. A line of areole around the margin;
these vary with the size of the zocecia from 10 to 16 or 18.

Avicularia: usually one; lateral, placed close beside the orifice.

Distribution. Bracklesham Beds, Bracklesham Bay.

Type. Brit. Mus. No. 49734, Edwards Coll. (Encrusting.)

Figure. Pl. XXXII. fig. 2. - x 55 diam.

Affinities. The shape of the orifice shows that this species is a true Lepralia, using
that term, of course, in its modern restricted sense. It was figured by Lonsdale as
Eschara brongniarti, a mistake due to his having failed to separate it from the
Bryozoan on which it is encrusting. A comparison of his figure 9* with his figure 9
shows that he has included two different forms under one name.

Among the species of Lepralia it most resembles Lepralia angiostoma, Reuss [No. 11,
pp- 291, 292, pl. xxx. fig.3], but it may be distinguished by the smallness of the orifice
in the Austrian species.

Genus Umsonuvta, Hincks.

Diagnosis. Hincks, No. 2, pp. 816 and cxxxviii.

Species 1. UMBONULA BARTONENSE, 0. sp.

Diagnosis. Zoarium adnate, encrusting ; forming a large and fairly thick crust over
shells.

Zoecia very crowded: quincuncially arranged; small, pyriform. Peristomial
aperture semicircular or slightly clithridiate ; lower margin straight; very large, some-
times occupying more than half the front of the zowecium. The front wall is occupied
by a large umbo, formed on an avicularian cell, the opening of which is just below the
aperture and is hidden by the prominence of the umbo. Around the margin of the
zocecium runs a line of areole ; those of the lower half are large, and from them furrows
run some way up towards the umbo ; the areole are small around the aperture.

Avicularia and external marsupia none.

Distribution. Barton Beds, Barton.

Type. Brit. Mus. No. 49741.

Figure. Pl. XXXI. fig. 4. Portion of zoarium, x 55. diam.

Gottardi’s Eschara prominens [No. 1, pp. 306, 307, pl. xiv. fig. 4] probably belongs
to this genus, but the species is so diagrammatically figured that I cannot be quite sure.
The genus is a convenient one, though, as Mr. Waters has pointed out, it is a very

close ally of Lepralia, and perhaps ought not to be separated from it. The aperture in
this species is typically lepralian.

BRITISH PALZOGENE BRYOZOA. 249

Species 2. UMBONULA CALCARIFORMIS!, n. sp.

Diagnosis. Zoariwn, a thick encrusting mass.

Zoecia roughly hexagonal in shape: short and thick. The front wall granular.
The aperture is suborbicular, somewhat irregular on the lower margin from the
ingrowth of the avicularia. The front wall is very tumid, and bears a large avicularian
cell, this is attached to the front wall and continues over it as a pair of sharp pointed
processes. The pore of the avicularium is raised and close beside the zocecial aperture.
The avicularium is always lateral and oblique.

Ocecia globose, low, and comparatively small.

Distribution. London Clay, Fareham.

Type. Brit. Mus. No. B 3831. (Growing on Hornera.)

Figure. Pl. XXXI. fig. 3. X 95 diam.

Affinities. The shape of the aperture, the tumid front wall, and the umbonate avicu-
larium all agree with the genus Umbonula. Amongst the other species it probably is
most nearly allied to U. bartonense, Greg., but from this it may be distinguished by its
suborbicular aperture, the lateral position of the avicularia, and the pores of these
being terminal instead of in the angle overhung by the umbo.

Subfamily TEICHOPORIN &.
Genus TEICHOPORA?, n. g.

Diagnosis. Zoarium foliaceous or encrusting: in large flat surfaces.

Zoecia pyriform, much expanded above; elongate. Aperture large, holostomatous,
orbicular ; surrounded by a raised ring. Usually long sinuous lines of marginal areole
continuous across successive zocecia.

Gonecia with the aperture contracted either marginally or by a bar or a central
spot.

Species 1. THICHOPORA CLAVATA.

Diagnosis. Zoarium in large foliaceous expansions.

Zoecia clavate, the lower part being much restricted in width. Orifice very large ;
the surrounding ring is continuous with the ridge on the front wall of the zoccia.
Punctures large and numerous.

Avicularia: usually one, just below the orifice; lateral.

* From the spur-shape of the avicularian cell.
* From retxos, the wall of a fortress.

250 Mk. J. W. GREGORY ON THE

Gonecia irregularly scattered. Orifice much restricted, either at the margin or by
the central calcareous plate, the knob of which has a small central pit.

Distribution. Barton Beds, Barton.

Type. Brit. Mus. No. 49733. Edwards Coll.

Figures. Pl. XXXI. fig. 5. Part of zoarium of the type. Fig. 6. Basal zocecia.
Fig. 7. Part of a large specimen (B. M. No. 49757) showing gonecia.

Remarks on the Genus. This isa Lepralidan with a simple orbicular aperture and
thickened peristome, and goneecia instead of external marsupia. The last character as
well as the form of the aperture distinguish it from Lepralia; the absence of a
secondary orifice separates it from the Smittide.

Affinities of the Species. The nearest ally of 7%. clavata is a specimen from the
German Oligocene, described by Stoliczka [No. 1, p. 87, pl. ii. tig. 8] as Eschara crena-
tula, trom which it differs by its plain margin. Lschara semitubulosa (Reuss) { No. 11,
p- 272, pl. xxxili. fig. 3] probably also belongs to Teichopora, though in the absence of
knowledge as to the ocecial characters one cannot be quite sure: the greater length of
the zocecia and the more uniform width of the Austrian species clearly distinguish
it. Mr. Waters has suggested that H. semitubulosa is a synonym of Reuss’s earlier
species, HL. syringopora [No. 1, p. 68, pl. viii. fig. 23, and No. 11, p. 269, pl. xxxii. fig. 1];
but in the latter the orifice is smaller, the zocecia expanded below, and the front wall
has a long furrow instead of being tumid and solid. Mr. Waters’s own figures [ No. 12,
p. 20, pl. iii. figs. 2-4] more resemble the English species, though the different nature
of the closure, the general form of the zocecia, and the proportions of the orifice show
them to be distinct.

Amongst other species that will probably prove to belong to this species are Eschara
stipitata (Reuss, MS. Manzoni) [No. 3, p. 60, pl. xii. fig. 3], and Eschara sulcimargo,
Reuss [No. 1, p. 58, pl. v. fig. 18].

Genus Meniscoporal, n. g.

Diagnosis. A genus of Lepraliide with a simple primary orifice, usually biconvex in
shape, with the lower margin a much flatter curve than the upper. Goncecia and no
external marsupia.

Affinities. This genus differs from Zeichopora by the shape of the orifice and the form
of the zovecia. From most other Lepraliide it differs by the presence of goncecia; when
this cannot be determined, it may be distinguished from Lepralia (sensu stricto) by the
form of the aperture, and from Umbonula by the absence of the umbo: these are the
two genera which it most resembles in general aspect.

* From pnvioxos, a lune, referring to the shape of the orifice.

BRITISH PALAOGENE BRYOZOA. 251

Species |. MENISCOPORA BIGIBBERA!.

Syn. Eschara brongniarti, Lonsdale, 1850 (non M.-Edw.), in Dixon, Geol. Suss. pp. 161, 162,
pl. i. fig. 9.

Diagnosis. Zoarium erect, bilaminar ; forming large flat foliaceous expansions.

Zoecia regularly quincuncial in arrangement. Surface plain. Shape pyriform.
Aperture large, semicircular, but with the lower margin somewhat curved outward.
The main part of the front wall is a raised triangular area; at the upper part are two
prominent humps. ‘The zocecia are separated by depressed furrows; a line of large
round punctures occurs along the margin.

Avicularia: one large marginal pair beside the lower corners of the aperture ; man-
dible pointing outwards.

Distribution. Bracklesham Beds, Huntingbridge.

Type. Brit. Mus. No. 49752. Edwards Coll.

Figures. Pl. XXXT. fig. 8. Part of the type from Huntingbridge. Fig. 9. Fragment
with goneecium.

Affinities of the Species. As this species was identified by Lonsdale with M.-Edwards’s
Eschara brongniarti it is necessary to compare the two: the fact that the English
species has the aperture wider than long, has two humps and a raised triangular
area of front wall, is quite sufficient to distinguish them. M.-Edwards’s figure [No. 2,
p. 335, pl. xi. fig. 9] leaves the generic position of his species quite uncertain; but even
should it prove to be a Meniscopora, which is not probable, there need be no confusion
between the species.

This species has a certain resemblance in general aspect to the Eschara fenestrata,
Reuss [No. 11, p. 290, pl. xxxii. fig. 5], which Waters [No. 12, pp. 18, 19] regards as
asynonym of Lepralia bisulca (Reuss) [No. 11, pp. 270, 271, pl. xxxii. fig. 10]; but
the latter has external occia.

Family CALLEPORID.
Genus CoNESCHARELLINA, D’Orbigny, 1851.

[D’Orbigny, No 2, pp. 446, 447. ]
Syn. Batopora, Reuss; Fedora, Jullien.

Diagnosis. A genus of Celleporide with a small, conical, hemispherical, or spherical
free zoarium ; uni- or multi-laminate. The zocecia are holostomatous; the aperture is
usually on the highest part of the front wall, and is generally orbicular or clithridiate.
Ocecia large and globose ; comparatively rare.

Distribution. Recent. S. Atlantic, Australia.—Fossil. Kocene: England. Oligocene:
Germany, Austria, Italy. Miocene: Austria.

‘ I. e. two-humped, referring to the prominences on the front wall.
VOL. XI11.—Part vi. No. 5.—June, 1893. 20

252 MR. J. W. GREGORY ON THE

Species 1. ConeSCHARELLINA CLITHRIDIATA, Ni. sp.

Syn. Cellepora, sp., Wetherell, 1837, Trans. Geol. Soe. ser. 2, vol. v. pl. ix. fig. 21; Busk, 1866,
Geol. Mag. vol. i. p. 301.

Cellepora sp. (pumicosa?), Vine, 1890, Proc. Yorks. Geol. & Polyt. Soc. xi. p. 164.
Diagnosis. Zoarium a small, thick, globular mass; base contracted (! attached).
Zoecia few in number and irregularly distributed and not arranged around a central

cell. The apertures of the zocecia are clithridiate in shape; they are large and terminal.
The zocecia are tumid and generally hexagonal in outline ; the front walls are granular
and steep; the zocecia are separated by deep depressions.

Oecia very large in proportion to the size of the zocecia ; globose, tumid, overhanging
the aperture. Only rare zoaria show them, but then they are numerous.

Distribution. London Clay: Highgate, Sydenham, &c.

Type. Brit. Mus. No. B 1357 (Wetherell’s specimen, No. 69554).

Dimensions. The largest zoarium is | mm. in diameter.

Figures. Pl. XX XI. fig. 10. A zoarium from the London Clay, Hampstead ; Brit.
Mus. No. 69554.—Fig. 11. A zoarium from Sydenham with ocecia.

Affinities. A charming little figure of a specimen of this species has been given by
Wetherell. The species, however, was not named and it has been missed by all subse-
quent workers. His specimen is in the British Museum collection, along with a great
number from the London Clay at Sydenham. This species belongs to the group of
which Cellepora globularis, Bronn [No. 2, p. 654], was the first described species; as
Reuss | No.9, pp. 118, 114] has, however, pointed out, several distinct forms have come
in time to be included under this name. ‘The specimen recently figured by Gioli
[No. 1, pp. 263, 264, pl. xiv. fig. 9] appears to be quite distinct. Pergens’s short
synonymy [No. 4, p. xvi] shows much discrimination.

The nearest ally of this species, C. scrobiculata (Koschinsky) [No. 1, p. 63, pl. vi.
figs. 2, 3], has a hemispherical or conical zoarium, the base being expanded instead of
contracted as in all the English specimens ; the aperture in the Bavarian species is also
circular and surrounded by a rim. ‘The new species differs from C. multiradiata, Reuss
[No. 11, p. 265, pl. xxxi. figs. 1-4, and Waters, No. 12, pp. 32, 33], as in that the
zocecia are barrel-shaped, the apertures flush with the surface of the zoarium, and it is
composed of several layers; the occia also are much larger. From the Miocene
Conescharellina rosula (Reuss) [No. 1, p. 78, pl. ix. fig. 17, and Manzoni, No. 3, p. 54,
pl. il. fig. 6] the London species differs in its clithridiate aperture and the less elevated
zoecia. The same characters also separate it from C. stoliczkai (Reuss) [No. 10,
pp- 223-226, pl. ii. figs. 2-4].

In agreement with the zoologists I accept the name Conescharellina in preference to
Reuss’s Latopora, which has been adopted by most paleontologists. There seems
little room for doubt as to the identity of the two. D’Orbigny’s genus was diagnosed

BRITISH PALMOGENE BRYOZOA. 253

exceptionally well for D’Orbigny, and its claims cannot be so quietly set aside as Reuss
has done in the two lines in which he refers to it. Batopora is the better name, but
that is of course a mere matter of detail.

Genus OrpiruLipora, Stoliczka, 1862.

[Stoliczka, No. 1, p. 90.]

Type species. O. haidingert, Stol. op. cit. p. 91, pl. iii. fig. 5.

Diagnosis. A Celleporidan with a bilaminar zoarium composed of a flat round disk
supported laterally by a short stem. The zocecia of the disk are usually arranged
around a small central zocecium. The zocecia are holostomatous, with a large and
typically orbicular aperture. The ocecia are narrow, but globose and elevated.
Small avicularia and vibracula may or may not occur. ;

Species 1. ORBITULIPORA PETIOLUS (Lonsdale), 1850.

Syn. Cellepora? petiolus, Lonsdale, 1850, Dixon, Geol. Suss. p. 151, pl. i. fig. 10; Morris, 1854,
Cat. Brit. Foss. ed. 2, p. 120; Mourlon, 1881, Geol. Belg. pp. 180, 191, 202; Vine,
1890, Proc. Yorks. Geol. & Polyt. Soc. xi. pp. 163, 164, pl. v. fig. 10; Reuss, 1867, Sitzb.
k. Ak. Wiss. Wien, Bd. lv. Abth. 1, p. 217.

Diagnosis. Zoarium: disks rather large for this genus; thick at the margins and
depressed in the centre. The stem is short and, so far as known, unjointed ; when
broken away it leaves a large round scar.

Zoecia numerous; usually in fairly regular radial rows; the apertures are orbicular
in the centre, but become elliptical at the margin; those adjoining ocecia have the
margin nearest incurved owing to the overgrowth of the ocecium. Separated by inter-
spaces which are often marked by punctures.

Occia very irregularly distributed ; sometimes absent from the whole of one surface
of a disk, at others there are a few irregularly scattered, at others nearly every zocecium
has one. They are globose, but narrow.

Distribution. Bracklesham Beds: Bracklesham, Bramshaw, Brook, Whitecliff Bay
(common).—oreign. Belgium: Bruxellien, Laekenien, Wemmelien, and Tongrien.

Type. Brit. Mus.

Figures. Pl. XX XI. fig. 12. Zoarium, x4 diam. Fig. 12a. Part of the same,
x 18 diam., to show the occia. Fig. 13. Another specimen, to show the stem.
Fig. 14. A young specimen in the Conescharellidan stage.

Affinities of the Species. This species differs from O. haidingeri mainly by the fact
that the peripheral zocecia open upwards instead of outwards, a point well seen in a
comparison of Stoliczka’s and Lonsdale’s figures. 0. haidingeri is the nearest ally of
the English species; if the two species should prove to be identical, Lonsdale’s name
will have the prior claim to adoption.

202

254 MR. J. W. GREGORY ON THE

Affinities of the Genus. The British Museum contains a large number of specimens
of this species, and these well show its range. One of the smallest specimens, having a
zoarium barely 1 mm. in diameter, is of interest as showing that this genus passes
through a Conescharellina (or Batopora) stage ; the small central zocecium is surrounded
by an irregular series of others having the tumid forms, granular walls, and terminal
apertures of that genus. This therefore shows that Conescharellina, and especially
such a species as C. clithridiata, is a more primitive form than Orbitulipora with its
remarkably specialized zoarium.

The species is also clearly distinct from O. lenticularis, Reuss [No. 11, p. 289,

opinion from Reuss’s figures.

Family SMITTID/.

Genus Mucronewua, Hincks, 1880.

Diagnosis. Hincks, No. 2, p. 360.

Species 1. MUCRONELLA ANGUSTO@CIUM, n. sp.

Syn. Porella concinna, var. eocena, G. R. Vine, 1891, Proc. Yorks. Geol. & Polyt. Soc. vol. xii. p. 61.

Diagnosis. Zoarium: unilaminar flat surfaces (? erect or encrusting alge).

Zo«cia irregular, but with a tendency towards a disposition along radial branching
lines. Shape approximately hexagonal. ‘The zocecia are tumid, rising from a flat
surface. Orifice suborbicular: the peristome is high and thickened, especially on the
lower margin; it here bears a small simple mucro. The thick bases of a pair of
marginal spines occur on the lower angles of the orifice. The thick lower lip has a
distinct median transverse depression. Surface granular. Zocecia separated. About
a dozen areole occur around the lower half of the zocecia.

Owcia numerous, granular, globose, but narrow. In one case there are two ocecia to
one zocecium.

Avicularia: none.

Distribution. Barton Beds, Barton ; London Clay, Fareham.

Type. Brit. Mus. No. 49739. Edwards Coll. From Barton.

Figures. Pl. XX XI. fig. 15. Barton Beds. Brit. Mus. No. 49739. Fig. 16. Part of
a zoarium from the London Clay, Fareham.

Affinities. This species reminds one at first sight of the common recent Mucronella
ventricosa (Hass.), and it clearly belongs to the same group; it differs from that species,
however, by the small simple mucro, the narrow instead of elongate ocecia, the position
of the marginal spines, and in less important points. Probably its nearest ally is
M. hérnesi (Reuss)', of the Middle Oligocene; the new species, however, may be
distinguished by its low instead of elongate ocecia. In this character it most resembles

* Lepraha hornesi, Reuss, No. 8, pp. 633, 634, pl. xiii. fig. 5, and No. 7, pp. 173, 174, pl. vii. fig. 12,

BRITISH PALAOGENE BRYOZOA. 255

M. chilopora (Reuss)!, but the general form of the zocecia and the structure of the
mucro are quite distinct in the two species.

Mr. A. Bell’s collection of Fareham Bryozoa having recently passed into the posses-
sion of the British Museum, I am able to identify with this species the specimen
referred to by Vine as Porella concinna.

Mr. Waters, in his ‘ Revision of the North Italian Bryozoa,’ does not quote Mucro=
nella from the Eocene deposits of that country. The genus occurs in the Austrian
Leithakalk (Helvetian), as at least two species, VM. serrulata (Reuss)? and WM. tenera
(Reuss) *, seem referable to it.

Mr. Waters [No. 11, pp. 14, 15] has shown that under the name “ mucro” several
distinct structures have been confused together, and he has proposed the dismemberment
of Mucronella and the incorporation of most of its species in Smittia. The generic
value of variations in the secondary orifice and its peristomial tube certainly appears
very doubtful, but there does seem sufficient difference between this group of species of
Mucronella and normal Smittie to justify the limitation and retention of Mr. Hincks’s
too comprehensive genus.

Genus Smirria, Hincks, 1880.
Diagnosis. Hincks, No. 1, p. 340.

Species 1. SmIrrra TUBULARIS 4, n. sp.

Diagnosis. Zoariwm erect; narrow cylindrical or shoot-like branches; branching
dichotomous.

Zoecia arranged alternately. Shape pyriform; ovate or elongate-ovate. Front wall
tumid; surface granular. Secondary orifice orbicular or a distinct spout-like depression
often shown on lower margin. Peristome thin. A row of large areole occurs around
the margin.

Owcia small, flattened, the lower side covered by the upper margin of the secondary
orifice.

Avicularia large, lateral, on a prominent tubercle obliquely below the orifice.

Distribution. London Clay, White Conduit House.

Type. Brit. Mus. No. 49744. Edwards Coll.

Figures. Pl. XXXII. fig. 1a. Zoarium, nat. size. Fig. 1. Several zocecia, enlarged.
Fig. 1 ¢. Basal zocecia.

* Cellepora chilopora, Reuss, No. 1, p. 91, pl. xi. fig. 4, and No. 14, p. 168, pl. iv. fig. 1.

* Cellepora serrulata, Reuss, No. 1, p. 85, pl. x. fig. 12; and Lepralia serrulata, Reuss, No. 14, p. 167, pl. ii.
figs. 2, 3 (? pl. iv. fig. 4).

® Lepralia tenera, Reuss, No. 14, p. 167, pl. ii. fig. 4, pl. iii. fig. 11.

4 Referring to the subtubular orifice.

256 MR. J. W. GREGORY ON THE

Affinities. This appears to be a very well-marked species, with its elevated
peristome, its tumid front wall, and its large lateral avicularia and marginal punctures.
The secondary orifice is so raised and subtubular that it first seemed that the species
belonged to Porella (or Tessarodoma) ; but its secondary orifice and external avicularia
show that the resemblance is superficial and that it is truly a Smittia. Its mode of
growth, however, is exactly that of Tubucellaria; it lacks, however, the peristomial pore
of that genus, and the peristome is not so raised. It is not improbable that some of
the specimens figured as fossil forms of ZT. opuntioides (Pall.) may belong to this
species. Such may be the specimen figured by Michelin [No. 1, pl. 46. fig. 21] as
Vincularia fragilis, Defr., and some of Reuss’s Cellaria michelini.

Smittia is well known in the Continental Upper Eocene and Oligocene!, but none
of the species with which I am acquainted sufficiently resemble this one to necessitate

a comparison.

Order CYCLOSTOMATA.

Family IDMONEID.
Genus IpmonnA, Lamouroux, 1821.

[Lamouroux, No. 2, p. 80.]

Diagnosis. Pergens, No. 3, p. 342.
Type species. Idmonea triquetra, Lamx. No. 2, p. 80, pl. 79. figs. 13-15.

Species 1. IpMoNEA GIEBELI, Stoliczka, 1862.
Syn. Idmonea (Tubigera) giebeli, F. Stoliczka, 1862, Ohg. Bry. Latdf., Sitzb. k. Ak. Wiss. Wien,
Bd. xlv. p. 81, pl. i. fig. 6; F. Schreiber, 1872, Bry. Mittelolig. Griinsand Magdeburg,
Zeit. f. gesammt. Naturwiss. Bd. xxxix. p. 479.
Idmonea giebeliana, F. Stoliczka, 1865, Foss. Bry. Orakei Bay, Novara Reise, Geol. Theil, Bd.
i. Abth. ii. Pal. p. 115, pl. xviii. figs. 4-6; F. W. Hutton, 1880, Man. New Zeal. Moll.

Coll. Mus. Geol. Surv. N.Z. p. 196.

Diagnosis. Zoarium cylindrical, straight, erect branches; mode of branching unknown.
The back of the zoarium is a full flat curve ; the front is well raised.

Zowcia in series of five ; one formsa median row, on each side of which are two pairs
placed on a line a little above the central zocecium. The outermost zocecia are the
longest, but only slightly exceed the others. The walls are granular. Peristome
entire, even.

Oecia small, replacing one of the median zocecia.

1 See e.g. Waters, No. 12, pp. 21, 22.

BRITISH PALAOGENE BRYOZOA. 257

Distribution. London Clay, Haverstock Hill.—Foreign. Oligocene: Latdorf, Magde
burg, &c., Germany. Paleogene: New Zealand.

Type. Brit. Mus. No. 49656.

Figures. Pl. XXXII. fig. 3.4. Part of zoarium, including an ocecium. Fig. 3.
Transverse section.

Affinities. Busk has divided the genus Jdmonea into two groups: in one the
zocecia all open in two lateral groups and the two innermost ones are the longest ; in
the second, corresponding to the genus Tervia of Jullien, the outermost are the longest
and between the lateral series there are some zocecia irregularly scattered. A third
group may, however, be added, including species, such as the present, in which the
outermost zocecia are the longest, but in which there is only a single median row of
zocecia, and the lateral series are opposite.

I am aware of the existence of only six specimens of Jdmonea from the London Clay ;
two of these are quite unrecognizable internal pyritous casts, one of which is identified
by Mr. Vine as [dmonea coronopus, Defr., and the other as J. gracillima. A specimen
which Mr. Vine tells me is that figured by him as the former is now in the British
Museum Collection, but it is labelled, and correctly so, from the London Clay of
Sheppey. Mr. Vine [B, p. 165, pl. v. fig. 12] has figured a third specimen also as
Idmonea gracillima, Reuss, but it is an Entalophora. The remaining three small
specimens belong one to each of these three groups of Jdmonea. ‘This helps one to
realize that the British Eocene Bryozoan fauna was a singularly diversified one.

Lonsdale [No. 2, pl. ix. fig. 24] has also figured a specimen as Jdmonea coronopus,
but the figure is unrecognizable and I have not been able to find the specimen.

The only noticeable difference between the London Clay specimen and the type
figure is in the greater length of the zocecia in the former; but that may be only due
to the fragments having come from a different position in the zoaria. The New
Zealand specimen is more doubtful ; Hutton quotes it, but Waters, in his paper on the
New Zealand Cyclostomata [No. 8, pp. 337-350, pl. xviii.], does not refer to it.
Miss Jelly [No. 1, pp. 118, 119] makes it a synonym of J. milneana, D’Orb., but I fail
to see why it should be included with this rather than any other species of the genus.

Species 2. IDMONEA BIALTERNATA, Nl. sp.

Diagnosis. Zoarium sinuous, in thin elongated branches, evenly rounded in front,
with a flattish curve at the back.

Zoecia of medium length, thick, with large apertures ; walls granular. They are
arranged in two pairs; each pair open close together; the two pairs are placed alter-
nately. Peristome thick, plain.

Ocecia:% a small dilatation at base of the mner zocecia.

Distribution. London Clay, Islington.

Type. Brit. Mus. No. 49662.

258 MR. J. W. GREGORY ON THE

Figures, Pl. XXXII. figs. 2a, 26, Zoarium and section,

Affinities. This species belongs to the first of the groups of /dmonea, including
those with the zocecia all in lateral series. It most closely resembles a specimen
figured by Manzoni [No. 4, p. 5, pl. iii. fig. 10] as 2. cavinata?, Rom. A comparison
with the figures both of Romer [No.1, p. 21, pl. v. fig. 20] and Reuss [ No.1, pp. 44, 45,
pl. vi. fig. 27] would seem to show that the query after the identification was very well
founded ; in the number of zoccia, the shape and structure of cross-sections, and other
points, Manzoni’s figures markedly differ from those of the larger pluriserial triangular
species figured by Romer from the North-German Chalk. From the typical . carinata
the London Clay species can be very readily distinguished.

Idmonea reticulata, Reuss [No. 11, pp. 281, 282, pl. xxxiv. fig. 13], belongs to the
same series, but differs in the smaller size and more regular arrangement of the zocecia
and apertures, which are grouped in triplets instead of pairs. ‘The same characters also
separate the new species from the J. laticosta, Mars. [No.1, p. 29, pl. ii. fig. 11], of
Danian age, which belongs to the same group.

Species 3. IDMONEA SERIATOPORA, Reuss (?).

Syn. Jdmonea seriatopora, Reuss, 1847, Foss. Polyp. Wiener Tertiirbeckens, p. 46, pl. vi. fig. 32;
Manzoni, 1878, Brioz. foss. Mioc. Austr. Ungh., Denk. k. Ak. Wiss. Wien, Bd. xxxviii.
Abth. 2, p. 6, pl. vi. fig. 12.

Diagnosis. Zoarium of thick irregular branches, composed of many zocecia, well
rounded at the back.

Zoecia very irregularly arranged, the lateral ones the longest. There are no regular
series arranged on either side of a medial line. Three zocecia often open in an oblique
line.

Peristome elliptic ; border irregular.

Distribution. London Clay, Haverstock Hill.—Foreign. Leithakalk (Helvetian),
Austria.

Type. Brit. Mus. No. B 4510.

Figures. Pl. XXXII. fig. 4a. Part of a zoarium, X 18 diam. Fig. 46. Mouth,
X 32 diam. Fig. 4c. Transverse section, X 18 diam. Fig. 5. Back view of zoarium.

This species belongs to the subgenus Tervia of Jullien.

Affinities. The irregular distribution of the zocecia of this species reminds one of
I. compressa, Reuss [No. 1, p. 46, pl. vi. fig. 22], but the zoarium is not so laterally
compressed. Its closest ally is Jdmonea seriatopora, Reuss, as figured by Manzoni
[No. 4, pp. 6, 7, pl. ii. fig. 8, pl. v. fig. 17]; to the original and no doubt
diagrammatic figure of Reuss it has a less decided resemblance. But the London Clay
specimen is not sufficiently large to allow of a more definite comparison; hence I do
not feel able positively to affirm the occurrence of the Austrian Miocene species in the

BRITISH PALZOGENE BRYOZOA. 209

English Eocenes. Among the species which M. Jullien [No. 2, p. 501, pl. xvii. figs. 72,
73] has referred to his genus Tervia it most resembles Tervia solidula.

Species 4. IpMonEA coronopus, Defrance, 1821.

Syn. Jdmonea coronopus, Defrance, 1821, Dict. Sci. Nat. t. xxii. p.565 (non Atlas, pl. xlvi. fig. 2, as
stated by Bronn) ; Blainville, 1830, ibid. t. lx. p. 385; id. 1834, Man. d’Actinol. p. 420;
Milne-Edwards, 1836, in Lamarck, Anim. sans Vert. ed. 2, t. ii. pp. 281, 282; id. 1838,
Mém. Crisiées, Ann. Sci. Nat. Zool. sér. 2, t. ix. pp. 215, 216, pl. xii. fig. 3; Michelin,
1844, Icon. Zooph. p. 172, pl. xlvi. fig. 16; Bronn, 1848, Index Paleeont. Nomencl. p. 606;
Lonsdale, 1850, in Dixon, Geol. Sussex, pp. 153-155, pl. ix. fig. 24; Hagenow, 1851, Bryoz.
Maastr. Kreidebild. p. 25; Lonsdale, 1878, in Dixon, Geol. Sussex, ed. 2, pp. 204-206,
pl. ix. [10] fig. 24; Harris and Burrows, 1891, Eoc. and Oligoc. Paris Basin, p. 61.

Retepora trigona, Morren, 1828, Desc. Corall. foss. Belgio, Ann. Gron. p. 37, pl. x. figs. 1-3
(identification fide Michelin) ; Galeotti, 1838, Mém. Géogn. Brabant, p. 187, pl. iv. fig. 13;
Nyst, 1844, Coq. et Polyp. foss. Terr. Tert. Belg., Mém. Cour. R. Ac. Belg. t. xvii.
pp- 619, 620.

Chrysisina coronopus, Mourlon, 1881, Géol. Belgique, t. i. p. 180.

Hornera flabelliformis, Vine (non Blainv.), Proc. Yorks. Geol. & Polyt. Soc. vol. xi. p. 166,
pl. v. fig. 15; id. ibid. vol. xii. p. 53.

Diagnosis. Zoarium small, erect, rising from an encrusting, expanded base. The
branches fork several times; they are triangular in section and well rounded behind ;
they end bluntly.

Zowcia in short transverse series, alternately arranged. ‘The zocecia are single at the
base, but rapidly increase to rows of four; this decreases to three above. The inner-
most zocecia are the longest.

Peristome even, usually oblong with rounded angles; younger and isolated zocecia
have oval or even circular apertures.

Wall granular.

Distribution. British: Bracklesham Beds, Bracklesham (Brit. Mus., Dixon and
Vine Collections).—Voreign: Calcaire grossier, Parnes, Grignon, Chaumont, &c.;
Laekenien ; Uccle (near Brussels), de Forét, d’Assche.

Figures. Pl. XXXII. figs. 6a, 6d.

Affinities and Differences. As this species belongs to the typical group of Zdmonea it
clearly differs from Idmonca (Tervia) seriatopora, Reuss. As there is no median line
of zocecia it differs from Zdmonea giebeli, Stol. From the third British Eocene species
it may readily be distinguished, as in that the zocecia are always in alternate pairs.

Genus Hornera, Lamouroux, 1821.
{Lamouroux, No. 2, p. 41.1
Diagnosis. Pergens, 1889, No. 3, p. 353.
Type species. Hornera frondiculata (Lamarck), 1816, No. 1, pp. 182, 183.
VOL. XII.—PART v1. No. 6.—June, 1893. 2P

260 MR. J. W. GREGORY ON THE

Species 1. HorNERA FAREHAMENSIS, 0. Sp.

Syn. Hornera ramosa, D’Orb., G. R. Vine, 1891, Proc. Yorks. Geol. & Polyt. Soc. xii. pp. 54-56.

Diagnosis. Zoarium thick, dichotomously branching tufts; the branches do not
anastomose.

Zowcia open somewhat regularly on the anterior side; the orbicular apertures form
straight lines around the branches. In the middle line there is often an irregular and
crowded series. The apertures are flush. The interzocecial pores are of medium size,
but not very abundant, numbering from twice to thrice as many as the zoecia. The
posterior side of the zoarium is deeply perforate, the punctures occurring in simple
series, occasionally branching.

Distribution. London Clay, Fareham.

Type. Brit. Mus. No. B 3831.

Figures. Pl. XXXII. figs. 7-9.

Affinities. ‘This species has been identified by Mr. Vine as H. ramosa, D’Orbigny
[No. 2, pp. 937, 938, pl. 608. figs. 6-10, pl. 773. figs. 1-3]; from that species it appears
to me to differ by the following characters: (1) the sections of the branches are round
and not subtriangular; (2) the central series of zocecial apertures are very irregularly
distributed ; (3) the species figured by D’Orbigny has the exceptional character of a
series of tubular prominences probably zocecial (see pl. 773. fig. 2); (4) the zoarium is
irregularly branched and does not form the cupuliform structure shown by D’Orbigny
(pl. 608. fig. 6).

The nearest ally of this species appears to me to be Hornera concatenata, Reuss
(No. 11, pp. 71, 72, pl. xxxv. figs. 5, 6), but in that species the pores on the back are
few and far between, the number of zocecia in a transverse series is less, the pores on
the front wall are much less numerous, and there is no irregular middle series.

Genus EytatopHora, Lamouroux, 1821.
[Lamouroux, No. 2, p. 81. ]
Diagnosis. Pergens, No. 3, p. 357.

Species 1. ENTALOPHORA TERGEMINA |, n. sp,

Syn. Jdmonea gracillima?, Reuss, Vine, 1889, Proc. Yorks. Geol. & Polyt. Soe. vol. xi. pp. 165,
166, pl. v. fig. 13.

Diagnosis. Zoarium thick, apparently short. In section it appears quadrangular,
with the angles well rounded. Surface minutely pitted.

Zoecia crowded, long, expanding above; series of three or four open together along
a straight line; there are four such triplets at not quite the same level in a series
around the zoarium. There are 12 or 13 in a complete series. The zocecia are some-
what infundibuliform, and have a somewhat quadrangular aperture.

1 Tergeminus, triple, reterring to the apertures being usually in triplets.

BRITISH PALAOGENE BRYOZOA. 261

Distribution. London Clay, Sheppey.

Type. Brit. Mus. No. B 4509.

Figures. Pl. XXXII. figs. 10 a, 100.

Affinities. The specimen which serves as the type of this species is that which
Mr. Vine figured as Jdmonea gracillima, Reuss, but as it belongs to a different family
there is no necessity to compare it with that species. It reminds one, in the form of
the zoarium, of Lntalophora clavula, Reuss!; from this it differs in the serial
arrangement of the apertures. The same character separates it from Entalophora pal-
mata, Busk 2.

This species seems to me to be most allied to Entalophora wanganuwiensis, Waters
(No. 5, pp. 340, 341, pl. xviii. fig. 1): but the New Zealand species has only 10 zocecia
in a series; these are verticillate, and the zocecia are not infundibuliform.

Family HETEROPORID/.

Genus Hetrropora, Blainville, 1830.
[Blainville, No. 1, p. 381.]
Diagnosis. Pergens, No. 3, p. 369.

Species 1. HETEROPORA GLANDIFORMIS %, n. sp.

Diagnosis. Zoarium very small, globular, free (the largest specimen is less than
3 millim. in diameter).

Zoecia irregularly bent tubes. The orifice varies from orbicular to subhexagonal in
shape; they are surrounded by a strong raised rim. The zoccia are crowded, but
interzocecial spaces occur on the surface of the zoarium; these are, however, entirely
filled in the interior. Secondary pores numerous, somewhat less in number than the
normal zocecia, irregularly scattered ; they also have a thickened, slightly raised rim.

Distribution. Barton Beds, Barton (common). Bracklesham Beds, Bracklesham Bay.
tLondon Clay, Highgate. (One somewhat doubtful specimen: Brit. Mus. No. 49596.)

Type. Brit. Mus. No. B 4511. Edwards Coll.

Figures. Pl. XXXII. fig. 11. A zoarium from Barton; external view. Figs.
12a, 6. Fragments to show internal structure.

Affinities. In the form of the zoarium this species resembles most closely some
specimens of Heteropora conifera (Lamx.) [No. 2, p. 87, pl. 83. figs. 6, 7; see Haime,
No. I, pp. 208, 209, pl. xi. figs. 1 a-c], figured by Haime, but the zocecial characters are
quite distinct. H. stellulata, Reuss (No. 1, p. 35, pl. v. figs. 21, 22; Manzoni, No. 4,

' Pustulipora clavula, Reuss, No. 1, p. £1, pl. vi. fig. 11. For later figures see Reuss, No. 1, p. 194, pl. ix.
figs. 3, 4.
* Pustulopora palmata, Busk, No. 6, p. 108, pl. xviii. fig. 2; and Manzoni, No. 4, p. 11, pl. ix. fig. 34.
* From glans, a bullet.
2P2

262 MR. J. W. GREGORY ON THE

p. 18, pl. xi. fig. 44), has certain affinities, but the raised triangular or oval zocecia and
numerous pores of that species are quite distinctive. H. stipitata, Reuss (No. 1, p. 39,
pl. v. fig. 19; Manzoni, No. 4, p. 19, pl. xi. fig. 45), has also more numerous cancellate
pores and a greater thickness of wall. Most of the specimens are less than 2 millim.
in diameter and are perfectly spherical; the largest is about 2°5 millim. in diameter,
and is somewhat flattened and presents a slight resemblance to some specimens in
the Conescharellina stage of Orbitulipora.

V. Miscellaneous Records.

As the Bryozoa are rare in the English Lower Tertiaries, the following records are
inserted in the hope that they may lead to search in those horizons.

Diacnoris InTERMEDIA, A. W. Waters, a, p. 224 (non Hincks); G. R. Vine, a, p. 673,
B, p. 160, c, p. 54.
Distribution. Middle Eocene, Bournemouth.
The British Museum contains some specimens of Bryozoa from the same horizon, but they are
quite indeterminable. Mr. Waters has also recorded Lepralia, sp., Membranipora, sp.,and Flustre,
from the same horizon.

Drraxta vartasitis, D’Orb., G. R. Vine, c, p. 58.
The specimen on which this identification was founded is now im the British Museum (B 4589),
but it seems to me to be generically indeterminable. It came from the London Clay at Fareham.

Criprinina RapIATA (Moll), A. W. Waters, 1883, in H. M. Klassen, a, p. 244; W. Whitaker,
B, vol. i. p. 237.
Distribution. Woolwich and Reading Beds (Blackheath Beds) ; Park Hill, Croydon.

Frustra crassa, Desm., J. Morris, No. 1, p. 37; T. H. Huxley and R. Etheridge, a, p. 332;
W. Whitaker, a, p. 594; G. R. Vine, a, p. 673; J. L. Lobley, a, p. 96.
Distribution. London Clay, Primrose Hill and London District.

Fiusrra, sp., W. Whitaker, T. H. Huxley, and R. Etheridge, a, pp. 574, 577, 581; T. H.
Huxley and E. T. Newton, s, p. 14; W. Whitaker, Bs, vol. i. p. 213.
Distribution. Thanet Beds, E. of Faversham. Woolwich and Reading Beds, Dulwich, Sund-
ridge.
Potyzoa, indet., H. W. Bristow, a, p. 284.
Distribution. Bembridge Beds. (This is the only evidence known to me of the occurrence of
Bryozoa in the British Upper Oligocene.)

Horners minuta, Vine, 8, p. 166, c, p. 58. Bracklesham Beds.
The specimen appears to have been lost.

Horner ? FLABELLIFORMIS, Blainv., Vine, 8, p. 166, pl. v. fig. 15, and c, p. 58.

The specimen upon which this record is founded is now in the British Museum; it is partly
immersed, with the zocecial orifices downwards, the basal portion alone being visible: it is likely
to belong to Jdmonea, and to be the same species as that figured by Lonsdale, No. 2, pl. ix. fig. 24,
as I. coronopus, Defr.

LicHENOPORA MEDITERRANEA ?, Blainv., Vine, c, p. 60.

BRITISH PALAOGENE BRYOZOA. 263

VI. Stratigraphical Distribution.

Bar-
ton.

Hea-

Bracklesham. a
on.

London Clay.

a
Lis}
o
a
Species. Author. Al os
P s|<
oO] .&
Wee} co 3 ; 3
: SO et |e -| 43
ofe3f = : z g ol] a] ah >| 2
Sa Soleal silat eal ss ds] sie! | 8) Ste sie
Alsi s| ais £/ 8] o] Bf) sf] s a0} a i= ric} boas | T=
8/21/38) 4) SS) 8] SS] S] al 4/-2/Sis] € [sls
ats] S|3| 3) a] 2) 2] a) SIS! s/o] Sis] SO] S els
2/2] 5} 3|3| 24] 5] 8) 2) 8] 2/2) S24] & [sie
BIB S|HIE elalalaal lalalime) ale} a fold
|
i I-I ——--}+— ee
Notamia wetherelli ...... (Busk) —
| i}
o |
Membranipora eocena ( 4 )I- =
buskalsye sy evsclese sence n. sp SN elbeeltes! baer dle lM (7H Le eel lee) (Aneee
, |
crassomuralis ...... , al — -|- =
. . i
tenuimuralis........ op |—}—|—}..J/—| |
F : : | | |e | |
virguliformis........ 9p He th |
| |
Gisyun tarry dcteneyetor. a a onl
| |
Lunulites transiens ...... lature olla -|..f—}—l—|—}.-|-p -
. . |
Biselenaria offa .......... a9 silio He _—
Micropora eribriformis ....) ,, aad

Onychocella magnoaperta . .
Cribrilinajvinet =.) 7...
Schizoporella magnoaperta .
-—— magnoincisa ........
Adeonellopsis wetherelli

ANCISA ee EE

Lepralia lonsdalei ........
Umbonula bartonense

calcariformis........

Teichopora clavata........
Meniscopora bigibbera ....
Conescharellina clithridiata .
Orbitulipora petiolus......
Mucronella angustocecium. .

Smittia tubularis) 4. ......

ILO ASE [ool lo of 2 oll> Olle olla oliom||aellacllac

n. sp. |. .J..[--

”

Oligocene, Germany.
Paleogene, New Zealand.

Idmonea giebeli.......... Stol.

pialternata) .5....... n. sp. ]..f.. ee |e

seriatopora ........ Reuss |. .J..]--|--

Cale. gross. France.
| Laekenien, Belgium

GOCINHWE Sascosoccc Defr.

Hornera farehamensis
Entalophora tergemina ....
Heteropora glandiformis

Lichenopora, sp. .......-

n. sp.
”
9?

”?

264 MR. J. W. GREGORY ON THE

VII. Affinities of the Fauna.

The preceding list shows that the Bryozva included in the present paper belong to
three fairly distinct faunas, but a comparison of the three shows that they possess
certain features in common. In the first place, each of the three faunas is numerically
small, both in species and individuals, in comparison with the wealth of forms that
inhabited the contemporary seas of the Mediterranean basin.

The stunted and dwarfed aspect of the three faunas is apparently due mainly to
climatic conditions. As has been pointed out ina recent revision of our Eocene
Echinoids 1, the British seas of that period were confined to the south by a land barrier
which stretched across France and Northern Germany. Hence to the south of this
area the Bryozoa flourished under favourable conditions in a tropical and subtropical
ocean, while on the other side the seas were open to the chilling influences of the
northern ocean. ‘The land barrier was breached in Middle Eocene times, but the
conditions were not seriously modified till later: then, with the gradual change to the
brackish and freshwater deposits of the Oligocene, the marine Bryozoa cease to be
represented in the British Paleogene.

The Echinoids of the period belong to the same genera as their contemporaries in the
Mediterranean basin, but their generally dwarfed aspect and rareness indicate that they
lived under unfavourable conditions. The Bryozoa present exactly the same parallel.

An effort has been made to explain the paucity of Bryozoa in English deposits of
this period as due simply to unfavourable lithological conditions of life and preserva-
tion. The prevalence of clay and sharp sand is quoted as unfavourable to the growth
of Bryozoa. But this is hardly sufficient. The shelly sands of the Bracklesham, on
the contrary, would seem to indicate the conditions that would be most favourable to
the existence and preservation of Bryozoa. That the clay shores of the London Clay
and Barton are wholly responsible for the rarity of the Bryozoa is not likely to be
accepted by any one who has dredged on the great mud-flats off the Essex coasts, where
it is often difficult to procure a shell not encrusted by them. In other districts, such
as the Paris basin, Belgium, and North Germany, which were also to the north of this
land barrier, and where the lithological characters of the sea-floors were quite different
from those of England, the Bryozoa are equally rare and stunted.

Hence, it is to geographical questions rather than to the lithological conditions of
the sea-floor that we must attribute the marked characters of our Paleogene Bryozoan °
fauna.

The singular diversity of the fauna is another feature which supports the view that
it is to be regarded as a remnant or an offshoot from one that was much greater and
richer. Mr. Waters [| Nos. 12 & 13], in his revision of the Oligocene Bryozoa of North
Italy, admits 88 species, representing 35 genera. But the British fauna contains only

* Gregory, Proc. Geol. Assoc. xii. 1891, pp. 51, 52.

BRITISH PALZOGENE BRYOZOA. 265

25 species, belonging to 17 genera. If we take the case of the species of a single
genus, we find the same point very instructively shown. Thus the genus Jdmonea is
represented by only five specimens, of which two from Mr. Vine’s collection appear to
me to be indeterminable ; Ldmonea may be conveniently divided into three groups or
subgenera, and one of the three recognizable specimens belongs to each of these three
groups. ‘This consideration ought to stimulate the search for more material, as the
specimens already known appear to represent but a fragment of the fauna.

The high proportion of peculiar species in this fauna would not excite surprise in
any of the higher groups, except the Bryozoa; but when we consider the vast range
both in space and time claimed for some species, a few words of explanation are
required. In the first place, rare though the Bryozoa are in the English beds, they
appear to have been even scarcer in contemporary deposits of other parts of the same
basin; the meagre lists given by Stremme (No. 1), Marsson (No. 2), Michelin (No. 1),
Milne-Edwards (No. 2), and Mourlon (No. 1) show the paucity of Bryozoa at this time
in Northern France, Germany, and Belgium.

The great range in time usually accorded to species of Bryozoa raises the general
question as to the value of species in this group; their growth in colonies is the main
reason for the ‘“‘lumping ” tendencies of zoophytologists. In the Cheilostomata species
are usually founded, if only on one specimen, yet on hundreds of zocecia: in a colony
of this size great variation is inevitable; many of the polypites are crushed out by
growth-pressure, and their zocecia are malformed or aborted; the older zocecia become
immersed and lose their characters; the younger zocecia at the tips of the branches
are immature. Hence it is easy to pick out two zocecia in a zoarium which differ far
more markedly than do two zocecia taken from different species; but that no more
proves that the two species should be merged than that two species of frogs are
identical because they resemble one another more closely than they do the tadpoles
from which they have developed.

Dr. Waagen— Pal. Indica’ (xiii.), ‘Salt Range Fossils,’ iv. pt. 2, ‘Geol. Results,’ 1891,
pp. 235, 236—has recently pointed out the disastrous effects that have been wrought
by paleontologists ‘‘ lumping” species and neglecting slight but definite differences ;
and one worker on Bryozoa has recently expressed his doubts as to the accuracy of the
identification of recent and Cretaceous species. With this opinion I feel strongly
disposed to concur, but will here only say that, so far, I have seen no Cretaceous
species of Cheilostomata identical with a living one. If there are such constant
differences, it seems certainly advisable to recognize them by name, whether we call
them species, forms (Smitt), or mutations (Waagen). Unless this be done, if we
accept species as ranging from the Jurassic to the present, then we must abandon all
hope of deriving from the Bryozoa any assistance in the study of the geographical
distribution of the past, though the group presents characters that should give its

evidence great value.

266 MR. J. W. GREGORY ON THE

VIII. Bibliography.

(Including only works referred to.)

A. General.
Avpovuin, Victor.

No. 1.—1826. Explication sommaire des planches de Zoophytes de Egypte et de la Syrie. . .
offrant un exposé des caractéres naturels des genres avec la distinction des
espéces. In ‘ Description de l’Egypte.’? Histoire Naturelle, i. 1809, Polypes,
pp. 225-244, pl. 14. Fol. Paris.

Buainvittz, H. M. D.'pe.
No. 1.—1830. Zoophytes. Dict. Sci. Nat. lx. pp. 1-546, pls. 67, 68.
No. 2,—1834. Manuel d’Actinologie ou de Zoophytologie. 8vo. Paris. 2 vols.

Bosc, L. A. G.
No. 1.—1830. Histoire naturelle des Vers. Ed.2. 12mo. Paris.

Bronenzart, ALEx., and Cuvinr, G.
No. 1.—1822. Description géologique des Couches des environs de Paris parmi lesquelles se
trouvent les gypes 4 ossemens. In Cuvier, Oss. foss. ed. 2. t. 11. pt. 2, Ato,
pp. 2389-648 ; 15 pls. & maps. :

Brown, H. G.

No. 1,.—1825. System der urweltlichen Pflanzenthiere. Fol. Heidelberg. iv+47 pp. 7 pls.

No, 2.—1831. Uebersicht der fossilen Ueberreste in den tertiaren subapenninischen Gebirgen.
In Ergebnisse meiner naturhistorisch-dkonomischen Reise. Th. u. Skizzen
und Ausarbeitungen iiber Italien. Heidelberg. 8vo. Bd. ii. pp. 505-646.
pl. iii.

No. 3.—1848. Handbuch der Geschichte der Natur. Bd. iti. Index paleontologicus. A.
Nomenclator palzontologicus. 8vo. Stuttgart, 1848. Pp. Ixxxiv+ 1381.

Busk, G.

No. 1.—1852. An account of the Polyzoa and Sertularian Zoophytes collected in the Voyage of
the ‘ Rattlesnake’ on the coasts of Australia and the Louisiade Archipelago.
In J. Macgillivray, Narrative of the Voyage of H.M.S. ‘Rattlesnake.’ 8vo.
London, 1852. Vol. i., App. No. iv. pp. 343-402, pl. 1.

No. 2.—1852. British Museum Catalogues. Catalogue of Marine Polyzoa. Pt. I. Cheilo-
stomata. Pp. vili+vi+54, pl. 68.

No. 3.—1854. Ditto. Pt. I]. Cheilostomata. Pp. viii+55-120, pls. 69-124.

No. 4.—1875. Ditto. Pt. III. Cyclostomata. Pp. viii+4, pl. 34.

No. 5.—1858. Zoophytology [9th part]. Quart. Journ. Mier. Sci. vi. pp. 124-180, pls. xviii.
& xix.

No. 6.—1859. A Monograph of the Fossil Polyzoa of the Crag. Paleont. Soc. Pp. xiii+ 136,
pl. 22.

No. 7.—1866. Description of three new Species of Polyzoa from the London Clay at Highgate,
in the collection of N.'T. Wetherell, Esq., F.G.S. Geol. Mag. iii. pp. 298-
302, pl. xii.

BRITISH PALZOGENE BRYOZOA. 267

No. 8.—1884. Report on the Polyzoa collected by H.M.S. ‘Challenger’ during the years 1873-
1876. Part I. The Cheilostomata. Chall. Exp., Zool. vol. x. (pt. xxx.),
pp. xxiv+ 216, pls. 36.

No. 9.—1886. Ditto. Part II. The Cyclostomata, Ctenostomata, and Pedicellinea. Ibid.
vol. xvil. pp. vui+47, pls. 10.

Destonecnames, Eve.
No. 1.—1824. In Encyclopédie Méthodique. Histoire Naturelle des Zoophytes ou Animaux
rayonnés. 4to. Paris.

Doturvs, G. F.
No. 1.—1889. Bryozoaires. Ann. géol. univ. v. (1888) pp. 1159-1170.

Epwarps, H. Mrunt-.
No. 1.—1836. Recherches anatomiques, physiologiques et zoologiques sur les Eschares. Ann.
Sci. Nat., Zool. (2) vi. pp. 5-53, pls. iv.
No. 2.—1836. Observations sur les polypiers fossiles du genre Eschare. Ibid. pp. 321-345,

pls. 1x.-xi.

FLEMING, JouN.
No. 1.—1828. A History of British Animals. 8vo. Edinburgh. Pp. xxiiit+565.

Gass, W. M., and Horn, G. H.
No. 1.—1862. Monograph of the Fossil Polyzoa of the Secondary and Tertiary Formations of
North America. Journ. Acad. Nat. Sci. Phil. v. pp. 111-179.

Grout, G.
No. 1.—1889. Briozoi neogenici dell’ isola di Pianosa nel Mar Tirreno. Atti Soc. Tose. Sci.

Nat. x. pp. 251-267, pl. xiv.

Goupruss, Avec.
No. 1.—1827. Petrefacta Germaniz. Fol. Diisseldorf (1827-1833). Vol. I. Bryozoa, pp. 23-41,

pls. vili.-xii.

Gorrarni, G. B.
No. 1.—1886. Briozoi fossili di Montecchio Maggiore. Atti Soc. Veneto-Trentina, ix. fs. ii.
Padova (1885), pp. 297-808, pl. xiv.

Grucorio, Marquis ANTonio Dr.
No. 1.—1890. Monographie de la Faune EKocénique de |’ Alabama et surtout de celle de Claiborne
de ’Etage Parisien. Ann. Géol. livr. vii., viii. 4to. Palermo. 316 pp.
46 pls.

Hacenow, Fr. von.
No. 1.—1839. Monographie der Riigenschen Kreide Versteinerungen. Abt. I. Phytolithen

und Polyparien. N. Jahrb. 1839, pp. 252-296, pls. iv., v.
No. 2.—1851. Die Bryozoen der Maastrichter Kreidebildung. Cassel. Pp. xi+111, pls. 12.
VOL. XII.—Part v1. No. 7.—June, 1893. 2a

268 MR. J. W. GREGORY ON THE

Haig, Jugs.
No. 1.—1854. Description des Bryozoaires fossiles de la formation Jurassique. Mém. Soc.
Géol. France, (2) v. pp. 156-218, pls. vi.—xi.

Herter, Cam.
No. 1.—1867. Die Bryozoén des adriatischen Meeres. Verh. k. k. zool.-bot. Ges. Wien, xvii.
Abh. pp. 77-136, pls. 1.-vi.

Hincxs, T.

No. 1.—1878. Notes on the Genus Retepora, with descriptions of new Species. Ann. Mag.
Nat. Hist. (5) i. pp. 853-365, pls. xviil., xix.

No. 2.—1880. A History of the British Marine Polyzoa. 2 vols. 8vo. London. Pp. clxi+
601, pls. 83.

No. 3.—1880. On new Hydroida and Polyzoa from Barents Sea. Ann. Mag. Nat. Hist. (5) vi.
pp. 277-286, pl. xv.

No. 4.—1880-1891. Contributions towards a General History of the Marine Polyzoa:—

Part I. 1880. Ann. Mag. Nat. Hist. (5) vi. pp. 69-92, pls. ix.—xi.

II. 1880. 55 ;\ Vi. pp. 376-384, pls. xvi., xvii.
HeessTe _ "vii. pp. 147-161, pls. viii—x.
LV lssie 55 vill. pp. 1-14.

V. 1881. 5 vill. pp. 122-136, pls. i.-v.
VI. 1882. BS ix. pp. 116-127, pl. v.

VII. 1882. as x. pp. 160-170, pls. vil., vill.
VIII. 1888. i xi. pp. 193-202, pls. vi., vil.
IX. 1884. 5 Xill. pp. 265-267.

X. 1884. 3 Xl. pp. 356-369, pls. xill., xiv.

XI. 1884. + Xiv. pp. 276-285, pls. viil., ix.
XG 1835; 5 xv. pp. 244-257, pls. vil.—ix.
XIII. 1891. 53 (6) vii. pp. 285-298, pls. vi., vil.

No. 5.—1891-1892. Ditto. Appendix :—
Part I. 189]. Ann. Mag. Nat. Hist. (6) vii. pp. 86-93.

MES Tse yk: 5 vii. pp. 169-176.
TES SON: 5 vill. pp. 471-480.
IV. 1892. as 1x. pp. 3827-334.

No. 6.—1886. The Polyzoa of the Adriatic : a supplement to Prof. Heller’s ‘Die Bryozoén des
adriatischen Meeres, 1867. Ann. Mag. Nat. Hist. (5) xvii. pp. 254-271,

pls. 1x., x.
No. 7.—1887. On the Polyzoa and Hydroida of the Mergui Archipelago collected... . by
Dr. J. Anderson. . . . Journ. Linn. Soc., Zool. xxi. pp. 121-135, pl. xii.

No. 8.—1887. Critical Notes on the Polyzoa :—
Part I. 1887. Ann. Mag. Nat. Hist. (5) xix. pp. 150-164.
II. 1890. 5 (6) v. pp. 88-103.

Hurton, F. W.
No. 1.—18380. Manual of the New Zealand Mollusca. Misc. Public. Col. Mus. Geol. Surv. N.
Zeal. No. xii. 8vo. Wellington. Pp. xvitiv+224. Bry. pp. 178-199.

BRITISH PALHOGENE BRYOZOA. 269

Jameson, R.
No. 1.—1811. Catalogue of Animals of the Class Vermes found in the Frith of Forth and other
parts of Scotland. Mem. Wern. Soe. i. pp. 556-565.

Jetty, E. C.
No. 1.—1889. A Synonymic Catalogue of the Recent Marine Bryozoa, including Fossil
Synonyms. 8vo. London. Pp. xv+822.

Jounston, GErorGu.
No. 1.—1838. A History of British Zoophytes. 8vo. London. Pp. xii+338, pls. 44.
No. 2.—1847. Ditto. Ed.2. 2vols. Pp. xvit+ 488, pls. 74.

JULLIEN, JULES.

No. 1.—1882. Note sur une nouvelle Division des Bryozoaires Cheilostomiens. Bull. Soc. Zool.
France, vi. pp. 271-285.

No. 2.—1888. Dragages du ‘ Travailleur.’ Bryozoaires, espéces draguées dans l’Océan Atlantique
en 1881. Ibid. vi. pp. 497-529, pls. xili.—xvii.

No. 3.—1886. Les Costulidées, nouvelle famille de Bryozoaires. Ib. xi. pp. 601-620, pls. xvii.—xx.

No. 4.—1888. Bryozoaires. Mission Scientifique du Cap Horn, 1882-1883. Zool. vi. 4to.
92 pp., 15 pls.

KircHENPAUER, G. H.
No. 1.—1880. Ueber die Bryozoen-Gattung Adeona. Abh. Naturwiss. Ver. Hamburg, vii.

pp. 1-24, pls. 1.11.

Kirxpatricr, R.
No. 1.—1888. Polyzoa of Mauritus. Ann. Mag. Nat. Hist. (6) i. pp. 72-85, pls. vil.—x.
No. 2.—1890. Report on the Zoological Collections made in Torres Straits by Prof. A. C.
Haddon, 1888-1889.—Hydroida and Polyzoa. Sci. Proc. Roy. Dublin Soc.
new ser. vi. pp. 603-626, pls. xiv.—xvii.

Kuopen, K. F.
No. 1.—1834. Die Versteinerungen der Mark Brandenburg, insonderheit diejenigen welche sich

in den Rollsteinen und Blécken der siidbaltische Ebene finden. 8vo. Berlin.
Pp. x+ 378, pls. 10.
Koscuinsry, C.
No. 1.—1885. Ein Beitrag zur Kenntniss der Bryozoenfauna der iilteren Tertiadrschichten des
siidlichen Bayerns. I. Chilostomata. Palaeontogr. xxxii. 1885, pp. 1-73,

pls. 1.-vii.

Lamarck, J. B.
No. 1.—1816. Histoire naturelle des Animaux sans Vertébres, t. 11. 8vo. Paris.

Lamovurovx, J. V. F.
No. 1.—1816. Histoire des Polypiers Coralligénes flexibles, vulgairement nommés Zoophytes.
— 8vo. Caen. Pp. lxxxiv +560, pls. 19.
No. 2.—1821. Exposition méthodique des genres de ordre des Polypiers. 4to. Paris, 1821.
Pp. vni+ 115, pls. 84.
i 2Q2

270 MR. J. W. GREGORY ON THE

No. 3.—1824. In Quoy & Gaimard, Voyage autour du Monde... .. exécuté sur les corvettes
de 8S. M. ‘V’Uranie’ et ‘la Physicienne’ pendant les années 1817, 1818, 1819,
et 1820. 4to. Paris. Pp. 604-641, pls. 89-95.

Lienengxavs, EH.
No. 1.—1891. Die Ober-Oligocin Fauna des Doberges. Jahresber. Naturwiss. Ver. Osna-
briick, vill. pp. 43-178, pls. 1., 11.

LonspaLze, Wm.
No. 1.—1845. Account of the species of Polyparia obtained from the Miocene Tertiary forma-
tions of North America. Quart. Journ. Geol. Soc. 1. pp. 495-509.
No. 2.—1850. In F. Dixon, The Geology and Fossils of the Tertiary and Cretaceous Formations
of Sussex. 4to. Pp. xvi+xvi+433, 40 pls.
No. 3.—1878. Ditto. Ed. 2.

Macertrivray, P. H.
No. 1.—1879-1890. In McCoy, Prodromus of the Zoology of Victoria.

Dee. IIT. pls. 24-26. 1879. Dec. IX. pls. 89-90. 1884. Dec. XV. pls. 146-148. 1887.

Dien BGeaiy ee X. 94-99. 1885. OV. WEN, TSS
V. 45-49. 1880. XA O5- 108s ue XV eeGb= 1685 ee
VI. 57-60.. 1881. XII. 116-118. 1886. XVIM. 175-178. 1889:
VII. 66-68. 1882. Xe 62128 SDS IRB,
VIL. 78. 1883. MINK, WENS, TGEY XX. 195-196. 1890.

No. 2.—1882-1891. Descriptions of new or little-known Polyzoa. Trans. Roy. Soc. Victoria.

Pt. I. 1882, xviii. pp. 115-121, 1 pl. Pt. VIII. 1885, xxi. pp. 106-119, 5 pls.
II. 1883, xix. pp. 130-138, 3 pls. IX. 1886, xxii. pp. 128-139.
pps lOl=t9 a 2m X. 1887, xxiii. pp. 34-38, 2 pls.
IV Sy) G8 PRs eo /—2 9342 Xa] » pp. 64-72, 3 pls.
V. 1884; xx. pp. 103-113, 3 ,, MG. 95 >> pp. 179-186.
Wale, S55 pp. 126-128, 1 pl. XIII. 1890, (new ser.) ii. pp. 106-110, pls. iv., v.
VII. 1885, xxi. pp. 92-99, 3 pls. XIV. 1891, Pe ill. pp. 78-83, pl. ix., x.

No. 3.—1887. A Catalogue of the Marme Polyzoa of Victoria. Trans. and Proc. Roy. Soe.
Vict. xxiii. pp. 187-224.

No. 4.—1889. On some South Australian Polyzoa. Trans. and Proc. Roy. Soc. S. Austr.
xi. pp. 24-80, pl. i1.

No. 5.—1890. An additional List of South Australian Polyzoa. Trans. Roy. Soc. S. Austr.
xiii. pt. 1, pp. 1-7, pl. 1.

Manzoni, A.

No, 1.—1869. Briozoi pliocenici Italiani. Pt. I. Sitzb. k. Ak. Wiss. Wien, lix. Abt. i. pp. 17-28,
2 plates.

No. 2.—1869-1870. Briozoi fossili Italiani. Pt.I1.-IV. Ibid. lix. Abt. i. pp. 512-523, 2 plates ;
Ix. Abt. pp. 930-944, 4 plates; Ixi. Abt. pp. 323-349, 6 plates.

BRITISH PALAOGENE BRYOZOA. 271

No. 3.—1877. I Briozoi fossili del Miocene d’Austria ed Ungheria. Parte II. Celleporidea,
Escharidea, Vincularidea, Selenaridea. Denk. k. Akad. Wiss. Wien, xxxvii.
Abt. il. pp. 49-78, pls. i.-xvii.
No. 4.—1878. Ditto. Parte III. Crisidea, Idmoneidea, Entalophoridea, Tubuliporidea, Diasto-
poridea, Cerioporidea. Ibid. xxxviii. Abt. ii. pp. 1-24, pls. i—xviii.
Marsson, Tu. -
No. 1.—1887. Die Bryozoen der weissen Schreibkreide der Insel Riigen. Pal. Abh. iv. Heft i.
pp. 112, pls. 10.
No. 2.—1888. In Fr. Noetling, Die Fauna des samlandischen Tertiirs. Th. ii. Lf. v. Bryozoa.
Abh. geol. Specialk. Preuss. vi. Heft 4, pp. 555-560.
Meunier, A., & Percens, Ep.
No. 1.—1886. Les Bryozoaires du Systeme Montien. 8vo. Louvain. Pp. 15, pls. 3.
No. 2.—1886. Nouveaux Bryozoaires du crétacé supérieur. Ann. Soc. Malacol. Belg. xx. Mém.
pp. 32-37, pl. ii.
No. 3.—1887. La faune des Bryozoaires garumniens de Faxe. Ibid. xxi. (1886) Mém. pp. 187-
24.2, pls. ix.—xiii.
Micuetin, Harpovurn.
No. 1.—1840-1847. Iconographie Zoophytologique: description par localités et terrains des
Polypiers fossiles de France et pays environnants. Groupe Supracrétacé, pp. 149-
178, pls. 48-46. 1844.
Mout, J. P. C.
No. 1.—1803. Eschara ex Zoophytorum seu Phytozoorum ordine pulcherrimum ac... ... 4to.
Vindobone. 70 pp., 4 pls.
Morren, C. F. A.
No. 1.—1828. Descriptio coralliorum fossilium in Belgio repertorum. Ann. Ac. Groning. 4to.
76 pp., 7 pls.
Morris, J.
No. 1.—1843. Catalogue of British Fossils. 8vo. London. Pp. x +222.
No. 2.—1854. Ditto. Ed. 2. Ditto. Pp. vili+ 372.

Movrton, Micuet.
No. 1.—1881. Géologie de la Belgique. 8vo. Bruwelles. T. ii. pp. xvit+392.

Mutter, O. F.
No. 1.—1788-1806. Zoologia Danica, seu Animalium Danie et Norvegie rariorum ac minus
notorum Descriptiones et Historia. Fol. Havnie. Vols. i. 1788, 52 pp.;. ii.
1788, 56 pp.; i. 1789, 71 pp.; iv. 1806, 46 pp.

MOnstER, von.
No. 1.—1835. Bemerkungen iiber einige tertiire Meerwasser-Gebilde in nordwestlichen
Deutschland zwischen Osnabriick und Cassel. N. Jahrb. 1835, pp. 420-451.

Novdx, Orromar.
No. 1.—1877. Beitrag zur Kenntniss der Bryozoen der béhmischen Kreideformation. Denk. k.
Ak. Wiss. Wien, xxxvii. Abt. ii. pp. 79-118, pls. i—x.

272 MR. J. W. GREGORY ON THE

D’Orzieny, ALCIDE.
No. 1.—1839 & 1846. Voyage dans Amérique méridionale. v. pt. 4. Zoophytes. 4to. Paris.
28 pp., 15 pls.
No. 2.—1850-1852. Bryozoaires. Pal. Frang. Terr. Crét. v. 8vo. Paris. 1192 pp., pls. 600-

800.

Orrmann, A.
No. 1.—1890. Die Japanische Bryozoenfauna. Arch. f. Nat. 1890, I. Heft i. pp. 1-74,

pls. itv.
Percens, Ev. (See also Mruniur & PrRcens.)

No. 1.—1887. Pliocine Bryozoén von Rhodos. Ann. k.k. Naturh. Hofmus, Wien, i. pp. 1-34,
pl. i.

No. 2.—1889. Zur fossilen Bryozoenfauna von Wola Lu’zanska. Ann. Soc. Géol. Belg. Mém.,
Hydr. ui. pp. 59-72. ,

No. 3.—1889. Revision des bryozoaires du Crétacé figurés par d’Orbigny. I. Cyclostomata. Ibid.
ii, pp. 305-400, pls. xi1.—xiii.

No. 4.—1889. Les Bryozoaires du Tasmajdan 4 Belgrade. Ann. Soc. Malac. Belg. xxii. (1887)
Bull. pp. xii-xxviil.

No. 5.—1889. Note supplémentaire sur ditto. Ibid. pp. lix—lx.

No. 6.—1889. Note préliminaire sur les Bryozoaires fossiles des environs de Kolosvar. Ibid.
pp. XXX11—XXXVHl.

No. 7.—1889. Untersuchungen an Seebryozoen. Zool. Anz. xi. pp. 504-510, 526-533.

No. 8.—1890. Notes succinctes sur les Bryozoaires. Ann. Soc. Mal. Belg. xxiv, Bull. pp. xx—

XXIV.

Puitterr, R. A.
No. 1.—1844. Beitrige zur Kenntniss der Tertiirversteinerungen des nordwestlichen Deutsch-

lands. 4to. Kassel (1843). 87 pp., 5 pls.

Pourtatus, L. F. pe.
No. 1.—1868-1869. Contributions to the Fauna of the Gulf Stream at great depths (1867-8).

Bull. Mus. Comp. Zool. i. pp. 108-120, 121-142.

Revss, A. E. von.

No. 1.—1847. Die fossilen Polyparien des Wiener Tertiiirbeckens. Ein monographischer
Versuch. Haidinger’s Naturwiss. Abh. 11. 8.i. Wien. 4to. 109 pp., 11 pls.

No. 2.—1851. Ein Beitrag zur Paliontologie der Tertiarschichten Oberschlesiens. Zeit. deut.
geol. Ges. iii. pp. 149-184, pls. viii., 1x.

No. 3.—1855. Beitrage zur Charakteristik der Tertiaérschichten des nérdlichen und mittleren
Deutschlands. Sitzb. k. Ak. Wiss. Wien, xviii. Abt. i. pp. 197-273, 12 pls.

No. 4.—1864. Bemerkungen iiber die Bryozoengattung Cumulipora, vy. M. Jahrb. k. k. geol.
Reichs. xiv. Verh. pp. 21, 22.

No. 5.—1864. Ueber Anthozoen und Bryozoen des Mainzer Tertiirbeckens. Sitzb. k. Ak. Wiss.
Wien, 1. Abt. 1. pp. 197-210, pls. i., ii.

No. 6.—1864. Die fossilen Foraminiferen, Anthozoen und Bryozoen von Oberburg in Steier-
mark. Denk. k. Ak. Wiss. Wien, xxiii. pp. 1-38, pls. i-x.

7.—1865.

BRITISH PALHOGENE BRYOZOA. 273

No. Die Foraminiferen, Anthozoen und Bryozoen des deutschen Septarienthones. Hin
Beitrag zur Fauna der mitteloligocinen Tertiarschichten. Tbid. xxv. pp. 117-214,
pls. 1.—x1.

No. 8.—1865. Zur Fauna des deutschen Oberoligocins. Sitzb. k. Ak. Wiss. Wien, 1. Abt. 1.
pp. 614-691, pls. vi.-xv.

No. 9.—1867. Die fossile Fauna der Steinsalzablagerungen von Wieliczka in Galizien. - Ibid. lv.
Abt. i. pp. 17-182, pls. 1.-vin.

No. 10.—1867. Ueber einige Bryozoen aus dem deutschen Unteroligociin. Ibid. pp. 216-234,
pls. 1.11.

No, 11.—1869. Palaontologische Studien tiber die alteren Tertiirschichten der Alpen. Th. ii.
Die fossilen Anthozoen und Bryozoen der Schichtengruppe von Crosara. Denk.
k. Ak. Wiss. Wien, xxix. pp. 215-298, pls. xvii. —xxxvi.

No. 12.—1870. Ueber tertiare Bryozoen von Kischenew in Bessarabia. Sitzb. k. Ak. Wiss.

. Wien, lx. Abt. i. pp. 505-518, pls. 1., ii.

No. 13.—1872. Die Bryozoen und Foraminiferen des unteren Planers. In H.B. Geinitz, Das
Elbthalgebirge in Sachsen. IV. Palaeontogr. xx. pp. 97-144, pls. xxiv.-xxxill.

No. 14.—1874. Die fossilen Bryozoen des dsterreichisch-ungarischen Midcans. Abt. i.

Riptey, 8. O.
/ No. 1.—1881].

Romer, F. A.
No. 1.—1840.

No. 2.—1863.

ScureiBer, A.
No. 1.—1872.

Smirt, F. A.
No. 1.—1867.

No. 2.—1872.

No. 3.—1873.

Speyer, Oscar.
No. 1.—1864.

Salicornaridea, Cellularidea, Membraniporidea. Denk. k. Ak. Wiss. Wien, xxxiii.
pp. 141-190, pls. 1.—xil.

Account of the Polyzoa collected during the Survey of H.M.S. ‘Alert’ in the
Straits of Magellan and on the Coast of Patagonia. Proc. Zool. Soc. 1881,
pp. 44-61, pl. vi.

Die Versteinerungen des norddeutschen Kreidegebirges. 4to. Hannover. Bryozoen
in Lf. i. pp. 11-25, pl. v.
Beschreibung der norddeutschen tertiaren Polyparien. Palacontogr. ix. pp. 199-

246, pls. xxXXV.—XXxix.

Die Bryozoen des mitteloligocinen Griinsandes bei Magdeburg. Zeit. £. gesammt.
Naturwiss. xxxix. pp. 475-481, pls. iv., v.

Bryozoa marina in regionibus arcticis et borealibus viventia recensuit. Ofver. K.
Vet.-Akad. Forh. Stockholm, xxiv. pp. 443-487.

Floridan Bryozoa collected by Count L. F. de Pourtales:
Vet.-Akad. x. No. 11.

Ditto. Pt. II.

Pt. I. Handl. K. Svens.
20 pp., 5 pls.
Ibid. xi. No. 4. 83 pp., 13 pls.

Die Tertiarfauna von Séllingen bei Jerxheim im Herzogthum Braunschweig.
Palaeontagr. ix. pp. 247-337, pls. xl—xhi.

274

MR. J. W. GREGORY ON THE

SroxiczKa, F'rrp.

No. 1.—1862.
No.
No.

No. 4.—1865.

2.—1862.

3.—1864.

Srremme, E.

No. 1.—1888.

Waters, A. W.
No. 1.—1881.
No. 2.—1882.
No. 3.—1882.
No. 4.—1883.
No. 5.—1884.
No. 6.—1885.
No. 7.—1887.
No. 8.—1887.
No. 9.—1887.
No.

No.
No.

No.

Wauiretrcen, T.

Oligocine Bryozoen von Latdorf in Bernburg. Sitzb. k. Ak. Wiss. Wien, xlv.
Abt. i. pp. 71-94, pls. i-ii1.

Ueber heteromorphe Zellenbildungen bei Bryozoen, Coelophyma, Reuss. Verh.
k. k. zool.-bot. Ges. Wien, xii. Abh. pp. 101-104.

Kritische Bemerkungen zu Herrn Fr. A. Rémer’s Beschreibung der nord-
deutschen tertiiren Polyparien. N. Jahrb. 1864, pp. 340-347.

Fossile Bryozoen aus dem tertiiren Griinsandsteine der Orakei-Bay bei Auckland.
‘Novara’ Reise, Geol. Th. i. Abt. 1. Pal. pp. 87-158, pls. xvii.—xx.

Beitrag zur Kenntniss der tertiiren Ablagerungen zwischen Cassel und Detmold,
nebst einer Besprechung der norddeutschen Pecten Arten. Zeit. deut. geol. Ges.
xl. pp. 310-354, pls. xx.-xxi.

On fossil Chilostomatous Bryozoa from South-west Victoria, Australia. Quart.
Journ. Geol. Soc. xxxvii. pp. 309-347, pls. xiv.-xviii.

On fossil Chilostomatous Bryozoa from Mount Gambier, South Australia. Quart.
Journ. Geol. Soc. xxxviii. pp. 257-276, pls. vil.—ix.

On Chilostomatous Bryozoa from Bairnsdale, Gippsland. Ibid. pp. 502-513,
pl. xxi.

Fossil Chilostomatous Bryozoa from Muddy Creek, Victoria. Ibid. xxxix.
pp. 423-443, pl. xii.

Fossil Cyclostomatous Bryozoa from Australia. Ibid. xl. pp. 674-697, pls. xxx., xxxi.

Chilostomatous Bryozoa from Aldinga and the River Murray Cliffs, South
Australia. Tbid.xli. pp. 279-310, pl. vii.

On Tertiary Chilostomatous Bryozoa from New Zealand. Ibid. xlii. pp. 40-72,
pls. vi.—viil.

On Tertiary Cyclostomatous Bryozoa from New Zealand. Ibid. xlii. pp. 337-350,
pl. xvii.

Bryozoa from New South Wales, North Australia, &. Pt. I. Ann, Mag. Nat.
Hist. (5) xx. pp. 81-95, pl. iv. Pt. II. Ibid. pp. 181-208, pls. v., vi. Pt. IIT.
pp- 253-265, pl. vii.

10.—1888. Supplementary Report on the Polyzoa collected by H.M.S. ‘ Challenger ’

during the years 1873-1876. Chall. Exp., Zool. xxxi. (pt. Ixxix.). 4to. 41 pp.,
3 pls.

11.—1889. Bryozoa from New South Wales. Ann. Mag. Nat. Hist. (6) iv. pp. 1-24,

pls. ii.

12.—1891. North Italian Bryozoa. Pt. I. Cheilostomata. Quart. Journ. Geol. Soc. xvii.

pp. 1-34, pls. 1.-1v.

13.—1892. Ditto. Pt. 11. Cyclostomata. Ibid. xlviii. pp. 153-162, pl. ii.

No. 1.—1887. Notes on some Australian Polyzoa. Proc. Linn. Soc. N. 8. Wales (2) ii.

pp. 837-347. [Reprinted 1888, Ann. Mag. Nat. Hist. (6) i. pp. 13-22.]

BRITISH PALAOGENE BRYOZOA. 275

B. Works bearing on the Distribution of the British Eocene Bryozoa.

Bristow, H. W.
4.—1889. Geology of the Isle of Wight. Ed. 2.

Huxury, T. H., and Ernerinen, R.
A.—1865. A Catalogue of the Collection of Fossils in the Museum of Practical Geology. 8vo.
Pp. Ixxixx +381.

Ditto, and Newron, Hf. T.
B.—1878. A Catalogue of the Tertiary and Post-Tertiary Fossils of the Museum of Practical
Geology. 8vo. 90 pp.

Jupp, J. W.
4.—1883. The Oligocene Strata of the Hampshire Basin. Geol. Mag. (2) x. pp. 525-527.

Kuaassen, H. M.
A.—1883. On a section of the Lower London Tertiaries at Park Hill, Croydon. Proc. Geol.
Assoc. vill. pp. 226-249.

Losury, J. Logan.
Aa,—1887. The Geology of the Parish of Hampstead. Trans. Middlesex Nat. Hist. Sei. Soc.
pp. 64-102.

Vine, G. R.
4.—1886. Report on Recent Polyzoa. Rep. Brit. Assoc. 1885, pp. 481-680.
B.—1889. Notes on British Eocene Polyzoa. Proc. Yorks. Geol. & Polyt. Soc. xi. pt. i. pp. 154-
169, pl. v.
c.—1892. Notes on some new or little-known Hocene Polyzoa from localities. Ibid. xii. pt. i.
pp. 52-61.

Warrrs, A. W.
a.—1879. In Gardner, J. S8., Description and Correlation of the Bournemouth Beds. Pt. I.
Upper Marine Series. Quart. Journ. Geol. Soc. xxxv. pp. 224-225.

Weraerc.t, N. T.
A.—1837. Observations on a well dug on the south side of Hampstead Heath. ‘Trans. Geol.

Soe. (2) v. pp. 181-135, pl. ix.

Wuirakrr, W.
A.—1872. The Geology of the London Basin. Pt. I. Mem. Geol. Surv. iv. pt. i.
B.—1889. The Geology of London and of part of the Thames Valley. Mem. Geol. Surv.
2 vols.

VOL. XIII.—PART VI. No. 8.— June, 1893. 2k

276

MR. J. W. GREGORY ON THE

IX. EXPLANATION OF THE PLATES.

PLATE XXIX.

_1. Notamia wetherelli (Busk), p. 226. Fig. 1a, front view; 14, lateral view.

London Clay, Highgate. Brit. Mus. No. 49731. x 37 diam.

. 2. Membranipora eocena (Busk), p. 228. London Clay, Highgate. Brit. Mus.

No. 49729. »x 16 diam.

3. Membranipora eocena (Busk). View of the back of the zoarium. London
Clay, Highgate. Brit. Mus. No. 6350. x 12 diam.
. 4. Membranipora eocena (Busk). Woolwich and Reading Beds, Croydon. . Brit.

Mus. X* diam.

. 5. Membranipora tenuimuralis, n. sp., p. 231. London Clay, Highgate. Brit.

Mus. No. 49736 (part of one of Busk’s types of MW. lacroivi). x *? diam.

. 6. Membranipora tenuimuralis, n. sp. London Clay, Highgate. Brit. Mus.

No. 49736. x * diam.

. 7. Membranipora tenuimuralis, n. sp. London Clay, Highgate. Brit. Mus.

No. B 4331. x 55 diam.

. 8. Membranipora virguliformis, n. sp., p. 232. London Clay, Highgate. Brit.

Mus. No. 49658. x 25 diam.

9. Membranipora disjuncta, n. sp., p. 232. London Clay, Highgate. Brit. Mus.
No. 69205. Fig. 9a. xX 4 diam., to show general arrangement of the
zoarium. Fig. 96. x 12 diam., to show structure of the zocecia.

. 10a. Membranipora crassomuralis, n. sp., p. 229. Barton Clay, Barton. Brit.

Mus. No. 49741. >» 32 diam.

.10b. Membranipora crassomuralis, n. sp. Barton Clay, Barton. Brit. Mus.

No. 49740. x 32 diam. Another specimen growing on a strongly ribbed
Pecten.

_ 11. Membranipora buski, n. sp., p. 229. Headon Beds, Colwell Bay. Brit. Mus.

No. B 4625. x 55 diam.

. 12. Membranipora buski, n. sp. Headon Beds, Colwell Bay. Mus. Pract. Geol.

Specimen with numerous ocecia. 55 diam.

. 13. Lunulites transiens, n. sp., p. 233. Barton Beds, Barton. Brit. Mus. No.

49724. »% 24 diam. View of the external layer of the zoarium.

_ 14. Lunulites transiens, nu. sp. Bracklesham Beds, Bracklesham. Brit. Mus.

No. B 4339. x 24 diam.

Fig.

» Os

BRITISH PALZOGENE BRYOZOA. 277

PLATE XXX.

. Lunulites transiens, n. sp., p. 253. Bracklesham Beds, Bracklesham. Brit. Mus.

No. B 49724. x er diam. ‘The centre of a zoarium, with the “ ancestrula.”
4

. Lunulites transiens, n. sp. Bracklesham Beds, Bracklesham. Brit. Mus.

No. B 4339. x 18 diam. In the lower part the front wall has been broken
away.

. Lunulites transiens, n. sp. Bracklesham Beds, Bracklesham. Brit. Mus. No.

49723. x o diam. Part of a worn zoarium resembling JL. urceolata,
4
Lamk.

. Biselenaria offa, nu. sp., p. 235. Barton Beds, Barton. Brit. Mus. No. 49766.

Upper surface of the zoarium. x 18 diam. Fig. 4a. A fragment of another
zoarium showing the zocecia of the under surface. Brit. Mus. No. 49766.
x 18 diam.

. Biselenaria offa, u. sp. Barton Beds, Barton. Another specimen: upper

surface. Brit. Mus. No. 49759. x * diam.

. Micropora cribriformis, n. sp., p. 236. Barton Beds, Barton. Brit. Mus.

No. B 4583. »x 55 diam.

. Onychocella magnoaperta, n. sp., p. 237. Brockenhurst Beds (Mid. Headon),

Brockenhurst. Brit. Mus. No. 49738. x a diam.
4

Cribrilina vinei, n. sp.,p. 238. LondonClay, Sheppey. Brit. Mus. No. B 4514.
(Vine’s type of Membraniporella nitida.)
Schizoporella magnoaperta, n. sp., p. 239. Barton Beds, Barton. Brit. Mus.

No. 49733. x v diam.

3

10. Schizoporella magnoincisa, n. sp., p. 240. London Clay, Hampstead. Brit.

Mus. No. B 4515. x 30 diam.

Fig. 11. Adeonellopsis incisa, n. sp., p. 247. London Clay, Haverstock Hill. Brit.

Mus. No. 49661. » 55 diam.

Figs. 12 & 13. Adeonellopsis wetherelli, n. sp., p. 245. London Clay, Haverstock Hill.

Brit. Mus. No. 49756. Fig. 12a. Azoarium, x 38 diam. Fig. 126. Upper
zocecia of the same, X 55diam. Fig. 12¢. Lower zocecia of the same, x 18
diam. Fig. 138. Basal zocecia: No. B 3832; x 18 diam

MR. J. W. GREGORY ON THE

PLATE XXXI.

ig. 1. Adeonellopsis wetherelli, n. sp., p. 245. London Clay, Fareham. Basal zocecia.

Brit. Mus. No. B4623. x 55 diam.

. 2. Lepralia lonsdalei, n. sp., p. 247. Bracklesham Beds, Bracklesham. Brit. Mus.

No. 49734. > 55 diam.

. 3. Umbonula calcariformis, n. sp., p. 249. London Clay, Fareham. Brit. Mus.

No. B 3831. x 55 diam.

Fig. 4. Umbonula bartonense, nu. sp., p. 248. Barton Beds, Barton. Brit. Mus.

No. 49741. > 565 diam.

Figs. 5-7. Teichopora clavata, n. sp., p. 249. Barton Beds, Barton. Brit. Mus. No.

Fig

sli:

jeielle

Sz

49733. x 55diam. Fig. 5. Normal zoewcia. Fig. 6. Basal zowcia: No. 49757.
Fig. 7. Part with a gonecium: No. 49659. xX 55 diam.

. 8. Meniscopora bigibbera, nu. sp., p. 251. Bracklesham beds, Huntingbridge. Brit.

Mus. No. 49732. x 565 diam.

. 9. Meniscopora bigibbera, nu. sp. Bracklesham Beds, Bracklesham. Brit. Mus.

No. 49734. 55 diam. Fragment with gonecium.

Conescharellina clithridiata, nu. sp., p. 252. London Clay, Hampstead. Brit.
Mus. No. 69594. > 18 diam.

Conescharellina clithridiata, n. sp. London Clay, Sydenham. Brit. Mus.
No. B1557.. x 18 diam. Another zoarium with oecia.

Orbitulipora petiolus (Lonsd.), p. 253. Bracklesham Beds, Bracklesham.
Brit. Mus. No. 49760. Fig. 12. Zoarium, x 4 diam. Fig. 12a. Zoecia,
x 18 diam.

3. Orbitulipora petiolus (Lonsd.). Bracklesham Beds, Bramshaw. Brit. Mus.

No. B 4349. »x 12. YZoarium with stem.

. Orbitulipora petiolus (Lonsd.). Whitecliff Bay. Specimen in Conescharellina

stage. Brit. Mus. No. B 4347.

5. Mucronella angustoecium, n. sp., p. 254. Barton Beds, Barton. Brit. Mus.

No. 49739.

. Mucronella angustoecium, n. sp. Brit. Mus. No. B 4579. x 55 diam.

PLATE XXXII.

_ 1. Smittia tubularis, n. sp., p. 265. London Clay, White Conduit House. Brit.

Mus. No. 49744. x 55 diam. Fig.1la. Nat. size. Fig. 14. Upper zoecia,
x 55 diam. Fig. le. Basal zocwcia, x 55 diam.

Fig. 2. Ldmonea bialternata, n. sp., p. 257. London Clay, Islington. Brit. Mus.

No. 49662. Fig. 2. Part of zoarium with occium. Fig. 2 6. Section.

BRITISH PALAOGENE BRYOZOA. 279

Fig. 3. Idmonea giebeli, Stol., p. 256. London Clay, Haverstock Hill. Brit. Mus.
No. 49656. X 55 diam. Fig. 5a. Part of zoarium including an occium.
Fig. 3b. Transverse section.

Fig. 4. Idmonea aff. seriatopora, Reuss, p. 258. London Clay, Haverstock Hill.
Brit. Mus. No. B 4510. Fig. 4a. Part of zoarium, X 18 diam. Fig. 46.
Mouth, x 52 diam. Fig. 4¢. Transverse section, x 18 diam.

Fig. 5. [dmonea aff. seriatopora. Back view of a zoarium, X 55 diam.

Fig. 6. Ldmonea coronopus, Defr., p. 259. Calcaire grossier, Parnes. Brit. Mus.
Fig. 6a. Nat. size. Fig. 6 6. An entire colony, x 18 diam.

Figs. 7-9. Hornera farehamensis, n. sp., p. 260. London Clay, Fareham. Brit. Mus.
No. B 3831. Fig.7 a. A zoarium, nat. size. Fig. 76. View of back, x 18
diam. Fig. 8. Part of another zoarium, X 18 diam. Fig. 9. Basal zocecia
of another specimen, x 18 diam.

Fig. 10. Entalophora tergemina, n. sp., p. 260. London Clay, Sheppey Brit. Mus.
No. B 4509. Figs. 10a & 10 6. Two views of the same specimen, 55 diam.

Figs. 11 & 12. Heteropora glandiformis, n. sp., p. 261. Barton Beds, Barton. Brit.
Mus. No. B 4511. Fig. 11. An entire zoarium, xX 18 diam. Figs. 12a &
12 6. Broken transverse sections showing internal structure. No. B 4512.
< 18 diam.

Fig. 15. Lichenopora, sp. Barton Beds, Barton. Brit. Mus. No. B 4583. » 10 diam.

The numerator of the magnifying-power ‘ fractions’ represents the original macni-
to) J to) to] S

fication, and the denominator the reduction from the size of the field of the

microscope.

VOL. X1I.—PartT vi. No. 9—June, 1893. 28

~

s-M.Woodward deLet hth West, Newman 2

OGENE

pe

in.

4%

iil
one!
ie

ra
vitae

AN Oe
my eau hh
bet

.
-

Freee

oe

PNewentewtarrin

q
H
Ff
7
%
3
Oy
y
£
5
3
°
&
a
e

/ Jf
Sramnd, <vot

oc. Lot KMS C XXXII

i
Lan a
Bic aay

iy

a

Aone ces
ie
Gree

IX. On additional Bones of the Dodo and other Extinct Birds of Mauritius obtained
by Mr. Tufopore Savuzter. By Sir Epwarp Newron, A.C.I.G., F.LS., CUZS.,
and Hans Gapow, Ph.D., M._A., F.RS., F.ZS.

Received October 31st, 1892, read November Ist, 1892.

[Prares XXXIIL-XXXVIL]

IN 1889 the Government of Mauritius appointed a Commission to enquire into the
“Souvenirs Historiques” of that island; and in furtherance of their object, at the
instance and under the able direction of their President, Mr. Théodore Sauzier, they
continued the exploration of the Mare aux Songes—the marsh in which the late
Mr. George Clark, upwards of five-and-twenty years ago, made the discovery of a vast
deposit of bones of the Dodo? and other animals, mostly now extinct, and the only
locality in Mauritius where remains of the Dodo haye been found in any quantity 2.

This exploration has been very successful, for not only have many Dodos’ bones,
some of them new and others represented only by imperfect specimens, been recovered,
but also a considerable number of the bones of other birds, materially adding to our
knowledge of those which had been but partially described, and proving the former
existence in Mauritius of species either vaguely indicated by old voyagers or wholly
unsuspected to have been members of its fauna. Besides these there have been found
many remains of the large extinct Lizard, Didosaurus mauritianus*, and several
carapaces, more or less entire, though none absolutely perfect, belonging to one or
other of the extirpated Tortoises.

Nearly the whole of these specimens have been sent by Mr. Sauzier, on behalf of
the Commission over which he presided, to the Museum at Cambridge, with a view to
their determination and to the description of such as are new, and this task has been
undertaken by the present writers.

Before proceeding to its execution, it may be as well to recall the fact that up to the
present time, beside bones of Didus ineptus, those of the following birds have been
obtained from this marsh and described as under :—

Lophopsittacus mauritianus (Owen). Lower Jaw. R. Owen, Ibis, 1866, pp. 168
et seqq.
Tibia. A. Milne-EKdwards, Ann. Sc. Nat.
sér. 5, vi. pp. 91 et segg. (1866).

' This, 1866, pp. 141 e seqq. * Proceedings of the Zoological Society, 1890, pp. 402 et seqg.
* Giinther, Journal of the Linnean Society, Zoology, xiii. pp. 322 et seqq.
VOL. XII1.—PART Vil. No. 1.— August, 1893. 27

282 SIR E. NEWTON AND DR. GADOW ON THE DODO

Astur, sp. indet. Metatarsus. Jd. op. cit. xix. art. 3 (1874).

Ardea garzetta, Linneus. Tibia. Id. loc. cit.

Aphanapteryx broecki (Schlegel). Lower Jaw, Tibia, Metatarsus. Jd. op. cit. x.
pp. 326 et seqg. (1868).

Fulica newtoni, A. Milne-Edwards. Pelvis, Tibia, Metatarsus. Jd. op. cit. viii.

pp. 195 et seqq. (1867).

All these are species which no longer occur in the island.

Bones of a species of Phenicopterus have also been found (G. Clark, Ibis, 1866,
p. 144, and A. Milne-Edwards, Ann. Sc. Nat. sér. 5, xix. art. 3).

The present collection contains not only bones of the above-named birds, but also
those of a Finch (?), an Owl, four other species of Heron, a Bittern, a Darter, a Gannet,
a Goose, a Duck, a Grebe, two species of Pigeon, one of which is probably the extinct
Funingus (Alectorenas) nitidissimus, a Water-hen, and two Petrels, of which we
proceed to describe and characterize as new :—

Stria sauziert,

Astur alphonsi,

Butorides mauritianus,
Plotus nanus,

Sarcidiornis mauritianus, and
Anas theodori.

In naming these species we wish by the first and last to commemorate the services to
science of Mr. Sauzier ; while the Astwr, being in all probability identical with that
recognized but left unnamed by Professor Milne-Edwards, may be appropriately
dedicated to him.

Of birds previously distinguished we have now for the first time the following
parts :—

Didus ineptus.—Atlas, Prepelvic or ‘ intermediate” (18th) Vertebra, complete Pubic
Bones, and Metacarpals.

Lophopsittacus mauritianus.—Sternum (?), Femur, Metatarsus, beside Lower Jaw far
larger than that first described.

Aphanapteryx broecki.— Upper Jaw, third Cervical Vertebra, Pelvis, Humerus,
Femur !.

Fulica newtoni.—Cervical vertebre (third and ninth or tenth), Sternum, Sacrum,
Humerus, Ulna, and Femur}.

* There is a large series of tibie (39 right and 50 left), which must belong to one or the other of these two
species, but except in a few cases it is impossible to distinguish between them.

AND OTHER EXTINCT BIRDS OF MAURITIUS. 283

One specimen at least of each of the bones now first described has been kindly
presented by Mr. Sauzier, on behalf of the Commission of which he is President, to
the Museum of the University of Cambridge, as well as a series of other bones in
proportion to the extent of the collection. The remainder, including a magnificent
skeleton, which has been mounted in that Museum and is doubtless the most complete
im the world, of Didus imeptus, will be ultimately deposited in the Museum of
Mauritius at Port Louis.

1. LopHopsirracus MauRITIANUS. (Plate XX XIII. figs. 1-8.)

A complete tibia obtained previously from Mauritius and having been assigned,
although not described, by M. Milne-Edwards to Lophopsittacus mauritianus, made it
easy to recognize 46 other tibie taken from the Mare aux Songes as belonging to the
same species of Parrot.

Several femora, varying from 58 to 63 mm. in length, are likewise easily referable to
the same species.

There is also a left tarso-metatarsus of 35 mm. in length, typically flattened and
broadened out, with the outer condyle turned backwards and outwards in accordance
with the reversed fourth toe. The plantar tuberculum near the proximal end of the
bone is partly broken off, but sufficiently preserved to show the two canals lying side
by side, through which the tendons of the deep flexors of the hallux and other three
digits passed. Near the inner or tibial margin of the second metatarsal is a deep
impression, caused by the insertion of the tendon of the m. tibialis anticus. The
position of this insertion, near the inner side of the second metatarsal, instead of near
the middle of the third metatarsal, is typical of Parrots. Above this impression is a
deep oblique groove, in which lodged the tendon of the m. extensor digitorum in its
oblique course from under the bony tibial bridge to the inner side of the foot. This
peculiar groove exists also in Necropsittacus rodericanus, Calyptorhynchus funereus,
Cacatua galerita, Licmetis tenuirostris, and Macrocercus macao, but apparently not in
Stringops, Domicella, or Trichoglossus, although the tendons run in precisely the same
direction, passing over the tarsus without leaving any impression upon the bone. The
erratic occurrence of this groove, intensified by age, but absent in a fully adult Stringops,
detracts from its taxonomic value.

The following measurements show that the relative lengths of the femur, tibia, and
metatarsus from Mauritius are so similar to those of other Old-World Parrots that the
bones in question can without doubt be referred to one species only. ‘The measure-
ments show also that this species was considerably larger than Mecropstttacus roderi-
canus, agreeing in the length of its hinder extremity with Cacatua galerita.

284 SIR E. NEWTON AND DR. GADOW ON THE DODO
Necropsittacus | Lophopsittacus |Calyptorhynchus | Cacatua | Paleornis
rodericanus. mauritianus. fumereus. galerita. | alewandri.
mm. mm. mm. mm. mm.
MOmur eircom erie 46-49 58, 61, 63 55 60 37 |
Q a )
Tibia ess ue ee 59-63 { Bey | 74 86 50 |
Metatarsus ............ 22 35 25 27 1. |
Total length of hind limb 127-134 181-197 154 173 105
Width of sternum at level
OH Mel) “Gapoasenac 20-0 27-5 32
Distance from spina ex-
terna to height of crista
Risa poaodsep eo coe 20-0 22:0 32
Distance from spina in-
terna to subclavian ridge 13°5 16:0 20
Greatest length of man- |
dilbles. 4 Reece wtensasrs 57 65, 71, 78 53
Greatest width of man- |
Po WebNS, Gaooeoebbno0c0 50 65 oeG 4] |
|
The most interesting part of this Parrot is the enormous underjaw. One pair of

underjaws is absolutely complete but for a few particles of bone being broken off from
the anterior margin. The left mandibles of two other specimens are nearly complete.
A fourth specimen is represented by the posterior half of the left mandible only.
These four jaws vary somewhat in size. The distance from the posterior angle (p in
fig. 5, Plate XX XIII.) to the anterior end of the complete symphysis is in the largest
and best preserved specimen 78 mm., in the next 71, in the third only 65 mm. The
smallest specimen of the extinct Mauritian Parrot is consequently still 8 mm. larger
than that of Necropsittacus rodericanus.

proportions.

The width of the mandibles shows the same
Each underjaw has a distinct additional articulating facet, about 7 mm.
in length, for the ventral surface of the outer process of the quadrate, which carries
the jugal bone. Such an additional facet, besides the usual one at the ventral end of
the quadrate, is indicated in Cacatua galerita, broad and well developed in Stringops,
Calyptorhynchus, and Ara: in fact, in many Parrots with powerful and broad underjaws.

It seems rather improbable that such an enormous jaw should be associated with a
Cockatoo of moderate size ; but, curiously enough, the comparison of the greatest length
of the mandibles with the total length of the hinder extremity shows that Necropsit-
tacus rodericanus had actually a proportionately larger jaw than the species of
Mauritius, because the length of the jaw should not be more than 50 or 51, while it
is in fact 57 mm. Of course it is hardly necessary to observe that there can scarcely
be any correlation between the length of the whole leg and the size of the bill and
head in a Parrot; but, having to deal with scanty remains of birds whose anatomical
structure is otherwise unknown, we have to be grateful for small mercies, At any
rate, we find that the Parrots from Mauritius and from Rodriguez not only resemble
each other in the proportions of the bones of their hinder extremities, but also in the

AND OTHER EXTINCT BIRDS OF MAURITIUS. 285

enormous development of their jaws, a feature which makes them unlike any other
Parrots. Moreover, these considerations enable us to discuss with some amount of
certainty, or at least probability, the only other bone of a Parrot which has been found
in the Mare aux Songes: to wit, the sternum.

The sternum is preserved only in its anterior part. The large spina externa agrees
in shape and direction exactly with that of VV. rodericanus, and excludes any possibility
of this sternum belonging to any other bird but a Parrot. The ventral margin of part
of the keel is broken off, but the line of the m. subclavius is well marked; the whole
of the anterior margin and the articulating facets for several ribs are likewise uninjured.
This sternum appears at the first glance undoubtedly far too small for L. mauritianus,
but if we measure its width in level of the first pair of ribs, the height of the keel, the
distance from the middle of the anterior margin of the sternum (at the place where the
spina interna would be if it existed in these Parrots) to the highest curve of the keel,
or to the muscular ridge at the point (Plate XX XIII. fig. 7, 8), we find that this
fragmentary sternum by all its dimensions indicates that it belonged to a larger bird
than NV. rodericanus. In fact, the size of this sternum would fit one of the smaller
specimens of LZ. mauritianus; and this is corroborated by the following calculation,
which gives a result which we should not have expected :—Average of total length of
hind limb of WV. rodericanus (150): width of its sternum (20)=Length of hind limb of
smallest L. mauritianus (181): width of its sternum would be 27:8, while our single
sternum from Mauritius actually measures 27°5 mm. across!

There can be no doubt that the extinct Mauritian Parrot was a larger but other-
wise nearly allied form of the Parrot from Rodriguez; it is, however, questionable
whether both might not be included in the same genus Necropsittacus, for while we
know from old drawings that the Mauritian form had a sort of ornamental crest, we
know nothing to the contrary of V. rodericanus.

2, ASTUR ALPHONSI, sp. nov. (Plate XX XIII. figs. 9, 10.)

Amongst numerous Asturine remains a pair of tibia, a pair of metatarsals, and the
metacarpals of the left side are probably referable to one individual bird of prey.
The two metatarsals, with a length of 81 mm., agree perfectly with that figured by
M. Milne-Edwards (plate 53. fig. 2). He rightly referred them to the genus Astur, and
remarked that they belonged to a bird which was undescribed and unknown, unless it
was identical with A. melanoleucus from the Cape of Good Hope. We have been able
to measure the length of the tarso-metatarsus of an A. melanoleucus, and have found
that it agrees in this respect with the two bones in question. It would therefore seem
reasonable to assign these bones to A. melanoleucus, unless the absence of this South-
African species from Madagascar, and the numerous instances of insular forms or species
of Hawks, be deemed arguments sufficiently strong to distinguish the bird to which
these bones belonged as Astur alphonsi.

286 SIR E. NEWTON AND DR. GADOW ON THE DODO

The greatest length of the two tibie is 117 mm., which agrees proportionately with
that of the two metatarsi, so as to justify us in connecting them with each other as those
of the same Hawk, a view which is corroborated by the tibial and metatarsal articulating
facets fitting well upon each other. The bony bridge for the m. flexor digitorum
communis is very strong, the fibula reaches far down the tibia, the peroneal crest is
straight and long, the cnemial crest slants gradually into the anterior inner edge of the
shaft of the tibia.

It is of course impossible to state with certainty whether the metacarpal bones, the
total length of which is 55 mm., belong to the same individual ; that they belong to the
same species is more than probable, and that they are those of a diurnal bird of prey of
the size of Astur melanoleucus is unquestionable. All the facets, tendinous impressions
and processes, and the sharp, blade-like, deeply scooped-out third metacarpal bone mark
the specimen.

3. STRIX SAUZIERI, sp. nov. (Plate XX XIII. figs. 11-18.)

The Owls are generally classified according to cranial, sternal, and various purely
external characters. None of these points will serve our purpose, because the only
bones of Owls in the present collection are those of the humerus, tibia, and metatarsus.

There is one character, namely the relative length of the tibia to that of the meta-
tarsus, which is not only very constant but also very characteristic of the various
families and even genera of Owls. From the quotient resulting from the division of
the length of the tibia by that of the metatarsus we have come to the conclusion that
the majority of the bones in question, namely four metatarsals, three tibie, and, by
inference, two humeri, belonged to a member of the long-footed Owls, of which Stria
flammea and its allies is the most pronounced type, while Heliodilus soumagnii from
Madagascar closely approaches it, to the exclusion of Carine murivora from Rodriguez,
Scops, Sceloglaua nove-zealandie, Spiloglaux, Gymnoscops, Asio, and Bubo as examples
of the several subfamilies and principal genera of the so-called Bubonide.

We have much pleasure in distinguishing this new Owl from Mauritius as Strix
sauzieri, referring it to the genus Strix, and not to Heliodilus, on the strength of most
of those very characters which induced M. A. Milne-Edwards to establish the new
genus Heliodilus!. These characters are, first, the relative length of the tibia

1 It will be convenient to mention here at least those characters which could be tested with the material at
our disposal :—

‘“ T/Heéliodile est un Strigide 4 pattes robustes, 4 ailes plus courtes et a téte plus large que les Effraies( Strix).
Le tibia est plus long et les proportions en sont différentes, car l’extrémité inférieure est plus robuste et le
corps de l’os est aussi gréle; la eréte péronniére est courte et le péroné ne se prolonge pas autant que chez les
Chouettes et les Hiboux. Si l’os de la jambe est plus long que celui de |’Effraie, celui du pied est au contraire
plus court ; mais ses caractéres sont & peu prés les mémes que dans ce dernier genre.”—-A, Milne-Edwards,
Comptes Rendus (1878), vol. 85, p. 1282.

AND OTHER EXTINCT BIRDS OF MAURITIUS. 287

?

and metatarsus; secondly, the length of the “péroné” or fibula, which, at least
in the two larger specimens, is continued far beyond the level of the tubercle
of the hallux attachment, as far down as the epicondyle; thirdly, the relatively
greater length of the peroneal crest, which in our specimens extends to the end
of the upper third of the tibia, while in Heliodilus it ends a little below the upper
fourth ; the actual peroneal connexion, ¢. ¢. the ridge of the tibia which touches the
fibula, is absolutely and relatively larger in our specimens than on the tibia figured by
M. Milne-Edwards (Grandidier, Ois. de Madagascar, plate 36 ¢, fig. 8).

On the other hand, there are differences, notably the longer and higher cnemial
process of the tibia and the shortness of the humerus, sufficient to justify the specific
distinction of this Mauritian Owl from Stria flammea, with its numerous varieties.

Humerus. Tibia. Metatarsus. Quotient. in ue
Metatarsus.
mm. mm. mm. mm.
ISuHb-< EWE do cogo bobEee ll 90, 92, 93 | 63, 63, 64, 64, 66 1:42 \ Longest Metatarsus.
HM Hovoapccas saves 46 a 56 pair.
NEM, 5.50.5 5000 84 85 60 1:42
Heliodilus soumagnil...... 72 87 57-60 1:52
Athene murivora ........ | 64-69 69-76 41-46 1:65-1:70
Asio capensis ............ ae 95 56 1:70
Scops rutilus ............ errr 50 28 1:80
Sceloglaux novee-zealandiee . 58 64 35 1°83
Bubo virginianus ........ ) 163 146 75 1:94
madagascariensis ....| 80 82 41 2:0 Y Shortest Metatarsus.

The pair of metatarsi measuring 56 mm. in length are at the same time much more
slender than the other five metatarsi. We do not feel justified in explaining this
considerable difference in size and strength by difference of age, because the bones
are fully ossified and show all the characteristic markings in the same pronounced
degree. Only the bony bridge over the tendon of the m. tibialis anticus is broken,
and was moreover certainly incomplete in both of the smaller metatarsi. We
naturally tried to fit the shortest tibia of 90 mm. length on to the metatarsus of the
corresponding side, but the tibial condylar facets are a little too large. If they fitted,
the quotient of this shortest tibia with the shortest metatarsus being 1:61 would
indicate an Owl different from any of those which are mentioned in our list. Asio
capensis cannot be thought of, because its metatarsus is several times stronger than
the two in question, nor do we feel inclined to explain the shortness and slender shape
of these two bones by sexual difference of Strix sauziert. Unless we assume, what is
unlikely, that the island of Mauritius possessed two different species of Strix, we have
to conclude that the short pair of metatarsals belonged to a small individual of Strix

288 SIR E. NEWTON AND DR. GADOW ON THE DODO

sauzieri, although it is rather improbable that this species, restricted to a small island,
varied as much as British specimens of Strix flammea, of which latter the British
Museum Catalogue records the length of the ‘ tarsus” as 2:2 inches, 7.é. 55 mm.,
while the measurements taken from an English specimen in the Cambridge Museum
give the length of this bone as 60 mm.

4, PLOTUS NANUS, sp. nov. (Plate XXXIV. figs. 1-5.)

The humerus, pelvis with sacrum, and tibia of the genus Plotus possess so many
diagnostic characters that the three bones figured on Plate XXXIV. can easily be
recognized as belonging to this genus of Steganopodes.

The Humerus shows the following characteristic points:—The sulcus transversus is
very deep and strongly marked, extending from the tuberculum mediale halfway
across the head of the humerus as a groove of equal width and depth. The crista
superior is straight, and shows well-marked impressions of the insertions of the great
pectoral muscle. The supracoracoidean or subclavian muscle has an inserting surface
upon the corner where the caput humeri meets the proximal end of the crista superior.
The tuberculum inferius s. medianum is a very prominent knob, serving on its dorsal
and ventral surfaces for the attachments of the m. coraco-brachialis posterior and
m. biceps humeri respectively. The pneumatic foramen hes at the bottom of a wide
and deep recess. ‘The dorsal lip of this recess is sharply marked by an oval impression
from the tendon of the m. scapuli-humeralis posterior (m. infraspinatus, m. teres major,
of other anatomists); from this impression the low but sharp ridge for the m. latissimus
dorsi is continued down the middle of the ventral or inner surface of the humerus.
The two grooves above and upon the ventral surface of the outer and inner condyles
are produced by the origins of the pronator and short flexor muscles of the forearm.
The m. brachialis inferior s. internus arises from a strongly marked impression on the
dorsal or outer surface of the distal part of the shaft of the humerus.

The Pelvis and Sacrum are easily referred to the genus Plotus by the deeply notched
or curved lateral margin of the pre-acetabular part of the illum, the prominent and
sharp antitrochanter, the sharp ventral ridge springing from the three anterior sacral
vertebra, and by the position of the single primary sacral vertebra closely behind the
acetabular axis. The individual peculiarity of the specimen described is the lopsided
position of the two halves of the pelvis with reference to the sacrum.

The 7ibia is much flattened anteriorly ; its anterior or cnemial crests are high, but
not anchylosed with the patella; the peroneal crest for the attachment of the fibula is
long and straight. ‘The condylar portion of the tibia is turned considerably inwards,
and the bridged-over groove for the passage of the tendon of the m., extensor digitorum
communis is very deep and placed obliquely.

There remains the question of the specific differences of the bones before us. They
all belong, to judge from their appearance, to one adult individual, but their small

AND OTHER EXTINCT BIRDS OF MAURITIUS. 289

size excludes at once the possibility of their being referable to any of the species
hitherto known, as the following measurements will show. We distinguish it, on
account of its small size, as Plotus nanus.

Plotus nanus, P.anhinga. P. melanogaster. P. nov.-holland.

mm, mm. mm, mm.
Meeftwh tm erms’ cs epeuckeeieier-s oxotersereyetvers 89 112 132 120
Teftitibiayyenttcetcatcetsc auaterie een ore 61 77 86 78-95
Distance from acetabular axis to ante-

TIOTENd [OL SAChUMNs se hclyelae cic ite 30 35
Distance between ventral inner margins

ofstheracetabulageereneeririe cece 145 15

5. PopIcEPES, sp. inc.

The proof of the former existence of Grebes in the Mare aux Songes rests upon one
single bone only, the right ulna; but the latter agrees in all the essential points with
the corresponding bone of the genus Podicepes, notably the configuration of the
proximal and the distal articulations, and the existence of a sharply marked groove at
the upper outer distal condyle for the passage of the tendons of the extensor muscles,
and differs in all these characters from the corresponding bones of any other birds
which might otherwise possibly be taken into consideration, that doubts are excluded.
The total length of this ulna is 82 mm. It is consequently far too long and strong
for P. pelzelni, P. minor, or P. philippensis. On the other hand, it is much too
short for P. cristatus and by 10 mm. shorter than that of P. rujicollis. It is,
however, slightly longer than the ulna of either P. cornutus or P. auritus, so that it
probably belonged to an insular form of one of these last-named species.

6. BUTORIDES MAURITIANUS, sp. nov. (Plate XXXIV. figs. 6-8.)

It is surprising that of all the Ardeine bones, referable to at least twelve individuals
of five different species, none belong to Ardea (Butorides) nigricollis, the only species
of Heron at present existing in Mauritius, while a pair of ulne, one radius, four
metatarsi, and one coracoid must be considered as belonging to a species of short-
footed Heron hitherto unknown. ‘The bones in question are all considerably shorter
than the corresponding bones of A. (Nycticorax) megacephala. The metatarsi agree
otherwise in every detail with those of the latter species; this relative stoutness
indicates that they belonged to a Night-Heron or Bittern like A. megacephala. The
two ulne cannot, unfortunately, be compared with those of A. megacephala; their
length, 110 mm., compared with the length of the humerus of A. megacephala,
119 mm., shows, however, likewise that they were those of a considerably smaller
bird.

VOL. Xii.—Part vu. No. 2.—August, 1893. 20

290 SIR E. NEWTON AND DR. GADOW ON THE DODO

The single left coracoid agrees in all the features of its dorsal or scapular half with
A. megacephala, but its ventral or sternal half differs considerably, first by the much
more strongly marked ridge of the linea intermuscularis on its ventral surface, secondly
by the almost straight instead of inwardly curved margin between the processus
lateralis and the lateral distal corner of the sternal articulation, thirdly by a very low
but very distinct and sharp ridge which arises from the median margin of the coracoid
a little above its median articulating corner. ‘This roughness or prominent ridge is
entirely absent in A. megacephala and in all other Herons which we have been able
to examine, but at least a slight indication of it occurs in an individually varying
degree in Nycticoraz and in Botaurus. That this coracoid bone belonged, however,
to an Ardeine bird is clearly indicated by its whole configuration, notably by the
shape and position of the precoracoid process, the various articulating facets at the
dorsal end, and the prominent lip on the visceral or internal surface of the median
portion of the sternal articulating facet.

Butorides mauritianus. Nycticorax megacephala.
mm. mm.
Meng thvotulna i. .acneeiae 111-112 oe
Length of metatarsus...... 81-87 96
Length of coracoid........ 48 55

7. SARCIDIORNIS MAURITIANUS, sp. nov. (Plate XXXIV. figs. 9, 10.)

The most tangible proof of the former existence of this form in the island of
Mauritius rests at present upon one specimen of left metacarpal bones. However,
this solitary specimen is sufficiently well preserved to show its affinities by various
well-marked characters. It agrees in size with the corresponding combination of
bones of Bernicla brenta, while it is considerably smaller than those of the common
domesticated Anser cinereus, and too large for the Madagascar and Kast-African
Sarcidiornis africanus and the Indian 8. melanonotus.

The generically diagnostic feature of the bones of the middle hand of Sarcidiornis
is the very prominent process which arises from the side of the first metacarpal,
proximally from the articulating facet of the pollex. The apex of this process is
covered in Sarcidiornis by a skin, which, although thickened and bare of feathers, is
not transformed into a horny callosity or spur. The same peculiar feature exists in
the bone before us; the apex is rough and irregularly shaped, and since this part of
the process never serves for the origin or insertion of muscles or tendons, its roughness
plainly indicates the same purpose as that of the Madagascar species of Sarcidiornis,
namely its use as a fighting knuckle, although in an either arrested or incipient state.
Such a weapon, furnished with a sharp and long horny spur with bony core, carried
by the first metacarpal bone, is fully developed in Chauna. This American genus

AND OTHER EXTINCT BIRDS OF MAURITIUS. 291

can of course not be compared with the bird from Mauritius except by analogy. The
only truly “spurred Goose” of the Ethiopian region is Plectropterus; but the spur is
carried by the radial carpal bone and therefore is at once removed from comparison
with our specimen, which belonged to a rather large-sized species of Sarcidiornis,
and, having probably been restricted to the island of Mauritius, may be distinguished
as S. mauritianus.

Another part of this bird consists of the somewhat incomplete left half of the
pelvis ; it agrees in size with that of Bernicla brenta, consequently by inference with
Sarcidiornis, measuring 70 mm. from the anterior brim of the acetabulum to the
posterior end of the os ischii. ‘The few characters which are preserved in this portion
of a pelvis agree with those of Anas and Anser and other Lamellirostres.

8. ANAS THEODORI, sp. nov. (Plate XXXIV. figs. 11-17.)

The fragment of a sternum, a pair of coracoids, eight humeri, and a pair of
metatarsi are referable to a Duck which was considerably larger than Netiapus
auritus, Anas berniert, and Dendrocygna, but smaller than Anas melleri, of which
we have a skin and breast-bone with shoulder-girdle for comparison.

Of the sternum only the anterior portion is preserved, which is, however, sufficient
to show its affinities. Its width between the two lateral muscular ridges of the
sternum is 28 mm., 7. e. slightly less than in Anas melleri and agreeing with Dendro-
cygna arcuata; the sternum differs, however, from that of Dendrocygna by its well-
developed, although broken-off spina externa, by its lower keel, and lastly by the
much smaller and shallower entrance to the pneumatic foramen; it differs also from
that of A. melleri by the lesser height of the keel, moreover by the shape and
direction of the anterior margin of the latter.

The single left coracoid is in a perfect state of preservation and fits well into the
sternal fragment, so that it might belong to the same species, although certainly not
to the same individual. This coracoid differs from that of Dendrocygna by its greater
length, by the shape of its sternal end, and by its very smooth, almost plain ventral
surface. Nettapus and Anas bernieri are to be excluded on account of their much
smaller and shorter coracoids. The coracoid in question is much shorter than that of
A. melleri, but it agrees closely with the latter by its shape, and especially by the
almost plain ridgeless ventral surface of the shaft.

The seven humeri are much like each other, but vary from 70 to 78 mm. in their
greatest length; they are exactly of the same shape and size as those of different
specimens of Anas punctata, i. e. of a much smaller Duck than A. melleré.

The two metatarsi are in a bad condition; the right one measures 42 mm. in length,
indicating a bird much more short-footed than A. melleri.

202

292 SIR E. NEWTON AND DR. GADOW ON THE DODO

A. theodori. A, melleri.

mm. mm.
Width between pectoral ridges ............ 28 30
Distance from spina interna to top of crest .. 20 25
Pengthrofcoracoid sy ej crete ese siete eieleiere 42 52
Wen gthyofphumoerusimerrrerenrecrersitrrl eel: 70-78 89
Length of metatarsus ..............-..- 42 41

The result of our investigations is that the bones figured (Plate XX XIV. figs. 11-17)
belong to a Duck which differs from any of those found in Madagascar, while it agrees
more closely with A. mel/eri but for its dimensions, which are so much smaller that they
cannot well be accounted for by individual variation. It is moreover the only Duck
of which remains have hitherto been found in Mauritius; we distinguish it therefore
as Anas theodori, in honour of Mr. Théodore Sauzier.

9. FULICA NEwTonI. (Plate XXXV. figs. 1-11.)

The remains of this large Coot are numerous. The femur, sternum, humerus, and
four cervical vertebre are new and hitherto not described, while the whole pelvis and
sacrum, the tibia, and the metatarsus have been described by M. Milne-Edwards. The
bones belonged to at least 24 different individuals, and show accordingly a considerable
amount of variation in their dimensions.

The smallest thigh-bone is 76, the longest 90 mm. long. ‘The latter is larger in all
its dimensions, otherwise alike the others. The outer or superior trochanteric crest
is high and curved inwards; the two principal arms of the tendinous loop for the
m. biceps cruris have left two very distinct impressions on the lateral surface of the
distal end of the shaft and near the popliteal region. The external condyle has a
deep and smooth notch for the reception of the head of the fibula.

The pelvis agrees with that of Fulica proper and with that of Tribonyx because of
the peculiar dip of the dorsal margin of the pre-acetabular ilium, which does not reach
up to the level of the dorsal spinous processes, leaving a long groove through which
passed the tendons of the usually obliterated dorsal spinal muscles. In the possession
of this groove and in its elongated and laterally contracted shape this pelvis agrees
with that of a typical Fulica, and it differs much from that of Aphanapteryx and
Ocydromus, while the pelvis of Porphyrio melanonotus and that of Tribonyx are less
contracted than in Fulica and Gailinula.

The sternum of F. newtoni resembles in several points that of Aphanapteryx, Ery-
thromachus, and Ocydromus, and differs from Tribonyx, Fulica proper, and Porphyrio,
first in the configuration of the whole anterior margin of the sternum, especially in
the double or basally divided spina externa, which is moreover broad and flat, while in
the other genera this spine is single and furnished with a ventral longitudinal sharp
ridge; secondly, by the receding and broad anterior margin of the keel, which,

however, is well developed, although less than in Tribonyx and Fulica atra, but the
tendency towards a reduction of the keel is apparent.

AND OTHER EXTINCT BIRDS OF MAURITIUS. 293

The shortest humerus is 8°5, the longest 92 mm. in length. They all differ from
that of Aphanapteryx by being far less curved, stronger throughout, and furnished with
a large pneumatic foramen; the sulcus transversus upon the head of the humerus is
deeper, but the tuberculum medium is lower.

The cervical vertebre of the Rallide can be easily recognized by their shape and by
the numerous articulating facets, processes, and median crests. On the whole, these
vertebre of F. newtoni resemble more closely those of Porphyrio and Ocydromus than
those of Fulica proper. Dorsal spinous processes are absent in the 9th and 10th
vertebrae, they are rather low in the 6th and 5th, sharp and high in the 4th. Ventral
median processes are absent in the 10th to 5th, high in the 4th and 3rd.‘ The latter
two vertebre are marked by a deep round notch on each side, this notch being often
turned into a complete foramen. Most of the lower and middle vertebre of the neck
are very broad in comparison with their length.

In conclusion, we feel inclined to think that the Fulica newtoni combines important
characters of the true genus Fulica with those of Porphyrio, Tribonyx, and Ocydromus,
and that on the whole it more resembles these last three than the true Waterhens.

10. APHANAPTERYX BROECKI. (Plate XX XV. figs. 12-20.)

Only the tibia, tarso-metatarsus, and underjaw were hitherto known, described and
figured by M. Milne-Edwards. Besides the tibiee and tarso-metatarsi of many indi-
viduals, we have now before us the pelvis with sacrum, femora, and humeri; one
sternum, one third cervical vertebra, and one nearly complete premaxilla, together with
fragments of the upper and lower jaws. It has been comparatively easy to determine
most of these new bones because of their close resemblance to the corresponding parts
of Hrythromachus.

The pelvis, with the sacrum, of one specimen is extremely well preserved. It is much
more compact, stouter, shorter, and broader than that of Fulica newtoni; the dorsal
margin of the pre-acetabular part of the ilium reaches up to the dorsal spinous crest of
the anchylosed presacral vertebra, as is also the case in Ocydromus and Porphyrio. In
their general configuration the pelvis and sacrum of Aphanapteryx agree with Erythro-
machus.

The femur, essentially similar to that of Fulica, Porphyrio, and other allied Rails,
can be distinguished from that of Fulica newtoni by its smaller dimensions.

The tibia and metatarsus, having been described and figured previously, need not be
commented upon, beyond stating that they, especially the metatarsus, are relatively
stouter than those of Fulica.

The sternum and humerus are of particular interest, because of their small size, and
because of the absence of any large pneumatic foramina, indicating that this bird was
devoid of the power of flight. The sternum is not complete, its posterior portion being
absent. Its width across the level of the first rib behind the anterior lateral process is
only 25 mm.; the keel is very much suppressed, with its anterior margin broadened

294 SIR E. NEWTON AND DR. GADOW ON THE DODO

out and deeply grooved, as in Erythromachus. There is no trace of a spina interna ;
the feet of the coracoids, as indicated by their facets, were separated from each other
by a smooth groove of 9 mm. in length. The spina externa is represented by two
projections from the ventral lips of the median corner of the coracoid articulation. In
this respect Aphanapteryx agrees with Erythromachus, and also with Fulica newtoni.

The humerus is very short and slender for so large a bird; its typically Ralline
characters are, however, obvious enough to recognize it as belonging to Aphanapterya,
while it differs by its far greater length and strength from the humerus of Ga/llinula,
and by its much smaller dimensions from that of Fulica. A very interesting feature is
the absence of the usually wide and deep pneumatic foramen, which is indicated only
by a shallow depression which is smaller than even in Gallinula chloropus.

The third cervical vertebra could easily be recognized as such by its numerous
Ralline characters, which in these birds are strongly pronounced ; its dimensions remove
it from either Fulica or Gallinula, ti. e. from the only other Ralline birds hitherto
known to have occurred in Mauritius.

The premaxilla fits well upon the several fragments of underjaws, and still better
upon the underjaw figured by M. Milne-Edwards. The great length and the shape of
these bones closely resemble those of Hrythromachus (Phil. Trans. vol. 168, pl. xliii.
fig. A). The Mauritian bird is, in fact, nothing but a larger species of the same genus.

A number of measurements are given in the subjoined Table in order to aid the
comparison of the Mauritian Ralline birds with each other and with some of their

allies.
: . | Aphanapteryx Porphyrio Ocydromus
Fulica newtoni. | “4 aac # TEES apa:
mm. mm. mm. mm.

hens thvotapelvisherrsncre ieee ag 80 60
Distance between pectineal processes . 23:0 28°5
Distance across lat. dorsal iliac pro-

Gest Ri i reetaate ces PARRA 28 44
Distance across antitrochanters...... 32 38
IGM Orie Losoabdoanacodos 76, 78, 81, 90 69, 70,71 83 82

a og 120, 127, 98, 102, 108, |

Lenigihivo! tibialis eee ene { ices | mas yp 1
Length of tarso-metatarsus ........ 82-84 79 98 62
Ikensthvof humerus ta. eee 85, 88, 90, 92 60-66 88 57
tens thvotyulnaieemermce re ciccn eeet 74
Total length of sternum .......... 70 incomplete 68 55
Width of sternum froma toa ...... 29:5-30 25 22 20
Distance between coracoids at sternal

PINCH, sogccauscscedonoous 5 9 25
Greatest length of 9th cervical vertebra 17-21
Greatest width of 3rd cervical vertebra 15 9°5
Greatest length of 3rd cervical vertebra 16 125

AND OTHER EXTINCT BIRDS OF MAURITIUS. 295

11. TrocazA MEYERI.

Five fractured breastbones can easily be recognized as Columbine by the combination
of the following characters :—The very high and at the same time slender or thin crista ;
the presence of a well-developed spina interna, which is broad at its base, ending
anteriorly obtuse and slightly bifurcated, while the spina externa is by far less developed ;
lastly, the deep and regular grooves for the articulation with the coracoids, which do
not meet each other, but are separated by a smooth, scarcely prominent, median
ridge.

In order to determine the species we have compared the bones with those of Turtur
picturatus, Vinago australis, Funingus madagascariensis, and Trocaza meyeri. The
bones of the first three are figured by M. Milne-Edwards, of the last two the Cambridge
Museum possesses skeletons. The result of this comparison showed that four breast-
bones belong to Trocaza meyer.

In this species there exists a small but distinctly prominent tubercle on the labium
internum of the anterior margin of the sternum, midway between the anterior end of
the spina interna and the base of the lateral anterior process of the sternum ; it serves
for the attachment of the inner accessory sterno-coracoidal ligament. This tubercle
is well developed in the four breastbones, as in Trocaza, very small in Funingus, and
absent in Turtur and Vinago.

A second specific character is afforded by the spina externa of the sternum, which is
well developed in Trocaza and Vinago, small in Turtur, absent in Funingus.

The measurements on the following page show, moreover, additional characters,
which led to the determination of the species to which the breastbones belong.

Three tarso-metatarsal bones likewise are referable to Trocaza meyert, because of
their length and the configuration of the bony ridges and furrows on the posterior side
of the proximal end of the tarso-metatarsus, serving for the passage of the various long
flexor muscles of the toes. In this respect Trocaza agrees, but for the length of the
bones, with Turtur, and differs considerably from Funingus.

12. FUuNINGUS, sp. inc.

One sternum, unfortunately very incomplete, consisting only of the anterior end,
with the anterior margin of the sternum and the anterior margin of the keel, may
possibly belong to Fumingus, chiefly on account of the absence of the lateral tubercles
of the spina interna, and because of several of its dimensions as given in the following
Table, in which this specimen is marked M.S.=Mare aux Songes, while E.N. indicates
obtained in the flesh by Sir Edward Newton, and M.E. that the specimen has been
figured by M. Milne-Edwards.

296 SIR E. NEWTON AND DR. GADOW ON THE DODO

{

Vinago. | Turtur. Puningus. ° Trocaza.
M.E. | M.E. |E.NJM.E] M.S. |E.N. 9 MS.
mm. mm. |mm.|mm.| mm. | mm. |/mm.mm.mm. mm. mm,

Distance ab from anterior
end of spina interna to
pneumatic foramen ....| 7:5 8:0 8 | 10-5) 10-0 8 | 11 105 10 9 9

Distance ¢ ¢ across the ster- Hoe of ; i
num in level of the 2nd He i :
thoracic mibsir nei 20°5 18 18 }20 |broken| 21 | 26 26 25 incomplete

Distance af from anterior : : H p
end of spina interna to H : H
anterior ventral curve of Baer H
crista sterni .......... a bo 22% 2.0] 729. 28 | 31 34...31... 285 incomplete

Length of tarso-meta- :
CAYSUS! ener reivaseke rs 23 26°5 SMA eye bree x 34 34 |. 285

There are also five ulne of a Pigeon, which could not, however, be further deter-
mined except that they belong to either Trocaza or Funingus.

13. Dipus 1neptus. (Plates XXXVI., XXXVII.)

The new material of bones of the Dodo has enabled us to add to the restoration of
the skeleton the following parts which have hitherto not been known :—

1. The median distal portion of the furcula appears to be devoid of an “‘ apophyse
médiane” or hypocleidium, this region being rounded off. This may, however, be a
case of individual variation, considering that in the male specimen of the Cambridge
Pezophaps there is likewise no apophysis, while there is one, although small, in the
female specimen. Hence, Professor Owen’s restored drawing of this part in the British
Museum specimen of Didus cannot be pronounced to be incorrect.

2. Metacarpal bones of the right and left side, and the first phalanx of the second
finger. ‘These bones present no remarkable features, and agree in their small size with
the much reduced state of the other bones of the wings. There is, moreover, no
evidence of the existence of those peculiar exostoses on the distal end of the radius
and on the first metacarpal that are so characteristic of the male Solitaire, which
probably used them as fighting-knuckles.

3. The distal third of the pubic bones.

4, Phalanges of the toes (hitherto known from the Oxford specimen only).

5. The atlas or first cervical vertebra.

The most interesting result of the examination of the bones entrusted to us by
Mr. Sauzier is the determination of the number of vertebre and ribs which belong to
the various regions of the skeletal axis.

Hitherto our knowledge of these parts has rested upon the mounted specimen in the

AND OTHER EXTINCT BIRDS OF MAURITIUS. 297

British Museum (and the restored drawings by Professor Owen in the Trans. Zool. Soc.
vi. and vii.), which is faulty, and upon the Cambridge skeleton, which was incomplete.
Hence all the references to the number of vertebrae and ribs are also at fault (cf.
Fuerbringer’s ‘ Untersuchungen fiir Morphologie und Systematik der Vogel,’ tabb. xxi.
& xxii. pp. 778-781 ; and Bronn’s ‘ Thier-Reich, Vogel,’ p. 950).

The vertebre examined by us belong to an unknown number of individuals. More-
over, it is not possible to pick out a complete series from the atlas to the pelvis, which
without doubt belonged to one and the same individual. Lastly, it is a curious mishap
that only a single specimen has been found of that vertebra which fits into the gap
between the last of the three anchylosed thoracic vertebre and the first vertebra which
is overlapped by and fused with the pelvis.

The determination of the number of vertebre composing the various regions of the
vertebral column has consequently to rest upon circumstantial evidence. An unbiased
collection of facts from other Pigeons reveals certain correlations of number and shape
of vertebree and ribs, and the results thus gained can be applied to the restoration of
the Dodo’s skeleton with a considerable amount of probability.

It seems to be the rule in normal (not domesticated) Pigeons that :—

1. The 15th, 16th, and 17th vertebre are anchylosed together.

2. The 18th vertebra is free, articulating in front with the 17th, and behind with
the 19th vertebra, which latter in all cases is overlapped by, and partly fused with,
the pelvis. For the sake of convenience the 18th may be called the intermediate
vertebra.

3. The 14th and 13th vertebre each possess a spinous process which is hook-
shaped.

4, Complete ribs, 7. e. such as articulate with the sternum, vary from 3 to 4 in
number, and are restricted to the 15th to 19th vertebre, while the 16th to 18th always
carry complete sternal ribs.

5. Cervico-dorsal vertebre are those which carry movable short ribs; the dorsal
portion of such a rib articulates by a typical capitulum and tuberculum with one
vertebra, while the ventral or distal half of the rib is lost. Asa rule, at least, the last
of these short ribs carries an uncinate process. ‘The number of cervico-dorsal vertebrae
is two, rarely three.

6. The other neck-vertebre are true cervical vertebre ; with the exception of the
atlas and the epistropheus, they all possess a transverse foramen and immovable rib-
rudiments.

7. In recent Pigeons the last or hindmost pair of complete sternal ribs is frequently
followed by one pair of ribs which, attached to the 19th, or Ist pelvic, vertebra,
almost reaches the sternum: in rare cases there is present even a second, although
much shorter pair, which then belongs to the 2nd pelvic or 20th vertebra.

The following Table will show these modifications :—

VOL. XII.— Part vil. No. 3.—August, 1893. 2x

298 SIR E. NEWTON AND DR. GADOW ON THE DODO

C indicates the last true cervical vertebra. st indicates a sternal rib.
h ,, a hooked spinous process. u » an uncinate process (not mentioned
r » a short rib. on the sternal ribs).
Serial number of vertebrae .......... 12 13 14 15 | 16 | 17 18 19 20
~ inter- 1st | 2nd
anchylosed mediate. | pelvic. | pelvic.
ColumbarliniG nemo eine eit C,h hr r,w ft st aaa no rib
PROPS Chalcopterd (-\-lere\ojolelel-erel- lols! C hr \hyrju st » |longrib
Didunculus strigirostris ..........4. C hr | hr | st ei omrib
WRG ONE bosboodoso0gbboGc0eD A Ch | hr | St 3
Carpophaga pacifica ..........6+-- oe) || Chia ase | et hee | onrab
Gounalcononatdrc ieee iit as Cah | hs 7 | st a
aia i short
Pezophaps solitaria, $ ............ Ch \h,r,u | st st Aas
—— ——, Preece cere ces eeees ae C,h | hr | st st ro
Didus ineptus, properly restored, Cam-
bridge and Mauritius Museums....} .. Ch h, r st : sé |no rib
ihe, Ietnn Mornin, isa) Tl ft aes H
Sir Richard Owen, Trans. Zool. ;
2 i almost
NOCH ae retrwesl giao heac uss delle kets neue nercuete C r 7, U st st st st st ay

Didus agrees with Pezophaps in possessing 13 cervical vertebre, 2 short ribs,
4 sternal ribs, the last being carried by the first pelvic vertebra.

Treron, Carpophaga, and Gowra agree with each other in having 13 true cervical
vertebree, 2 short, 2 sternal, and 1 almost sternal pairs of ribs. They differ from
Didus and Pezophaps in the latter pair of ribs being withdrawn from the articulation
with the sternum.

Columba, Phaps, and Didunculus differ from the others in having only 12 true
cervical vertebra, 2 short, 4 sternal, and 1 almost sternal pair of ribs, because their
15th or first anchylosed vertebra (instead of the 16th or 2nd anchylosed vertebra)
carries the first pair of sternal ribs.

The restoration in the Trans. Zool. Soc. vi. pl. 15 contains one pair of sternal ribs
and one vertebra (the 15th in the figure) too many.

In conclusion we wish to say that, beside the Birds’ bones here described, the
explorations of Mr. Sauzier have produced very many bones of Reptiles, which will be
treated of by one of us in a subsequent paper, together with a considerable number of
shells of Mollusks, portions of Crustacean integument, and a few pieces of Coral.
The presence of these marine forms in the soil of the Mare aux Songes may be, it is
believed, attributed to the action of Land-Crabs, for there is no reason to think that

AND OTHER EXTINCT BIRDS OF MAURITIUS. 299

the sea has ever had access to the lake, from which it is separated by a ridge of some
height, while it is known to be the habit of those creatures to convey animal remains
a long distance inland from the shore. Nevertheless it may be as well to name the
shells as determined by Mr. A. H. Cooke, Honorary Curator of Conchology in the
Cambridge Museum. They are as follows :—Lanp Mottusca, Gibbulina sulcata, Lam..,
Pachystyla inversicolor, Fér., Cyclostoma carinatum, Lam. : Marine Mo.uvsca,
Cyprea caput-serpentis, L., Nerita polita, L., Turbo sp. incert. (fragment). In addition
to the foregoing the seeds of several plants have also been found, but these appear not
to need enumeration ; nor do the specimens of the soil collected by Mr. Sauzier (though
all are carefully preserved in the Museum) seem to call for any particular remark on
the present occasion.

EXPLANATION OF THE PLATES.
PLATE XXXIII.

Figs. 1-8. Lophopsittacus mauritianus (p. 283).

Fig. 1. Left tibia, front view. /, attachments of the transverse ligament across the
long extensor tendons.

Fig. 2. Right femur, posterior view.

Fig. 3. Left metatarsus, plantar surface.

Fig. 4. Left metatarsus, dorsal surface. g, groove for the tendon of the musc. extensor
digitorum ; @, insertion of the tendon of the m. tibialis anticus,

Fig. 5. Dorsal view of the underjaw. jp, posterior angle; f, facet for the quadrate ;
JS %, additional facet for the jugal process of the quadrate.

Fig. 6. Lateral view of the right jaw.

Fig. 7. Ventral view of sternum. Sp.e., spina externa sterni; p.l.a., anterior lateral
process; S, lateral line of m. subclavius.

Fig. 8. Right lateral view of sternum. Sp.i., spina externa sterni; S, median line of
m. subclavius.

Figs. 9 and 10. Astur alphonsi (p. 285).

Fig. 9. Left tibia, front view. (f, rest of fibula; 6, bony bridge over the tendon of the
m. extensor digitorum ; p, peroneal crest.

Fig. 10. Left metacarpals, lateral view. p, articular facet of the pollex; m,, meta-
carpale ITT.

Figs. 11-18. Strix sauzieri (p. 286).
Figs. 11, 12. Inner and outer views of humerus. f, pneumatic foramen.
Fig. 13. Left tibia, front view. p, peroneal crest; f, distal portion of fibula; /, attach-
ment of transverse ligament.

2x 2

300 SIR E. NEWTON AND DR. GADOW ON THE DODO

Figs. 14, 15. Posterior and anterior views of right tarso-metatarsus. 6, bony bridge
across the tendon of the m. tibialis anticus; h, facet for the hallux’s meta-
tarsal; 2, insertion of the m. tibialis anticus.

Fig. 16. Proximal end of the tarsus.

Figs. 17, 18. Posterior and anterior views of the small pair of tarso-metatarsals.

PLATE XXXIV.
Figs. 1-5. Plotus nanus (p. 288).

Figs. 1, 2. Dorsal and ventral views of pelvis. J, first vertebra fused with sacrum ;
T.z, primary sacral vertebra; pb, os pubis ; at, antitrochanter.

Fig. 3. Inner or median view of left humerus.

Fig. 4. Outer or lateral view of left humerus. C.s, crista superior ; S.¢, sulcus for the
humero-coracoid ligament; Z7.s, tuberculum superius s. externum; s.a./,
sulcus anconzi lateralis ; s.¢.m, sulcus anconei medialis ; 0.7, origin of the
m. brachialis inferior.

Fig. 5. Anterior view of left tibia. p, peroneal crest ; /.c, lateral cnemial crest.

Figs. 6-8. Butorides mauritianus (p. 289).
Figs. 6, 7. Dorsal and ventral views of left coracoid. J.2, linea intermuscularis ; p.J,
prominent lip; pr, precoracoid process; 7, rough ridge.
Fig. 8. Posterior plantar view of right tarso-metatarsus. h, attachment of meta-
tarsale I.

Figs. 9 and 10. Sarcidiornis mauritianus (p. 290).

Fig. 9. Lateral view of metacarpals of left wing.
Fig. 10. Median or ventral view. J/.z, rough knob of first metacarpal; P, articula-
tion of pollex ; f, facet for the os carpale ulnare.

Figs. 11-17. Anas theodori (p. 291).

ig. 11. Left lateral view of sternum. Sp.e, spina externa; J, lateral line of m.
subclavius.

Fig. 12. Anterior margin of sternum. 4K, keel.

Fig. 13. Ventral view of left coracoid.

|
3

Figs. 14, 15. Right humerus. - s, sulcus for ligament. humero-scapulare ; C.s, crista
superior ; 6.7, origin of m. brachialis inferior.

Fig. 16. Posterior or plantar view of right tarso-metatarsus.

Fig. 17. Proximal end of the same.

AND OTHER EXTINCT BIRDS OF MAURITIUS, 301

PLATE XXXV.
Figs. 1-11. Fulica newtoni (p. 292).

Dorsal view of premaxilla.

Dorsal view of third cervical vertebra.

Dorsal view of either ninth or tenth cervical vertebra.

Ventral view of either ninth or tenth cervical vertebra.

. Sternum from the right side. p.l.a, anterior lateral process; Sp.e, spina
externa ; s, ridge of m. subclavius.

Fig. 6. Sternum from the ventral side. aa, distance of 30 mm.

Fig. 7. Sternum from the dorsal side. C's, attachment of ligament between sternum

and coracoid.

Fig. 8. Lateral view of left humerus. S.¢, sulcus transversus.

Fig. 9. Median view of left humerus. Cs, crista superior ; ¢.m, tuberculum medium.

Figs. 10, 11. Anterior and posterior views of left femur. B, attachment of sling of

biceps muscle ; F, facet for the fibula.

i>]

_—
fe de de da
oT go bo Ht

Figs. 12-20. Aphanapteryx broecki (p. 293).

Fig. 12. Premaxilla and mandible; lateral view.
Fig. 13. Dorsal and ventral views of third cervical vertebra. p.t, posterior zygapo-
physis.
Figs. 14, 15, 16. Ventral, dorsal, and lateral views of sternum. 7, tubercle for attach-
ment of sterno-coracoid ligament ; A, anterior end of keel.
g. 17. Lateral view of left humerus. ¢.m, tuberculum medium, much higher than
in Fulica.
Fig. 18. Median view of left humerus.
Fig. 19. Ventral view of pelvis and sacrum. p.p, pectineal process; Sz, primary sacral
vertebra.
Fig. 20. Right lateral view of pelvis. A, antitrochanter ; p.i, posterior lateral dorsal
process of ilium ; P, os pubis.

Fi

e

PLATE XXXVI.
Didus ineptus (p. 296).

Fig. 1. The first correctly restored and properly mounted skeleton of the Dodo. The
specimen belongs to the Government Museum of Mauritius. The left wing
and ribs have not been drawn, in order to keep the drawing clearer. The
18th vertebra has been cross-shaded, because it was still unknown when the

skeleton was restored.

302 ON THE. DODO AND OTHER EXTINCT BIRDS OF MAURITIUS.

Figs. 2, 3, 4. Anterior, posterior, and lateral views of the atlas. Nat. size.
Fig. 5. Dorsal and ventral views of the metacarpals and of the first phalanx of the
index.

PLATE XXXVII.
Didus ineptus (p. 296).
Figs. 1 A, B. Dorsal and lateral views of the posterior portion of the pelvis. Nat. size.
Figs. 2A, B, ©. Lateral, anterior, and posterior views of the 18th vertebra. Nat. size.
Only one single specimen of this 18th vertebra was found amongst the
hundreds of other vertebree of the Dodo which have passed through our
hands. Thisspecimen is unique. The corresponding vertebra of the mounted
skeleton in the British Museum is a cleverly executed artificial substitute.

Yoo Vol MM TY

FIG1-8 LOPHOPSITTACUS MAURITIANUS. FI6.9-10 AS

FIG. 1-18 STRIX SAUZIERI

Cambridge Engraving Company.

SEMEN.

FIG. 1-5 PLOTUS NANUS. FIG 6-8 BUTORIDES MAURITIANUS. F16.9-10 SAI
FIG.1-17 ANAS THEODORI.

Of

C

Too x

POD sé

MAEL

&

ol, KU SURED

KH
qi

CA NEWTO!

Sl eNO

m
w)

Te

ene

ue

Tram Lo0k, be Vol MULL HU.

FIG. 1-5 DIDUS INEPTUS

Cambridge En

(Soc Viel MUTE OU

Cambridge Engraving Compony.

Q
ie)

Ie

—
Fy
pu)

DIDUS IN

8) ees AL
re alls
UNSEAT

[ 303 ]

X. Description of a remarkable new Sea-urchin of the Genus Cidaris from Mauritius.
By F. Jurrrey Beui, M.A., Sec.R.M.S., Professor of Comparative Anatomy and
Zoology in King’s College.

Received September 12th, 1892, read November Ist, 1892,
[Puate XXXVIII.]

A FEW months since the Trustees of the British Museum obtained from M. de
Robillard, of Mauritius, another of those rarities in the collection of which he has
so much distinguished himself !. The specimen is unique, and its general facies would
be so much altered by the removal of a fifth of the spinulation that I propose, on this
occasion, to limit myself to a description of the external appearance of the most
remarkable Cidaris it has ever been my good fortune to see. In the hope that further
examples might be discovered I have delayed, longer perhaps than I should, the publi-
cation of a notice of this extraordinary specimen.

The primary spines are exceedingly long, some of them being more than 150 millim.
in length, or about three times the diameter of the test. They are, however, most
remarkable for being curved, slightly indeed, but yet distinctly curved in an upward
direction. The base of the spine is flattened on its lower side; there are two sharp
edges, and the upper side is formed of two halves set at a wide angle to each other,
and ending in a distinct ridge. This ridge may be dentate and ornamented with a few
minute tubercles. At a distance of about 20 millim. from its base the upper ridge
disappears, and the spine becomes flat above as well as below. At about this point
most of the spines become completely altered in colour (in the dry specimen), for
while the basal part is creamy yellow, the rest of the large spine is of a reddish-
brown colour. In many, near the tip, there are a few bands of brown and pale yellow.
Where the brown colour begins a distinct striation also commences, and there are ten
striz on both the upper and the lower surfaces. The spine is widest at its base, and
as it narrows very regularly the whole has the form of a greatly elongated triangle.
Gradually and almost imperceptibly, the form of the spines in cross section alters,
and instead of being depressed and flattened it becomes almost regularly circular.

The spines just described are arranged very regularly in pairs in each inter-
ambulacrum ; only one or two are more than 150 millim. long; in each inter-
ambulacrum there are seven or eight primary spines, and the shortest are, as usual,

1 T cannot let pass this opportunity of putting on record the regret with which all who are interested in
marine zoology have heard of the recent death of this distinguished collector.

304 PROF. F. J. BELL ON A NEW SPECIES OF CIDARIS.

those nearest the mouth. It is usual for all to have the form and coloration already
described ; but those nearest the mouth are more spatulate in form than the rest.

The secondary spines are crowded in great quantities round the bases of the
primaries and in the ambulacral areas; they are sharply pointed, and creamy or
yellowish in colour. The apical area is very extensive and about half the diameter of
the corona. ¥

With regard to the affinities of this species it is not possible to say much. The
amount of ostracum seen in transverse sections of the spines is slight, and there are
no swellings, transverse crowns, or ridges, and no parasitic deposits. The long simple
spines with striz appear to be most like those members of the genus which have been
distinguished as Dorocidaris.

A number of, as I think, very unnecessary genera have been founded for various
examples of Cidaris!. To Cidaris in a wide sense there is no doubt that the present
specimen belongs, and I do not expect that the investigation of the denuded test will
lead to the establishment of any new generic division for it. From the characteristic
shape of the spines I propose to call it Cidaris cwrvatispinis.

EXPLANATION OF PLATE XXXVIII.

Cidaris curvatispinis, from a photograph, reduced to two-thirds of the natural size.

1 Of. Catal. Brit. Echinoderms Brit. Mus. (1892), p. 139.

“ARI £
SINTEASTIVAEIAS Lay GD

‘dunt qaequey

—_ emacs

TY 2b IE 726 45. GT PO

XI. On Remains of an Extinct Gigantic Tortoise from Madagascar
(Testudo grandidieri, Vaillant). By G. A. BouLENGER.

Received October 25th, 1892, read November 15th, 1892.

[Puates XX XIX.-XLI.]

On the 14th December, 1868, Prof. H. Milne-Edwards announced to the French
Academy of Sciences’ the discovery by M. Grandidier of remains of some gigantic
Tortoises in Madagascar, contemporaries of Apyornis and Hippopotamus lemerlii.
These bones were referred to two species, named Testudo abrupta and Emys gigantea
by Grandidier, without, however, any descriptions being given by means of which
some idea could be obtained of their affinities. So matters stood until 1885, when
Prof. L. Vaillant published some notes on these remains*, which had in the meantime
been restored for exhibition in the Paleontological Gallery of the Paris Museum, at
the same time showing that both species belong to the genus Testudo. The name
gigantea being twice preoccupied in that genus, Grandidier’s Emys gigantea was
renamed Testudo grandidiert. It is, however, not impossible that Grandidier’s tortoise
will ultimately have to be regarded as a form of the true Testudo gigantea of
Schweigger. To 7. grandidieri belong the remains the description of which has
been kindly entrusted to me by Dr. H. Woodward. ‘They consist of two nearly perfect
shells, skilfully restored by Mr. Barlow, fragments of others, an imperfect skull, and
numerous bones belonging to several individuals, found by Mr. Last in South-west
Madagascar, and now preserved in the Geological Department of the British Museum.

Mr. Last writes, from Nossi Vey, as follows about the specimens :—‘ They were
found in large caves in the rocks some two miles from the beach. These caves were
formed by the sea long ago, when either the sea was higher or the land lower than it is
at present. The Tortoises are found in pairs partly bedded in the fine loose sand of the
caves, and, owing to the fact that they are only partly buried, many of the small bones
get lost, the natives making use of these caves as hiding-places for themselves in time
of war, and for their goods in time of peace. In one case, where the shell of one
animal was completely underground, though broken all to pieces, I have sifted the
soil and found nearly all the small bones and a head, which seems to me very small
for so large an animal.”

Unfortunately, Mr. Last did not keep apart the bones which he secured in connexion
with the shell, but sent them home mixed up with those of several other individuals.
This has necessitated considerable labour on my part in identifying them, and in several

’ Comptes Rendus, Ixvii. 1868 p. 1165. 2 Comptes Rendus ec. 1885, p. 874.
VOL. XIL.—Pakt vill. No. 1.—April 1894. 2Y

506 MR. G. A. BOULENGER ON AN EXTINCT

cases caused some uncertainty; in dealing with the smaller carpal and tarsal bones
and phalanges I had to give up the attempt at sorting out the bones according to
specimens. The material, though thus inadequate for the complete restoration of any
single specimen, yet affords information on almost every portion of the skeleton, and is
therefore of the greatest value in fixing the characters of this extinct tortoise, our
knowledge of which was still very imperfect.

1. Shell.—In addition to fragmentary remains of several specimens, we have two
nearly perfect shells, referable to male and female. The former measures 116 centim.
in a straight line and 150 over the curve, thus agreeing very closely in size with the
type specimen in Paris, which measures 12] and 152. In the characteristic flatness of
the vertebral region, together with the sinuous protuberances and the deep grooves
separating the epidermal scutes, it agrees entirely with the original, as I am informed
by Prof. Vaillant, to whom I showed the specimens during a recent visit he made to
the Natural History Museum. The figures (Plate XX XIX.) appended to this paper
relieve me from giving a description of the shape of the carapace, which, after all,
differs but slightly from that of the existing Aldabra forms, Testudo elephantina and
allies. Its width is 85 centim. and its height 49. The anterior margin is feebly
notched in the middle and turned outwards on the sides, with notches between the
marginal scutes; the posterior sides are likewise expanded, and the pygal incurved.
The nuchal shield is very small, as long as broad; the vertebral shields are broader
than long, a little broader than the costals; the supracaudal is single.

The plastron agrees with that of 7. gigantea and T. hololissa. Its length is 92 centim.;
its front lobe 27 long and 33 broad at the base, 12 at the apex, which is truncate. The
bridge measures 44. ‘The hind lobe is rounded, without anal notch, 19 centim. long
and 47 broad. The gular shield is divided; the suture between the humerals three and
a half times as long as that between the gulars; pectorals very short; abdominals as
long as pectorals ; femorals three-fifths of pectorals; anals as long as gulars.

The second specimen, which is a female, differs from the preceding in its less deeply
excavated plastron, its greater thickness, the dorsal plates being about 1} inch thick,
whereas in the larger male their thickness is only # to | inch; the anterior and posterior
borders are not spread out nor notched; the sutures between the bones are almost
obsolete; and the nuchal is larger and a little longer than broad. The carapace
measures 97 centim. in a straight line, 120 over curve; width 73; height 44. The
plastron is imperfect, wanting the anterior lobe. ‘The bridge measures 40 centim.

In the characters of its shell, 7. grandidieri clearly pertains to the section known as
*¢ Aldabra Tortoises ””*. Aldabra being a group of small islands north-west of Madagascar,
this section of the genus would, with our present extended knowledge of its distribution,
more properly be termed the “ Madagascar Gigantic Tortoises.” On referring to the
synopsis of the species as given by me”*, it will be seen that an analysis of the characters

1 Giinther, ‘ Gigantic Land-Tortoises,’ p. 10. * Cat. Chelon. &e. p. 158.

GIGANTIC TORTOISE FROM MADAGASCAR. 307

of the carapace and plastron in the tortoise now described leads to 7’. hololissa, which
differs from 7’. elephantina in the smaller plastron not notched behind; from 7. gigantea
in the undivided supracaudal shield; and from 7. daudini in the wider plastral bridge.
T. grandidieri may be distinguished from T. elephantina, hololissa, avd gigantea by the
greater depression of the carapace.

2. Skull.—The remains on which Testudo grandidieri was established did not include
any portion of the skull. It is therefore highly gratifying to find among the bones
secured by Mr. Last a nearly complete skull, 15 centim. long (to the extremity of the
supraoccipital crest), wanting merely the zygomatic arch and the right quadrate. This
skull was associated with the female shell noticed above. The mandible of this
specimen is also present, together with a symphysial fragment of another.

The structure of the skull fully confirms the conclusion arrived at by Prof. Vaillant,
from the study of the shell, as to the close affinity of 7. grandidieri to the Aldabra
Tortoises. The differences, in fact, are rather slight and, in some respects, show an
exaggeration of the features which differentiate the Aldabra forms from their congeners.

The naso-frontal region is moderately convex, with the nasal fossa extremely large
and produced to between the anterior portion of the orbits, sloping obliquely down-
wards, and longer than broad. When the skull is viewed from above, the anterior
portions of the choane, separated by a narrow septum, are visible through the nasal
fossa, whilst the premaxillaries terminate on a line with the anterior borders of the
orbits. The interorbital region is formed entirely by the prefrontal bones, the upper
surface of which is much more developed than that of the frontals. The postorbital
arch is slender and the parietal bones narrow. ‘The pretrontals form a broad suture
with the postfrontals, and the frontals are enclosed between these two elements and the
parietals. The prefrontals are longer than broad, and their median suture measures
two-fifths that between the frontals; the latter are very slightly longer than broad, and
measure half the greatest length of the parietals. The diameter of the tympanic cavity
equals that of the orbit.

The lower aspect of the skull presents this peculiarity, that the pterygoids do not
meet on the middle line, being separated by the basisphenoid, as I have recently
described in 7. microtympanum*. ‘The vomer is produced posteriorly far beyond
the line of the postorbital arch, and its length equals nearly four times that of the
choane; it bears a feeble median keel, which is continued on the anterior half of
the basisphenoid; the palatines extend but little beyond the vomer. The suture
between the premaxillaries and the vomer falls a considerable distance behind the
inner angle of the alveolar edges of the maxillaries. The alveolar surface of the
maxillary is broad, with a strong denticulate median ridge, which is equally distant
from the likewise denticulate inner and outer margins. ‘The occipital condyle is
tripartite, and the posterior margin of the opisthotic is not excised.

? Proc. Zool. Soc. 1891, p. 5, figs. 1, 2, 3.

305 MR. G. A. BOULENGER ON AN EXTINCT

The mandible has a double denticulate ridge, between which the alveolar surface is
deeply concave; its outer surface slopes outwards and is concave; the symphysis is
short and without a backward dilatation of the lower margin.

3. Cervical Vertebre.—Among the remains mentioned by Vaillant was a third
biconvex cervical vertebra, thus affording a valuable diagnostic character confirming
the deductions arrived at from the examination of other bones. Our material is,
unfortunately, very scanty, consisting merely of two second vertebra, one sixth, one
seventh, and three eighth, one of the latter being reduced to the arch. I have there-
fore no special remarks to offer on this part of the skeleton.

4. Sacral and Caudal Vertebre.—A great number of vertebre were collected, which
I have been able to sort out and refer to three specimens, as follows :—

A. The largest specimen: 7 vertebra, referable to the 5rd, 4th, 6th, 10th, 11th, 12th,
and 13th caudals. In all these the arch is anchylosed to the centrum, and so are the
costoids to the vertebre. I regard this specimen as a male.

B. A smaller specimen, with centrum and arch and costoids likewise anchylosed, but
the arch less elevated and less produced posteriorly; the series consists of a sacral
and 11 caudal vertebre, viz. Ist and 2nd, 4th and 5th, and 7th to 13th. I regard this
specimen as a female, probably the same as yielded the shell and skull which were
found associated.

C. This specimen, which agrees nearly is size with the preceding, differs in having
the centra thicker and joined by suture to the arches, which in some of the vertebre
have even become detached; the costoids were also loosely attached and have been
lost. These vertebree form a complete series from the 2nd to the 18th.

The bones in specimens A and B agree so well with those of 7. elephantina, male
and female respectively, that I have no doubt they belong to the same form as yielded
the skull described above. As regards specimen C, there are, on the other hand,
several differences, so that I have to consider whether they are referable to the same
species. These differences consist chiefly in the greater vertical diameter of the
centrum, the lesser excavation of the articular cavity, the persistent suture between
arch and centrum in the anterior vertebra, 8th inclusive, and the autogenous costoids
throughout.

Owing to the fact that the vertebra are nearly equal in size in specimens B and C it
would seem, at first, that age cannot be made to account for the differences. However,
we must bear in mind that the tail differs so much in size according to the sexes in
these Tortoises that it may be as large in a half-grown male as in a full-grown female
of the same species; and how great the differences in the shape of the bones are may
be gathered from a perusal of Giinther’s descriptions on pp. 29 and 37 of his memoir,
which deal with what I regard as male and female of one and the same species. I have
therefore carefully compared the caudal vertebre of the adult male 7. elephantina with
those of the female of the same species (I regret to have no half-grown or young male

GIGANTIC TORTOISE FROM MADAGASCAR. 309

skeleton with which to compare), and have come to the conclusion that the differences
between specimens B and C may, provisionally at least, be ascribed to both age and sex.
On comparing the caudal vertebrae of a young Galapagos tortoise (7. elephantopus) with
those of an adult of a closely allied form (7. vicina) I find the articular facets to be
nearly plane in the former, whereas the cup-and-ball system is strongly developed in
the latter. As to the anchylosis or non-anchylosis of the costoids, I think the
examination of more material would reveal a great amount of individual differences
on this point, irrespective of age or sex. Both specimens of 7. elephantina show no
trace of suture, but I find a great amount of individual variation in other species. In
the adult specimen of 7. vicina most of the costoids show a distinct suture with the
centrum, whilst those of the 6th vertebra, the right one of the 10th, the left of the
16th and 17th, and the right of the 18th and 19th are more or less completely united
with the centrum.

After describing the tail of the adult male 7. elephantina, Dr. Giinther adds :—
“Nearly always the animal carries it [the tail] bent sidewards under the carapace,
generally towards the left side; and therefore I anticipated to find a want of symmetry
in some portion of the root of the tail; however, nothing of the kind can be observed.”
This statement is not quite correct. Since the above lines were written, Dr. G. Smets!
has pointed out that the basal caudal vertebre of 7. sulcata and other Land-Tortoises
are characterized by a remarkable asymmetry, especially with regard to the zygapo-
physes, and I find his statement borne out by the gigantic species as well. Smets
remarks of 7. sulcata that the right postzygapophysis of the first sacral vertebra is
obliterated, whilst the left is well developed; on the second sacral the left zygapophysis
is more developed than the right; likewise on the first caudal. On the second caudal
the right postzygapophysis is slightly more massive than the left, but its articular
facet is smaller; in the 3rd, 4th, and 5th the right postzygapophysis is more developed
than the left, whilst in the 6th to 8th it is the reverse. ‘The first and second caudal
vertebree have the diapophyses more developed than the left; on the third vertebra the
left diapophysis is less massive but a little longer than the right; fourth and fifth with
the left, seventh with the right, diapophysis longest. From the ninth vertebra any
striking asymmetry ceases.

In the large male specimen described by Giinther’ the second sacral vertebra has
but one prezygapophysis, the left; the facet of the right postzygapophysis of the first
caudal is much larger than that of the left, but little larger in the second; in the third
vertebra the left postzygapophysis is the largest, and in the fourth it is the right. In
the female specimen the asymmetry is much less marked.

In the 3rd and 4th vertebra of specimen A of 7. grandidieri the right postzygapo-
physis is more developed than the left, in the 6th the left.

In specimen B the left postzygapophysial facet of the second sacral is much higher

1 Muséon, 1887, p. 394. 2 Op, cit. p. 29.

310 MR. G. A. BOULENGER ON AN EXTINCT

up than the right; the right postzygapophysis of the first caudal vertebra is abortive,
its facet being sessile and directed upwards; that of the second vertebra is less
developed than the left; the other vertebre being nearly symmetrical.

In specimen C the right postzygapophysis is the more developed on the second
caudal vertebra, the left on the third and fourth.

It is therefore clear that in these Tortoises the asymmetry of the caudal vertebre is
subject to much individual variation, and that the identification of single vertebra
cannot be attempted by means of this character.

5. Pectoral arch.—The complete pectoral arch of a specimen which, judging from
the size of the bones, must be the one of which Mr. Last states he found most of the
bones, that is to say, the female previously noticed, is preserved. The coracoid is free
from the scapula, which measures (from the proximal extremity of the suture with the
coracoid) 174 centim., the acromial process (so-called precoracoid) measuring (likewise
from the suture with the coracoid) 9 centim. Length of coracoid 11 centim., greatest
width 9. These proportions agree very closely with those of 7. elephantina.

Besides these bones, a smaller right coracoid is preserved.

6. Pelvis.—This is represented by the ischia of a large specimen, and the ilia and left
pubis and ischium of a smaller specimen, no doubt the female of which shell and bones
were found associated. In general configuration, and especially in the narrow bridge,
they agree with 7’. elephantina. The surface of the ilium which articulates with the
sacrum is directed more upwards than inwards, thus differing from the specimen of
T. elephantina with which I have compared it, and approaching the arrangement
described by Vaillant.

7. Limb-bones—Numerous carpal and tarsal bones and phalangeals and long bones
belonging to at least six specimens :—

A. The largest (femur measuring 20 centim.): left femur, right tibia, and left fibula,

B. A smaller specimen, probably the female (femur 165 centim.): right humerus,
right and left radius, and right ulna; right and left femur and right tibia and fibula.

C-F. Right radius; two right tibie ; four left tibie ; one right and one left fibula.

There are also a few bones, apparently dermal ossifications from the limbs, which I
am unable to determine with precision, having been unsuccessful in finding anything
similar with which to compare them.

Now that most of the bones have been identified, it will perhaps be possible to mount
the skeleton of the female specimen found undisturbed by Mr. Last, and to exhibit it
in that condition in the Geological Galleries of the British Museum. The neck will,
however, have to be omitted.

GIGANTIC TORTOISE FROM MADAGASCAR. dll

EXPLANATION OF THE PLATES.

PLATE XXXIX.

Shell of male (Zestudo grandidieri), $ nat. size. Upper, lower, and side views.

PLATE XL.

Shell of female, $ nat. size. Upper, lower, and side views.

PLATE XLI.

Skull of female, nat. size. Upper, lower, and side views, and upper view of
symphysial portion of mandible.

E.C.& G.M-Woodward. del.et lith.

“Wes t, Newman. imp.

THESTUDO GRANDIDIERI.

E.C & GM Woodward del.et kth. West, Newman imzp .

THESTUDO GRANDIDIERI.

My
teh

S

cen

EC &GMWoodward del.et lth.

THSTUDO GRANDIDIERI.

a

XII. On the Remains of some Gigantic Land-Tortoises, and of an extinct Lizard,
recently discovered in Mauritius. By Hans Gapvow, Ph.D. WA. FRS.,
Lecturer on Advanced Morphology of Vertebrata, and Strickland Curator,
University of Cambridge.

Received November 29th, 1892, read December 20th, 1892.

[Puares XLII.—XLIV. |

THE collection of bones of birds from the Mare aux Songes, in Mauritius (described
in Trans. Zool. Soc. vol. xiii. (1893) p. 281), was accompanied by an equally interesting
collection of reptilian remains from the same locality. Mr. Théodore Sauzier, as President
of the Commission des Souvenirs Historiques, made the stipulation that a typical
selection of these bones should be given to the University Museum of Zoology, pro-
vided the whole of the material was worked out in Cambridge. This task has been
entrusted to me, and I now take the opportunity of thanking Mr. Sauzier for his
generous liberality.

Considering that a great portion of the collection has to be returned to Mauritius, it
is necessary to figure most of the important specimens.

Dr. Giinther’s Monograph, ‘ The Gigantic Land-Tortoises, living and extinct,’
London, 1877, naturally forms the basis of the following descriptions. By having dis-
tinguished several species, notably Testudo triserrata and T. inepta, when studying
previous collections from the Mare aux Songes, and by having, moreover, assigned
names to the numerous disconnected skulls, carapaces, plastra, pelves, and shoulder-
girdles, he has established a case of precedence which naturally has to be acknowledged
as potentially correct, until at some future time complete specimens, with all their
bones associated, shall be found, and either corroborate or correct his identifi-
cations.

I follow his plan of distinguishing by different names at least some of the most obviously
differing carapaces and plastra, referring, however, to many of the other bones by
letters and numbers only. The same letters and numbers, with references to this
paper, having been attached to all the specimens, recognition has been secured.

There remains the question of the specific value of these names. It is immaterial to
the descriptive purpose of this paper, whether they be considered as indicating species,
subspecies, varieties, or races. So long as we knew that Mauritius was inhabited
at the utmost by three species, namely 7. triserrata and T. inepta—T. indica s.
perraulti being only supposed to have come from Mauritius, and since the name
T. leptocnemis was suggested only on account of the femur, pelvis, and scapula—this

vou. Xt.—Part vill. No. 2.—April, 1894. 22

314 DR. HANS GADOW ON REMAINS OF SOME GIGANTIC

view was to be accepted as possible, considering that the island of Aldabra has yielded,
according to Dr. Giinther, five living species, which Mr. Boulenger has reduced to four.

But now, through this last collection, there have come to light so many different
forms of Tortoises that, proceeding upon the old lines, at least the following forms
have to be distinguished :—

T. INDICA, provided carapace No. V. belongs to this form.
T. TRISERRATA.

T. INEPTA.

T. SAUZIERI.

T. SUMEIREI, 7. é. the name given by Mr. Sauzier to the specimen which is still living
in the court of the Artillery Barracks at Port Louis.

Lastly, 'T. Leprocnemis, if need be.

This makes five or six different forms, and to suppose that these represent as many
species reduces the idea of a species to absurdity, unless the very presence of these
Tortoises on this little island (and the same applies to Aldabra) be explained by the
thrilling assumption that during the supposed process of subsidence of the surrounding
country—now the Indian Ocean—the Tortoises fled to the highest districts, now the
islands of Madagascar, Aldabra, Mascarenes, &c. This assumption implies the
supposition, equally gratuitous, that South-western “ Lemuria” was inhabited by at
least 11 to 14 different species of gigantic Tortoises, namely 5 or 6 now in Mauritius,
4 or 5 now in Aldabra, 1 in Rodriguez, and 1 or 2 in Madagascar, not counting
the species which possibly never reached these islands.

How these islands ever received their Tortoises is a mystery, but this is quite another
question.

The five or six forms of Galapagos Tortoises were, or are, to a certain extent
peculiar to different islands, and this isolation is in favour of their specific value, but
five of the Mauritian forms were all found in the same swamp !. With plenty of food,
a congenial climate, and without formidable enemies, they grew to a gigantic size, could
interbreed to their hearts’ content, for all we know to the contrary, and variation
within harmless bounds received no check from natural selection. ‘The very thinness
of the shells of some of these gigantic Tortoises, especially 7. vosmaert of Rodriguez
and several Galapagos forms, seems to indicate that strength of the dermal armour was
no longer required in these Elysia of Tortoises.

1 I do not suggest that different genera, and even different species of one genus, do not inhabit the same
locality. In the marismas of Andalucia I have found, in the Laguna de los Patos, Hnys europea and
Clemmys sigris s. leprosa in equal numbers, a somewhat unexpected fact, because Hmys is the almost exclusive
Tortoise in North Portugal, while Clemmys is extremely abundant in the Alemtejo, where Lmys is very rare.

LAND-TORTOISES AND AN EXTINCT LIZARD FROM MAURITIUS. 315

Tue Carapaces. (Plate XLII.)

Carapace No. I.—Testudo sauzieri. 'The whole shell, together with the plastron, is
complete, with the exception of the second, third, and fourth vertebral plates. The
dorsal profile differs from that of the typical 7’. inepta by the shape of the hump of
the fifth vertebral plate, the much steeper hump of the fourth plate, and the appa-
rently much steeper hump of the first vertebral.

The marginals are likewise different; the first is in broad contact with the first
costal, even more so than in 7. triserrata, while in T. inepta the first marginal and
first costal do not touch each other. The last marginal or caudal is much thicker than
in T. imepta, measuring 10 by 3°35 cm., with a thickness of 2°3 cm.; instead of being
concave ventrally, it is decidedly convex.

The total length of carapace no. I. is 51°5 cm.; its greatest breadth across the
inguinal region is 36 cm.

The plastron, 36 cm. long, resembles that of 7. triserrata (that of T. inepta is still
unknown), but the pectorals are wider than in the plastron figured by Dr. Giinther,
while the markings of the shields in the axillary region agree with it. However, there
is another male plastron in the Cambridge Museum, determined by Dr. Giinther as
belonging to 7. triserrata, in which the pectorals are just as wide as in 7’, sauziert,
while the axillary impressions are different.

The whole shell is rather thick, like that of 7. inepta; the sides are steep and as
decidedly convex as in 7. inepta.

Carapace No. II.—7. inepta. Fragment of posterior two-thirds, typical 7’. inepta.

Carapace No. III.—7. triserrata. Fragment, consisting of the posterior six
marginals, with portions of the adjoining right and left fourth costals, and the fifth
vertebral plates.

Carapace No. 1V.—Z. triserrata. Fragment, a little more complete than no. III.

-Both specimens are easily recognized as belonging to the typical 7. triserrata,
because of the large, ventrally concave caudal plate, which is thin, strongly curved, and
measures 17°5 cm. in greatest width, 10 cm. in height.

Carapace No. V.—TZ. indica. This fragment consists of the complete first, second,
third, and portion of the fourth vertebral plates; portions of the right and left first
marginals, portions of the right and left first, second, and third costals. Greatest
length of fragment 43 cm.

Its dorsal longitudinal profile is almost a straight line, only with a slight concavity
across the middle of the first vertebral plate. All the vertebrals are nearly flat, and
there is no indication of a swelling or hump on the fourth plate.

The first marginals are likewise in the same flat dorsal level. Apparently the first
marginal scutes did not touch the costal plates.

The anterior margin of this carapace is very peculiar (cf. Pl. XLU. fig. 10).
222

316 DR. HANS GADOW ON REMAINS OF SOME GIGANTIC

Ist. The two marginals form a straight line, instead of being curved as in T. inepta,
T. triserrata, and T', sauzieri.

2nd. The median notch is very slight dorsally ; absolutely wanting on the ventral
margin.

3rd. Ventrally the two marginals are strongly concave, forming a sharp and very
prominent ridge.

Dr. Giinther, Monograph, p. 43, remarks that “a carapace with so straight a verte-
bral profile as that delineated and described of 7. perraulti is not represented among
the specimens collected by Messrs. Bouton and Newton.” Moreover, none of the
species described in his Monograph possess such a flat carapace. Our carapace
no. V., in its flatness and almost straight profile, agrees rather well with the figure
given by Perrault of his male Grande Tortue des Indes (‘Mémoires pour servir &
Histoire des Animaux et des Plantes,’ Amsterdam, 1736, p. 395), but the anterior
marginals are very different. ‘This may, however, be due to the inexactness of the
drawing, which also exhibits the curious anomaly of showing only four instead of the
usual five vertebral plates.

Perrault remarks that the length of the shell was 8 feet, the tail 14 inches long, and
ending in a point “ garni d’un bout semblable & une corne de beeuf.” The length of
this horny spur is not mentioned; judging from the figure, it would scarcely amount to
half an inch.

Duméril et Bibron, ‘ Erpétologie Générale, vol. ii. p. 126, mention among other
points “la suscaudale simple, trés élargie ; la derniére de la rangée vertébrale bombée.”
If this implies that only the last vertebral plate possesses a hump, then this specimen
differs from both 7. triserrata and T. inepta, because in the former all the vertebrals
are humped and in the latter the fourth and fifth; on the other hand, 7’. indica s.
perraulti agrees by the large caudal plate with Z. triserrata, and differs from 7. inepta
and 7’. sauzieri.

It is very probable that carapace no. V. belongs to a 7’. indica, and in this case there
can be no longer any doubt that Perrault’s specimen came from Mauritius, a corrobo-
ration of Dr. Giinther’s surmise.

THE Puastra. (Plate XLIL.)

Plastron A.—Intermediate between 7. triserrata and T. sauzieri. This plastron is
complete. Its greatest length is 39 cm., its greatest breadth 35:5 cm., indicating a
much broader Tortoise than 7. sauzieri. It agrees in its ventral impressions with that
of T. sauziert, but differs from the latter as follows :——

1. The markings or shield-impressions in the inguinal region are more like those of
T. triserrata.

4. The posterior margin of the plastron is decidedly and sharply curved upwards,
instead of showing a slight triangular swelling; more like 7. ¢riserrata.

LAND-TORTOISES AND AN EXTINCT LIZARD FROM MAURITIUS. 317

3. The fourth to seventh marginals, which connected the plastron with the three
middle costal plates, are very much steeper, and almost flat vertically instead of being
convex.

Plastron B. T. triserrata. Typical. Represented by the two disconnected anterior
and posterior two-fifths of a male specimen.

Plastron C. T. triserrata. The anterior half of a large male plastron; greatest
width of fragment 42 cm.

Plastra D, FE, F, G. T. sumeirei. When Dr. Giinther wrote his Monograph he
could state categorically :—

1. That the specimens with a nuchal plate, and with a double gular, came from
Aldabra.

2. That the specimens without nuchal, and with a single gular, came from the
Mascarenes.

3. That the specimens without nuchal, and with a double gular, are Galapagos
Tortoises.

Now this statement cannot be upheld any longer, because among the materials
brought by Mr. Sauzier from the Mare aux Songes are the anterior portions of four
very large plastra, which differ from all the others previously received from Mauritius
and Rodriguez in the following points !:—

1. The anterior lobe of the plastron is very much elongated.

2. It ends in a fork instead of being rounded off.

3. There were two gular shields, a right and a left, as indicated by the deep
impressions left upon the bones.

Another difference is exhibited by the posterior portion of the plastron (Pl. XLII.
fig. 8), which, from its size, thickness, and colour, I suppose to belong to the same
Tortoise as the anterior portion of the plastron (fig. 6).

The posterior margin of this specimen ends ventrally in a much swollen and rugose
tuberosity ; dorsally it possesses a somewhat triangular, very strong tuberosity, which
seems to have fitted upon the ischiadic symphysial tuberosity of the pelvis, and which,
to judge from its roughness, seems partly to have been anchylosed with the pelvis.
None of the Mauritian specimens, hitherto known, show any such tuberosities; but
they exist in some of the Aldabran forms, namely in 7. elephantina, T. daudini, and
T. hololissa, not, however, in T. ponderosa, which latter has, by the way, been recognized
by Boulenger as the female of 7. elephantina.

In the configuration of the pectoral impressions, and in the whole shape of the
anterior lobe, the plastra D, E, F, G agree mostly with 7. daudini.

1 There are also five precisely similar specimens of anterior plastral portions in the Cambridge Museum,
which had probably been received together with those Tortoise-remains from Mauritius which Professor

Haddon has catalogued and described in Trans. Linn. Soc. ser. ii., Zoology, vol. ii, (1879) pp. 155-163, pl. 13,
They have, however, remained undetermined and do not seem to have been mentioned.

318 DR. HANS GADOW ON REMAINS OF SOME GIGANTIC

All these specimens are extremely thick and heavy, in every respect different from
the plastra of 7. triserrata and 7. sauzieri. The greatest width of fragment Kis 38 cm.

The fact that they have been found in the Mare aux Songes, together with the other
Tortoise material, excludes the possibility of their having been introduced by Man.

Tabular Comparison of the Forked Plastra D, BE, F, G of Mauritius with Plastra of
Aldabra Tortoises. (-++ means agreement, — means difference.)

L. elephantina. | 1’. ponderosa. | T. daudini. | T. hololissa.

Anterior end of plastron .... a | +, most. ~ ee
Double gulars ............ + ae ak an
Pectoral impressions ........ + +, most. _— —
Posterior plastral tuberosity .. + — te ai

T. sumeiret (P1. XLIV.).--The indigenous existence in Mauritius of Tortoises with
a double gular and with a long forked plastron having been proved, it is quite within
the range of probability that the solitary specimen which is still living in that island
is a native and not an imported creature.

Mr. Sauzier has given a description, with photographic views, of this specimen in
‘La Nature,’ no. 1016, 19 novembre, 1892, pp. 395-398, and he has distinguished it
as Testudo swmeire?, in honour of M. Camille Sumeire, of Mauritius. Mr. Sauzier has
presented to us several of the original photographs of this Tortoise, accompanied by
the following notes :—

“ Lors de la conquéte de l’Ile de France (Maurice), le 3 décembre 1810, il existait
dans la cour des casernes de |’ Artillerie, 4 Port Louis, une gigantesque tortue de terre,
qui a fait partie du matériel laissé aux Anglais.

‘“‘ Cette béte vit encore dans cette méme cour, dont les batiments ont été convertis
en mess pour les officiers.

“Tl est facile de voir, par son aspect général, qu’elle doit étre d’un grand age. Si, en
1810, d’aprés les plus anciens habitants, elle avait attenue sa taille actuelle, ou peu
aprés, elle aurait pour moins deux siécles--ce qui ne l’empéche pas, bien qu’aveugle
depuis quelques années, de porter avec aisance sur sa carapace deux hommes repré-
sentants ensemble le poids de 150 kilos.

“Tl est A regretter que l’absence d’échelle [in the side-view photograph ], ou mieux
encore, d'un objet de comparaison, dans la photographie, ne permette pas d’apprécier
exactement la taille de cette gigantesque tortue, dont on ne connait pas le lieu
dorigine.”

When walking this Tortoise stands 63°5 cm. high, leaving 15°5 cm. between the
ground and the plastron; its carapace is grey and measures in its “ grande circonfé-
rence” 259 cm. = 8 feet 6 inches, and 213 cm. “ de circonférence en largeur.” The

LAND-TORTOISES AND AN EXTINCT LIZARD FROM MAURITIUS. 319

fore legs are 45, the hind legs 30 cm. long; neck and head 39-5 em., tail 30°5 cm. in
length.

The back view shows a very large, broad, and sharply-curved caudal shield, which
strongly resembles that of 7. triserrata, and differs from that of any Aldabran or
Galapagos specimens. Front and side views show that there is no nuchal shield
whatever.

The under view shows a slightly forked projection of the anterior end of the plastron,
with two gular shields, indicating two gular bony plates as in the forked plastra
D, E, F, G.

The first marginal shield is very large and in broad contact with the first costal,
agreeing in this respect with Aldabran specimens.

The profile of the carapace, the scarcely serrated marginals, and the markings of the
vertebral and costal shields most resemble the corresponding parts of 7’. ponderosa =
female of 2. elephantina.

The anterior portion of the plastron, which is well shown in the photograph, agrees
in length, narrowness, and forked termination with 7. daudini.

According to Boulenger’s Key, pp. 153-154, Cat. Chelonians, British Museum, this
specimen would come nearest to 7. nigrita and T. nigra s. elephantopus (nuchal absent,
gulars distinct, shields of carapace concentrically striated in the adult, profile of cara-
pace declivous in front). But 7. nigrita differs considerably in the shape of the
anterior end of the plastron, the profile of the distinctly humped vertebral shields, and
the serrated marginals. 7. elephantopus differs likewise in the shape of its plastron.
Both 7. elephantopus and T. nigrita, moreover, differ in the shape of their much
smaller caudal shield, and above all in the shape of the head. The head of the
photographed specimen, 7. swmetrei, agrees much more with that of the Aldabran
type, while 7. ¢riserrata and T. inepta appear to have resembled the Galapagos
types.

There remains the question whether the forked plastra D, E, F, G belong to the
same race of Tortoises as 7. swmeiret.

This question is difficult to settle; we do not know the carapaces which belonged to
the plastra D, KE, F, G.

However, this much is certain—(1) that the plastra D, E, F, G cannot have
belonged to Galapagos Tortoises, because of the double gulars and because of their
locality ; (2) that the type of 7. swmeire? cannot be a specimen introduced from the
Galapagos Islands, because of the shape of its head, plastron, and double gulars;
(3) that 7. swmeiret cannot be one of the true Aldabran species, because it has no
nuchal shield and because of its different caudal shield; (4) 7. swmeirei exhibits
quite a new combination of characters, namely double gulars, without nuchal, and
is indigenous in Mauritius. At any rate, we have here a Mauritian Tortoise which is
fundamentally of the Aldabran type, but combines with Aldabran features several

320 DR. HANS GADOW ON REMAINS OF SOME GIGANTIC

peculiarities which are characteristic of the Mauritian 7. indica, T. triserrata, and
T. inepta, and also resembling in several points some of the Galapagos species '.

Complexes of terminal Caudal Vertebre (Plate XLIII. figs. 1, 2, 3)—Until proof
to the contrary is forthcoming, I assign two completely preserved specimens of
anchylosed terminal vertebra to the species which possesses the cleft or forked
plastron, namely 7. sumeiret. The largest of these curious specimens measures
12 cm. in length, with an anterior concave, almost saddle-shaped, articulating facet of
53 cm. in width. The anterior half of this vertebral complex consists clearly of three
or four anchylosed vertebre, while the posterior half, strongly curved downwards and
tapering to a blunt point, shows by its surface-mouldings that it was covered with a
horny sheath which completely surrounded the terminal half like a spur. The length
of this spur was at least 6 cm., to which, of course, the probably considerable thickness
of the horn itself has to be added.

Dr. Giinther says in his description of 7. elephantina (Monograph, p. 30) that “the
last seven vertebree are quite rudimentary and coalesced into a single bone.” ‘The
total number of caudal vertebre of 7. elephantina is 25, 7. e. 18 free vertebre besides
the coalesced complex; the shell of the large stuffed male specimen in the Natural
History Museum is not less than 49 inches long, but the caudal complex is far less
completely anchylosed, and its anterior articulating facet is one third smaller than is
the case with the two specimens in Mr. Sauzier’s collection. They either belonged to
a Tortoise of gigantic dimensions (as indicated by the large plastron E), or the caudal
spurred complex is relatively larger than in any of the Aldabran races. According to
Perrault’s description, 7’. indica likewise possessed a distinct horny spur. Dr. Giinther
continues as follows :—“ In individuals of the male sex the tail plays a very important
part as an external prehensile or, rather, steadying organ, which also differs externally
from that of the female in its greater length and by being provided with a large
terminal claw. Nearly always the animal carries it bent sidewards under the carapace,
generally towards the left side, and therefore I anticipated to find a want of symmetry
in some portion of the root of the tail; however, nothing of the kind can be observed.”

I have much pleasure in corroborating the sagacious anticipation of Dr. Gtnther
concerning an asymmetrical development, not, however, of the root of the tail, but of
the terminal half of the anchylosed complex, which shows a distinct deviation towards
the left side (see figure of dorsal view, Plate XLIII.).

Among gigantic Land-Tortoises such an archylosis has hitherto been observed only
in specimens from Aldabra, Mauritius, and, to a lesser extent, from the Galapagos.
In one specimen of 7. elephantopus the vertebre are, according to Giinther, ‘irregular,
and asymmetrically confluent towards the end of the tail.”

? Should the post-mortem of the solitary surviving type of 7’. swmetrei reveal that it differs in its plastron
from those referred to as D, E, F, G, I herewith reserve to myself the claim of distinguishing these forked
plastra as belonging to a Testudo quenthert.

LAND-TORTOISES AND AN EXTINCT LIZARD FROM MAURITIUS. 321

Adult specimens of Chelone midas have a similar, although much smaller, caudal
claw; in a large male specimen in the Cambridge Museum the horny claw or spur is
about 2 cm. in length, and covers about three vertebra, two anchylosed and one free.

None of the textbooks of zoology, comparative anatomy, and herpetology written
since 1877 have as yet condescended to mention this important instance of an anchy-
losed terminal caudal vertebral complex in the class of Reptiles.

Cervical Vertebre (Pl. XLIV. figs. 20-25).—Three atlas vertebra, representing two
different types, one with slightly joined neural arches and with a ring-shaped perforated
body, the others with a completely solid unperforated body—differences which cannot
be explained away by age. Although the atlas of the various families of Tortoises
exhibits many modifications, the specimen A, with the solid body, is peculiar; its
body does not contain the odontoid process, because the latter has left the three typical
articulating facets or impressions upon the body of the atlas. Specimens B and C
agree more with the atlas vertebre figured by Dr. Giinther.

Pelves.—Five nearly complete specimens, numbered I. to V. (Pl. XIII. figs. 4 & 5).

Pelvis No. I. has to be assigned to 7. triserrata according to Dr. Giinther’s defini-
tion. It is the largest known, measuring 23 cm. in height and 23 cm. in width.
The bridge between the obturator foramina is very broad, namely 3°8 cm. Unfor-
tunately, the tuberosity of the ischiadic symphysis, resting upon the plastron, is lost.
The ventral ridge of the ischiadic symphysis is very prominent.

Pelvis No. H. belongs to a smaller specimen, its measurements being 17 and
15-5 cm. The obturator foramina are wide, the bridge consequently narrow, resem-
bling that of Aldabran Tortoises. ‘The lateral ridge or crest of the shaft of the ilium
is very prominent, more so than in 7’ ¢riserrata. This pelvis differs markedly in two
peculiarities from those which have been determined as belonging to 7. triserrata and
T. inepta. First, the longitudinal ridge on the ventral side of the ischiadic symphysis
is very low, instead of being very prominent. Secondly, the ischiadic tuberosity has a
deep cavity on its ventral surface, and is rough instead of being smooth, suggesting
that it fitted upon and was partly fused with a corresponding tuberosity of the posterior
end of the plastron. Such a plastron is that which I have distinguished as belonging
to T. sumeiret.

Pelves III., IV., and V. resemble each other, and those of 7. énepta more than other
species; but it has to be noted that they exhibit a certain amount of variation in the
extent of the lateral iliac ridge—in fact, that they are intermediate between the typical
specimens of 7. triserrata and T. inepta. ‘They may belong to TZ. leptocnemis, the
pelvis of which is diagnosed as resembling that of 7. triserrata, but with a narrower
ilium.

Scapule and Coracoids.—The four specimens are all different. Two are more like
those of 7’. inepta; one more like that of 7. triserrata; the fourth has a very flat
scapular shaft, resembling in its transverse configuration that of 7. ¢triserrata and still

VOL. XII.—Part vil. No. 3.—April, 1894. 3A

322 DR. HANS GADOW ON REMAINS OF SOME GIGANTIC

more in general the Aldabran species. I feel inclined to associate this specimen with
the other remains of 7. swmeire?.

It has to be borne in mind, however, that the bones of the shoulder-girdle of all
these extinct Tortoises are subject to a very great amount of variation in size and
shape}. Itwould not be difficult to select out of the extensive material at our disposal
at least half a dozen different types, provided the intermediate forms were neglected
or suppressed.

Phalanges.—Three large terminal and seven middle and proximal phalanges.

Skulls (Pl. XLIIL figs. 6 a-8).—Mr. Sauzier’s collection contains 19 skulls and two
mandibles. Two of these skulls and one pair of underjaws belong to 7. triserrata
according to Dr. Giinther’s definition. Six skulls and the other pair of underjaws
agree with those of 7. inepta. The remaining eleven skulls differ from those of
L. triserrata and T. énepta chiefly in the shape of the ventral surface of the long supra-
occipital crest. This surface is broad, triangular, and concave, while it is narrow and
ridge-like in 7. inepta, narrow and doubly ridged (or, in other words, with a narrow
longitudinal groove) in 7. triserrata.

However, all these skulls exhibit a considerable amount of individual variation in
their general aspect, slope and size of the crest, relative strength of the various
parts of the skull, naso-frontal profile, &c. Most of them approach to a slight
extent the Aldabran skulls by the convexity of their frontal region, and, according to
Dr. Giinther, the “ posterior margin of the paroccipital crest is deeply excised” in
T. daudini. Whether these variations are due to age or sex, or are of specific or
subspecific value, cannot be determined. It has to be borne in mind that we have no
criterion whatever by which we can associate any of these numerous skulls with any
particular form of carapace, plastron, or limb-girdle. It is quite possible that the
typical skulls of 7. triserrata belong to the carapaces which have been distinguished
as those of 7. inepta, or to T. sumeirei, or vice versd, and the same remark applies with
equal force to the various sorts of pelves.

Unfortunately, this uncertainty is inevitable, because, owing to the circumstance that
many of the bones from the Mare aux Songes had to be fished out of a morass just as
the labourers happened to come across them, no record of the juxtaposition of the
various bones could be, or at any rate has been, made. Until, by happy chance or by
a much more careful and extensive mode of research, all the principal parts of one
unquestionable individual are found, the association of these bones will be a matter
of speculation without any valid basis.

“ Habt alle die Theile in der Hand,
Fehlt leider nur das geistige Band.”

* See A. C. Haddon, Trans. Linn. Soc. ser. ii., Zoology, vol. ii. (1879) pp. 156-158.

LAND-TORTOISES AND AN EXTINCT LIZARD FROM MAURITIUS. 323

DIDOSAURUS MAURITIANUS. (Plate XLIV. figs. 1-16.)

A short fragment, with three teeth, of the maxilla, five fragments of the mandible,
seven more or less perfect femora, and portions of three humeri have been described
and figured (with the exception of the maxillary fragment) by Dr. Giinther in the
‘Journal of the Linnean Society,’ Zoology, vol. xiii. (1878) pp. 322-324. All these
bones had been collected in the Mare aux Songes.

Mr. Sauzier has obtained many more specimens in the same locality :—4 complete
left mandibles ; 4 complete right mandibles; 10 right and 9 left mandibular dentals:
14 right and 14 left proximal halves of mandibles; 3 complete frontals, of two large
specimens and one small; 3 bases cranii; 1 atlas vertebra; 3 thoracic vertebrae, two of
which are successive and belong to one individual; 4 lumbal vertebre of a smaller spe-
cimen ; 1 lumbal rib; 1 sacrum; 2 fused vertebre; 4 post-sacral vertebree (first, second,
third ?, and fourth ?); 4 right humeri; 4 left humeri; 4 ulne; 3 right femora; 7 left
femora; and 5 left ossa innominata or pelvic halves, one of which indicates a small
specimen.

The largest of the complete underjaws measures 76 mm. in length.

As was to be expected, the number of teeth is variable, namely 22, 23, 24, 24, 25
in the five most complete left dentals, and 20, 22, 25, 26 in the right dentals.

The shape and proportions of the underjaws, of the frontal bone, and of the basis
cranii indicate that in the shape of the skull Didosaurus resembled the genus Cyclodus.

The largest humerus measures 42 mm. in length, the smallest 35 mm.; all the
specimens possess a distinct entepicondylar foramen.

The ulne vary from 50:5 to 32 mm. in length.

The largest femur measures 49 mm., the shortest 43 mm.

P.S.—Since this paper was read, Mr. Sauzier has published a memoir, which contains not only numerous
historical accounts, but also several excellent woodcuts, representing side and back views of the large Tortoise
living at St. Louis. It is entitled ‘Les Tortues de terre gigantesques des Mascareignes et de certaines autres
iles de la mer des Indes.’ 8vo. Paris, 1893. 32 pp.

EXPLANATION OF THE PLATES.
PLATE XLII.

Fig. 1. Type of Testudo sauzieri. Carapace No. I.

Fig. 2. Carapace No. I.; posterior view.

Fig. 5. Longitudinal vertical section through type-specimen of 7. sauzieri.
Fig. 4. 7. sauzieri; ventral view.

Fig. 5. Plastron A.

Fig. 6. Plastron F. Dorsal view, posterior portion, 7. swmeire?.

Fig. 7. Plastron F. Profile section through posterior portion.

Fig. 8. Plastron F. Ventral view, posterior portion, 7. swmeirei.

Fig. 9. Carapace No. V. Probably 7. indica of Perrault.

324 DR. GADOW ON REMAINS OF GIGANTIC LAND-TORTOISES.

Fig. 10. Profile section through the anterior portion of carapace no. V., in level of
the arrow.

Fig. 11. Profile section through corresponding part of 7. triserrata.

Fig. 12. Plastron E. Dorsal view. TZ. swmeiret.

Fig. 13. Plastron E. Ventral view. 7. swmeiret.

PLATE XLIII.

Figs. 1-3. Dorsal, anterior, and ventral views of complex of terminal caudal vertebre,
referred to Testudo sumeirei. Nat. size.

.4,. Pelvis No. II. Ventral view. Probably 7. sumeirei.

. 9. Pelvis No. II. Dorsal and anterior view.

. 6a. Cranium of Testudo, sp. ?

. 65. View of the “ posterior margin of the paroccipital crest.”

. 7. Cranium and posterior view of paroccipital crest of 7’. inepta.

ra Se % Ss a SS T. triserrata.

de’ da’ da gq’ de’ da

—

(Figs. 6, 7, 8 are drawn to the same scale, namely about 8, nat. size. The
paroccipital crests are drawn of the natural size.)

PLATE XLIV.

Figs. 1-16. Didosaurus mauritianus. Nat. size.
Figs. 1, 2. Inner and outer view of left mandible.

Fig. 5. Dorsal view of frontal bone.

Fig. 4. Ventral view of basis cranii.

Fig. 5. Upper figure: dorsal view of atlas.

Fig. 5. Lower figure: side view of a thoracic vertebra.

Figs. 6 & 7. Dorsal and ventral views of two successive thoracic vertebra.
Fig. 8. Dorsal and ventral views of sacrum.
Fig. 9. Dorsal view of first post-sacral vertebra.
Fig. 10. Dorsal view of second post-sacral vertebra.
Fig. 11. Dorsal view of third? or fourth ? post-sacral vertebra.
Fig. 12. Left outer view of pelvis.
Fig. 13. Posterior view of femur.
Figs. 14 & 15. Anterior and posterior view of humerus.
Fig. 16. Left ulna.
Fig. 17. Testudo sumeirei; the type specimen living at Port Louis, Mauritius.
Fig. 18. Ventral view of plastron of the same specimen.
Fig. 19. Side view of the same specimen.
(Figs. 17-19 after photographs procured through the kindness of Mr. Th. Sauzier.)
Figs. 20, 21, 22. Lateral, anterior, and posterior views of atlas A. Nat. size.
Figs. 23, 24, 25. Anterior, lateral, and posterior views of atlas B. Nat. size.

FrariLool Soc Yoo A PEXLI

MAURITIAN LAND=-TORTOISES.

Edwin Wils on Cambridge

Mc Vol MM PY ALM

Zool:

Spires

noe

Lemienetm

WARE

yeu
=
€

\a

(

MAURITIAN LAND-TORTOISES

hy

Edwin Wils on Caubri

Srans Lol Soe Ub AW PLLALLY

pan a

FIG.1-16 DIDOSAURUS MAURITIANUS
FIG.17-19 TESTUDO SUMEIREI
FIG. 20-25 sp ATLAS A anp B

Edwin Wilson Cambridge

XIII. A Revision of the Genera of the Alcyonaria Stolonifera, with a Description of
one new Genus and several new Species. By Sypnuy J. Hickson, M.A. Cantab.,
DSc. Lond., F.Z.S., Fellow of Downing College, Cambridge.

Received October 13th, 1892, read December 6th, 1892.
[Puares XLV.-L. |

IN a communication made to the Royal Society in 1883 (8) I pointed out the
advisability of separating those Alcyonarians in which the polypes spring independently
from a basal stolon into a special suborder, for which I preposed the term Stolonifera.
As my suggestions have not been very generally accepted by continental writers on
this group, it is necessary to preface my remarks by a short defence of the position
I now take in retaining this suborder.

No very serious argument has yet been brought forward against the retention of the
Stolonifera. Von Koch (19), in his monograph of the Gorgonide of Naples, refers
to my paper in a footnote only, and does not attempt either to describe or criticize my
classification. I cannot allow this opportunity to pass without reference to the extra-
ordinary and perfectly unjustifiable attack that he has made upon me in this footnote.

He says, in the first place, that I have quite falsely quoted his paper on the “ Skelet
der Alcyonarien.” ‘The sentence to which he probably objects will be found on
page 699. It runs as follows :—“ Recently von Koch has suggested a classification
that is based on the varieties of the skeleton, but it seems to me that the Pennatulide
and Gorgonidé are not so closely related as to justify their position in the same division
of the same group (Axifera).” The word Axifera, it is true, at the end of this sentence
was allowed to remain in this position in the text by an oversight, and for that I
apologize; but the general statement is perfectly true, for on p. 474, in describing his
third ‘‘ Hauptgruppe,” he says “Zu der letzten Abtheilung gehdren die beiden
Familien der Pennatuliden (VII.)...... und die der Axifera (VIII.)” (7. e. the true
Gorgonide).

In the second place, he says that I did not investigate a single true Gorgonid
(Axifera); but, as a matter of fact, I not only did make and examine several series of
sections through the polypes of Primnoa, which von Koch himself includes in his
family Gorgonide, but I actually gave and described a careful drawing of one of these
sections. Iwas induced to make the statement that the siphonoglyphe is probably
absent in the Gorgonide, partly because I could find no such structure in the two
forms I investigated myself, and partly because von Koch himself did not describe any

9

VOL. X11.—Part 1x. No. 1.—October 1st, 1894. 3B

326 DR. S. J. HICKSON—REVISION OF THE

such structure in any of the numerous Gorgonide he had examined (Isis, Gorgonia,
Sclerogorgia, &c.) ; but I was wrong in placing too much reliance on von Koch’s work,
as he failed to note in a monograph on the anatomy of the genus the well-marked
siphonoglyphe of Tub¢pora. Ever since the publication of my paper in the * Philo-
sophical Transactions’ von Koch has not brought forward one word of criticism upon
my system of classification that requires an answer, and I can only say now that I am
exceedingly sorry that he should have thought it necessary to make such a personal
attack upon me.

Viguier (35), in a valuable paper on a very remarkable Alcyonarian, Mascicularia
edwardsi, in which a number of small Alcyonium-like colonies are connected together
by expansions of the ccenenchym, remarks :—‘‘ Je ne parlerai pas, au cours de cette
discussion, de la classification proposée par Hickson, dans le mémoire cité plus haut.
En voulant séparer d’une maniére aussi absolue les types ot: la multiplication se fait par
bourgeons naissant directement sur les polypes, de ceux ow elle se fait par Vinter-
médiaire de stolons, cette classification, justement critiquée par Koch a un autre point
de vue, avait déja, au moment ow elle a été proposée, linconvénient de laisser en dehors
le Paralcyonium, ot les deux modes se trouvent réunis.”

Viguier’s argument would be perfectly conclusive if my group Stolonifera were based
entirely on the mode of origin of the young polypes. I was probably wrong in assuming
that there was sufficient evidence to lead us to believe that in the majority of
Alcyonaria they are formed by budding directly from the first-formed polypes; in fact,
it seems to be more probable now that in all the Alcyonaria, except, perhaps, the
Pennatulida, the buds are formed in the ccenosarcal canals, which connect the
coelentera of the older polypes. The essential feature of the Stolonifera, a feature in
which the genera of the group differ from all the other Alcyonaria, is that the polypes
arise independently from a creeping basal stolon or (in Tudbipora, in Clavularia viridis,
and the fossil Syringopora) also from horizontal platforms or connecting-tubes ; and
their walls never become fused or cemented together by a growth of the mesoderm
during the whole life of the colony.

The growth of the colony of a Stoloniferan usually takes place at the periphery of
the stolon, that is to say it increases in size horizontally; the only exceptions to this
rule being found in the forms just mentioned, where there is a very considerable
vertical growth, and new polypes are formed above the plane of the stolon. In all the
other Alcyonaria there is, after the youngest stages, very little basal horizontal growth,
but a very considerable distal vertical growth and multiplication by gemmation from
peripheral canals.

The important differences of the mode of growth of the five suborders of the
Alcyonaria may be seen,at a glance by reference to the diagrammatic figures here given.

It was naturally expected that the authors of the volume on the Alcyonaria of the
«Challenger’ expedition (30) would take the trouble to consider and discuss the

ow
bo
“I

GENERA OF THE ALCYONARIA STOLONIFERA.

Fig. 2.

Fig. 1.

NO DOTS ONO Ds

Fig. 1. Schematic section through a Stoloniferan. Fig. 3. Schematic section through a Gorgonid.
Fic. 2. Schematic section through an Alcyonid. Fig. 4. Schematic section through a Pennatulid.

3 B 2

328 DR. S. J. HICKSON—REVISION OF THE

value of classifications that have been seriously put forward in easily accessible publi-
cations; but not only is my group, the Stolonifera, passed over in silence, but the name
is actually employed for a division of the genus Clavularia, without one word of
comment or apology. Anyone who is not well acquainted with the literature of the
group might quite easily infer, on reading the ‘Challenger’ report, that the term
“ Stolonifera ” is used for the first time by these authors. It is quite in keeping with
such work as this that no attempt is made in the volume cited to discuss the value of
the genera of the family Cornulariide (Clavulariidze), which have recently been proposed
without sufficient reason or description; that the peculiar mode of budding of Clavu-
laria viridis is not referred to'; and that, in a word, the whole group remains in the
same state of confusion that it was in before the publication of that colossal memoir.

Classification of the Alcyonaria.

The Order Alcyonaria may be conveniently divided into the following five Sub-
orders :—
1. ProroatcyonaRta . Haimeia, Hartea.
Fam. 1. Clavulariide.

2. STOLONIFERA {
Me WLULUNIEPENA ..eene 5 5
» 2 Tubiporide.

Fam. 1. Alcyonide.
1 » 2. Helioporide.
uO Gio,
Section I. Sclerazxonia.
» Il. Holaxonia.

jhe Fam. Pennatulide.

4, GORGONACEA......... {

ox
ae)
a
2
A
be
3
g
SI
>
Q
I
>

The principal points by which this classification differs from those put forward in
recent times by other authors are the separation of the Protoalcyonaria and Stolonifera
from the rest of the Alcyonaria as separate suborders, and the grouping together into
one suborder the Tubiporide and the Clavulariide.

The value of a system of classification rests upon the correctness of the conception
of the relative values of the characters presented by the animals that are being
classified. A good classification is not necessarily the one in which the different groups
contain an approximately equal number of families or genera. It is generally recog-
nized now, for example, that it is not reasonable to include Amphioxus in the Class
Pisces, but that it is reasonable and far more correct to place this remarkable form in
a group by itself, the Acrania, which is to be considered of equal value to the whole

* To illustrate the importance of this point, I may be allowed to quote some remarks of the late Professor
Moseley (10):—“ The existence of transyerse communicating canals in Olavularia, extending between the
vertical tubes at successive heights above the stolon-tubes, as in Syringopora, is apparently a new fact, and
one of great interest.”

GENERA OF THE ALCYONARIA STOLONIFERA. 329

of the Chordata, with the exception of the Tunicata—the Craniata. Now the genera
Haimeia, Hartea, and Monoxenia differ from all the other Alcyonaria in the remarkable
character that they remain solitary—they do not, in fact, form compound colonies by
gemmation.

This feature surely, by itself, is quite sufficient to justify their separation into a
suborder. It is true there are many points in their anatomy and life-history that
require further investigation, but it is only misleading to group them, even temporarily,
with the Alcyonide, Helioporide, and other families with complicated growth and
gemmiation.

The Stolonifera, again, must be placed in a separate suborder, because in their
mode of budding and in their general anatomy they differ widely from the other
Alcyonaria. Anyone with the smallest experience of the group could distinguish
almost at a glance one of the Stolonifera. He could recognize it as such as easily as
he could recognize an Alcyonian, a Pennatulid, or a Gorgonian. ‘There are, of course,
in all these suborders some genera that present difficulties, but the majority of them
may be quite easily located.

I have very little to add to the remarks I made in two former papers in favour of
my proposition to classify Tubipora with the Stolonifera. J was not by any means
the first to point out the relations between this genus and Clavularia and Cornularia.
In 1834, de Blainville (1) placed these three genera together in one family, “ Les
tubipores ;” and von Koch many years ago regarded T'ubipora as a very primitive form,
closely related to the Cornulariide. The formation of new buds in Clavularia viridis,
from tubes connecting the polypes, similar to the condition which existed in the fossil
Syringopora, is a point which brings the genus Clavularia closer to Tubipora, and this
may be used as an additional argument in favour of my method of classification.

THE STOLONIFERA.

The Suborder Stolonifera may be defined as follows :—Colonial Alcyonaria, with a
membranous or ribbon-like stolon. Mesoglcea poorly developed. Polypes either
entirely free from one another, excepting at their bases, or connected by horizontal
platforms (Zudbépora) or connecting-tubes (Clavularia viridis). Skeleton composed of
calcareous spicules, which may be joined together to form firm tubes (Zudipora), free
from one another, or absent. In some cases the body-wall supported by a horny
secretion.

The Stolonifera contain two families, the Tubiporide and the Clavulariide. To the
former belongs the genus Tubipora alone, distinguished from all the other Stolonifera
by the fact that the spicules join together to form a firm skeleton, and by the presence
of horizontal connecting platforms. To the latter belong four living genera, namely
Clavularia, Cornularia, Stereosoma, and Sympodium, and probably the fossil genus
Syringopora.

530 DR. S. J. HICKSON—REVISION OF THE

The family Clavulariide is practically the same as the Cornulariide of other writers,
but there are very good reasons for changing the name and adopting the one that I
have proposed.

The genus Cornularia is distinguished from the other genera by the absence of
spicules and the presence of a considerable horny secretion on the polype-walls and
stolon. If this genus contained a large number of species, and it were at all a common
thing for the species of the other genera to have horny walls and be devoid of spicules,
Cornularia might be taken as the type of the family. Butit isnot so. Clavularia is
the genus with the largest number of species, and the absence of spicules and the
presence of a horny substance strengthening the walls of the polypes are phenomena
not very frequently met with im the family. It is better, then, to take Clavularia as
the typical genus.

At the time of the appearance of the famous ‘ Histoire Naturelle des Coralliaires,’
by MM. Milne-Edwards and Haime, seven genera were recognized, and they were
arranged as follows :—

( Simples, & polypiéroide tubulaire et 4 polype rétractile — »... 6... eee eee eee Haimeia.
(des stolons radiciformes. ( tubuliformes. ( rétractiles. ( ni costulées ni
| | spiculifécres :
sah Cornularia.
Cornulariés < |
| Polypes < Murailles costulées et
Agrégées, Polypiéroides < : spiculiféres :
| portées < Clavularia.
(Gesu: | |
(non rétractiles. | Rhizowenia.
| (Gvernuciionmes eee reese eer ir Sarcodictyon.
rétractiles .. Sympodium.
(une expansion membrani- Polypes
forme. non rétractiles Anthelia.

If subsequent naturalists had followed closely the characters here given by the
French naturalists, we might have been preserved from the extraordinary state of
confusion into which the group has now fallen; but new species have been included
in the old genera without reference to the characters here given, new genera have been
created without any reason, adequate or otherwise, given in the text, and the figures,
in some cases, have been hopelessly at variance with the descriptions. ‘To give here
just one example of the many I have come across:—The genus Ahizoxenia was estab-
lished by Ehrenberg (5) to include a species that he found in the Red Sea, charac-
terized by the fact that the polypes are not retractile, and Milne-Kdwards and Haime
rely upon this as the one and only character separating this genus from Clavularia and
Cornularia. “Genre IV. Ruizoxenia. Polypiéroides comme dans les genres précé-
dents, mais polypes non rétractiles.”

GENERA OF THE ALCYONARIA STOLONIFERA. 331

Notwithstanding this fact, Sars (26) described a new species as Rhizoxenia filiformis
with completely retractile polypes, and von Koch (17), in describing Dana’s old species,
Rhizoxenia rosea, says the polypes are extraordinarily contractile.

Without giving more examples illustrating the fearful state of confusion of the
group at the present time, I will merely express my opinion that all the old classifi-
cations must now be definitely abandoned, and a new one be formed to take their place.
In order to do this it is necessary to criticize the genera as they now stand.

Since the publication of the ‘ Histoire Naturelle des Coralliaires,’ several genera
have been added, so that we have now thirteen genera in all (omitting the non-
colonial forms Haimeia, Hartea, and Monoxenia).

The Genera of the Stolonifera.

In the report on the Alcyonaria of H.M.S. ‘Challenger,’ Studer and Perceval
Wright (30) include the following sixteen genera in the family Cornulariide :—

1. Cornularia, Lamarck. 9. Cologorgia, Milne-Edwards.
2. Rhizoxenia, Ehrenberg. 10. Cyathopodium, Vervill.

3. Clavularia, Quoy & Gaimard. ll. Scleranthelia, Studer.

4. Sarcodictyon, Forbes. 12. Anthopodium, Vervill.

5. Anthelia, Savigny. 13. Sympodium, Ehrenberg.

6. Gymnosarca, Saville Kent. 14. Erythropodium, Kolliker.

7. Cornulariella, Verrill. 15. Callipodium, Vevrrill.

8. Telesto, Lamouroux. 16. Pseudogorgia, Kolliker.

These sixteen genera fall into three groups :—

(1) Those that have been thoroughly well described and figured, and can be readily
identified as separate genera belonging to the Stolonifera, namely : Cornu-
laria, Clavularia, and Sympodium.

(2) Those that have been only imperfectly described, and had better be incorpo-
rated in the other genera: Rhizoxenia, Sarcodictyon, Anthelia, Gymnosarca,
Cornulariella, and Cyathopodium.

(3) Those that do not come within the limits of my definition of the Stolonifera,
and should be placed in other suborders: Telesto, Celogorgia, Scleranthelia.
Erythropodium, Pseudogorgia, Anthopodium, and Callipodium.

The three genera Rhizoxenia, Sarcodictyon, and Anthelia have been established for
many years, and I feel some hesitation in proposing that they should/now be abolished.
Increased knowledge of the varieties of the species of Clavularia, however, shows that
it isa matter of impossibility to draw any hard-and-fast lines between the forms described
under these generic names and some of the species of Clavularia.

The genus Rhizorenia was established by Ehrenberg (5) in 1834 for those forms
allied to Cornularia and Clavularia, but differing from them in the non-retractility of

332 DR. 8. J. HICKSON—REVISION OF THE

their polypes. Since Ehrenberg’s time new species have been named Rhizoxenia with
perfectly retractile polypes, so that the original character of the genus has been lost.
‘Thus von Koch (17) in a recent paper says that Rhizoxenia rosea (von Koch) differs from
Clavularia in the fact that the tentacles are partially invaginated in retraction as in
Corallium rubrum ; but this is not a character that can be raised to the value of a specific
distinction. I have myself noticed in some species of Clavularia that the base of the
tentacle is, to a certain extent, invaginated, and I have always considered that this.
feature depends upon the presence or absence of a dense deposit of spicules at the basal
part of the tentacles. In those Clavularias with very thick-walled polypes there is no
invagination of the tentacles—they are simply withdrawn ; but in those whose polypes
are thin-walled a partial invagination takes place. ven if there were any value in
this character it is not one, I think, that should be very readily adopted, as it is not
very easy to determine the mode of retraction of the tentacles with certainty without
making a series of sections through a large number of polypes, and it would consequently
be a character of great inconvenience to the systematic zoologist. Until some other
character has been described, the genus Khizovenia must go, and consequently

Rhizoxenia rosea, Dana (3), becomes Clavularia dane.”

Rhizoxenia primula, Dana (3), becomes Clavularia primula.

Rhizoxenia thalassantha, Ehrenberg (5), becomes Clavularia thalassantha.

Rhizoxenia filiformis, Sars (26), becomes Clavularia filiformis.

The genus Sarcodictyon was established by Forbes (12) for a delicate little form with
very thin-walled polypes, which may be completely withdrawn into a thin ramifying
stolon of ribbon-like bands.

There can be no doubt that the Rhizoxenia filiformis of Sars is closely related to
Sarcodictyon catenata of Forbes; but it was hardly fair of Sars to criticize Forbes’s
action in not including it in the genus /hizovenia. Surely it was Sars who was at
fault in placing his species with perfectly retractile polypes in a genus whose main
character was that the polypes were not retractile !

However, I agree with Sars that Sarcodictyon must be given up, and consequently

Sarcodictyon catenata of Forbes becomes Clavularia catenata.

It is difficult, too, to find any very definite character by which to distinguish the
genus Anthelia from Clavularia. ‘The stolon is a membranous expansion, the polypes
are not retractile as a whole, though their tentacles are, and there are numerous
spicules. All of these characters are to be seen in some Clavularias.

Dana says, “The Anthelie cover the rocks or any solid support at hand with thin
fleshy plates, which consist of an aggregation of polypes united by their bases. They
differ from the Xeniw in budding only at the base, which gradually spreads outward by
the process, producing finally the encrusting plate.”

* The name Clavularia rosea has been given to a species from Kerguelen by Studer (28).

GENERA OF THE ALCYONARIA STOLONIFERA. 309:

Anthelia glauca, Savigny, = Clavularia glauca.
Anthelia strumosa, Dana (3), = Clavularia strumosa.
Anthelia purpurascens, Dana (3), = Clavularia purpurascens.
Anthelia desjardiniana, Templeton,= Clavularia desjardiniana.
Anthelia capensis, Studer (28), = Clavularia capensis.
Anthela filippiti, Kolliker (20), = Clavularia filippii.

The name Gymnosarca was given by Saville Kent (13) to a form that exhibits
numerous creeping stolon-like thick expansions, which anastomose and give rise to
free cylindrical stolons on which the polypes are found. It may be that Gymnosarca
should more correctly be placed with Yelesto among the Alcyonida, but the deter-
mination of this point must rest upon some future microscopic examination of the
‘free cylindrical stolons.” If they are really “stolons” (that is to say, if they are not
mainly composed of the fused body-walls of the polypes), there is no reason for
separating this form from the genus Clavularia. Whatever view we take, Gymnosarca
as a separate genus must disappear from the Clavulariide.

The genus Cornulariella was described by Verrill (31) in 1874 in a footnote to a list
of specimens caught in a dredging-expedition on the coast of New England.

The chief point of importance in this form is the presence of large fusiform spicules
with sharp conical projections, which thicken and stiffen the walls of the polype-bodies.
The actual size of these spicules is not given.

It seems to me to be a mistake to place a species in a separate genus on the character
of the spicules alone. ‘he spicules vary enormously in size, in number, and in distri-
bution in the various species of Clavularia. Sometimes they are very large, as in
Clavularia viridis (Plate L. fig. 16), sometimes very small, as in C. garciw, and sometimes
altogether absent, as in one variety of C. australiensis, C. celebensis, &c.

If there were any other distinctive character of this form, it might be well to consider
it a separate genus; but it seems to me, from the very meagre description that is given
(without any figure), that it is nothing more than a species of Clavularia with large
and peculiar spicules. In what respects it is allied to “ Cornularia and Telesto” we have
no information.

It seems to me, then, that Cornulariella modesta should become Clavularia modesta.

The name Cyathopodium was given by Verrill (33) to a species formerly described by
Dana as Awlopora tenuis. (It is obvious that in Verrill’s paper Adlopora is a misprint
for Aulopora.) Verrill says, “It is, in fact, a Zubipora-like polyp with short cup-shaped
cells, connected by narrow calcareous stolons, which correspond to the transverse plates
of Tubipora, and from which the polyps spring.”

I must confess that I fail to see that we have any evidence for supposing that this
form is allied to Tubipora. We have no figure nor description of the polypes, and we
have no figure nor description of the microscopic appearance of the calcareous stolon.

VOL. XIIL.—Part 1x. No. 2.—October, 1894. 36

334 DR. S. J. HICKSON—REVISION OF THE

Moreover, the narrow calcareous stolons of Cyathopodiwm do not correspond in any way
with the transverse plates of Tubipora. They are homologous with the stolons of
Tubipora, and there is nothing like the horizontal platforms ix this form. From the
beautiful drawing given by Dana (3) it appears to be very similar macroscopically to
Forbes’s Sarcodictyon, a genus which, as I have pointed out, must now be abandoned.

Aulopora tenuis (Dana) = Cyathopodium tenue (Verrill) = Clavularia tenuis.

The genera Telesto, Celogorgia, Scleranthelia, and Pseudogorgia seem to me to be
very remotely allied to the Stolonifera, and I do not think it is necessary to discuss
their relations here. They all belong to the Alcyonida.

The genus Hrythropodium was established by Kolliker (20) for the Xenia cary-
beorum of Duchassaing and Michelotti(4). From the description and figures given by
the Italian observers, and the very brief description of the species given by Kolliker, it
is quite unreasonable to accept without further comment the genus Hrythropodium.
It seems to me to be highly probable that Duchassaing and Michelotti were perfectly
correct in their identification of the specimen they examined as a Xenia.

The genus Callipodiwm, established by Verrill (32) in 1869, would probably be more
correctly placed among the Alcyonide than the Clavulariide. ‘The polyps are rather
large and situated at the tops or summits of round-topped verruce, which are more or
less elevated above the surface of the ccenenchyma, and either distantly scattered or
closely crowded together; in the latter case often united laterally nearly to their
summits.” These points seem to indicate, as Verrill remarks, that it is ‘“‘more nearly
allied to the Briareide than to the Cornularide (Clavulariide, mihi), and I am
therefore inclined to regard it as an encrusting genus of the former family, since
even the typical species of the genus Lriarewm is sometimes found growing in broad
encrusting sheets on stones, or parasitically covering the dead axes of many species
of Gorgonide.

The genus Anthopodium, also established by Verrill (34), seems to be closely related to
Callipodiwm on the one hand, and Telesto on the other. There is no reason whatever
for retaining it in the suborder Stolonifera.

Genus CornuLaria, Lamarck (24).
Without spicules. Stolons with a simple cavity. The basal parts of the polypes
and stolons protected by a horny secretion.

Cornularia cornucopie. Naples. von Koch (16) = Tubularia cornucopie, Pallas, Elenchus.
Cornularia aurantiaca, China. Stimpson (27).

The species given by Quoy and Gaimard, Cornularia multipinnata and Cornularia
subviridis, belong properly to the genus Xenia.
A further and more detailed account of the anatomy of this genus is a desideratum.

GENERA OF THE ALCYONARIA STOLONIFERA. 335

Genus CLAVULARIA, Quoy & Gaimard (25).

The definition of this genus given by the French naturalists runs as follows :—

“Animaux cylindriques a huit tentacules pinnés continus dans des claviformes,
coriaces, striés, subpédicules fixes et agglomérés.”

This definition holds good in its entirety for the one species Clavularia viridis only.
The limits of the genus must be considerably expanded to include the species that have
been described since the return of the ‘Astrolabe.’

The genus Clavularia includes those Clavulariide that possess a membranous or
retiform creeping stolon into which the polypes cannot be completely retracted.
Spicules are usually present. No horny secretion of the ectoderm formed.

In order to render the task of identifying the species of Clavularia an easier one, it
is necessary to give detailed lists of some of the principal characters of all the species
that have now been described.

The species of Clavularia arranged approximately in the order in which
they have been described.

Name of Species. Locality and depth. Length of polypes. Authority.
OClavulaniaiuiridis 00230. --\ Vanikoro, shore (?). 50 mm. and over.| Quoy and Gaimard (25).
OL CME 305500065040008 ..| Vanikoro, shore (?). Only a few lines. | Quoy and Gaimard (25).
(OB ATID Polen a ACIS O TE CCRT St. Thomas. 2 Duchassaing and Michelotti (4).
Oh OSH coosdopsooogcndassoo Kerguelen, 120 fath. 9 mm. Studer (28).
Ch POGMUTERR cocccasn000608 Straits of Magellan, 5 mm. Studer (28).
42 fath.
(Ch WHORUHAE oo 5008000000000000 Mediterranean. 10 mm. von Koch (15).
C. crassa = Cornularia crassa | Mediterranean, shallow) 5 mm. Marion and Kowalewsky (23).
(M.-E. & H.) water.
Of FOATOUG 5656 59500000000000 Mediterranean, shallow) 6 mm. Marion and Kowalewsky (23).
water.
CE GRAIG > agockooooo poop 0 Vadso, 50-60 fath. 12-14 mm. Koren and Danielssen (21).
[150 fath., Grieg (6) |.
OC} WORTH e650 oo one00500K000 Throndhjemsfjord. 10-12 mm. Koren and Danielssen (21).
Ob SUP coc 606o0ceDabOOoD Te Throndhjemsfjord. 5 mm, Koren and Danielssen (21).
Ob ROGUE voonacecassooade asc Off Spitzbergen, 260 3mm. Koren and Danielssen (22).
fath.
Co CUVEE pool dancoccndo0GES Sombrero, W. Indies, | 18 mm. Perceval Wright and Studer (30).
450 fath.
Oh, GND 0.0 0000b00000000000 Off Azores, 1000 fath. | 13 mm. Perceval Wright and Studer (30).
OL, CIEE 9¢60000085000000 Off Tristan da Cunha, 5 mm. Perceval Wright and Studer (30).
100-150 fath.
Ob THRO. ao cood000oWD DOH OCD Mediterranean. 10 mm. von Koch (16). |
Oh COMPA ooocornnoaccogasoL AGoING lat., 499M long) ne aca Studer (29). |
1267 fath.
Ch CRISTRAUTISDS ociodooan0c00b0 Port Jackson, Australia,| About 3 mm. Hickson, present monograph.
shallow water.
Ob RUMI coococensagncc0o0se Port Jackson, Australia) About 4 mm. Hickson, present monograph. |
shallow water.
OL ffi o000n00500000000000¢ Port Jackson, Australia,) About 8 mm. Hickson, present monograph.
shallow water.
Ob GORE coonodsodocovogone Diego Garcia, shallow | 9-10 mm. Hickson, present monograph.
water.
Ob PIRLMS cooonon0ho0d000000 N. Celebes, 5-10 fath. 2 Hickson, present monograph.
Cb, COIGISIS coocc0n0g0n00D0c N. Celebes, 10 fath. 8 mm. Hickson, present monograph.

|

3c 2

336 DR. 8S. J. HICKSON—REVISION OF THE

To these must be added the species of those genera that I have shown can no longer
be separated from Clavularia :—

Name of Species. Locality and depth. Length of polypes. Authority.

Clavularia desjardiniana = An- Me de France. | ? Dana (3).

thelia desjardimana. |
C. strumosa= A. strumosa ...... | Red Sea. 1 inch. Ehrenberg (5).
O. glauca= A. glawed.. 1... ++ | Red Sea. 2 Savigny’s figure in de Blainville (1).
C. purpurascens= A. purpurascens, Red Sea. 2 Ehrenberg (5). |
C. capensis=A. capensis ...... | Cape of Good Hope, 9-10 mm. Studer (28).

50 fath. |
C. filippii=A. filippit ........ | 2 | % | Kolliker (20).
C. dane=Rhizoxenia rosea ....| Mediterranean. 2 lines. Dana (3).
CO. primula=R. priomula ...... | Fiji Islands. 4-5 lines. | Dana (3).
C. thalassantha = R. thalassantha .| Moluccas. | ? | Ehrenberg (5).
C. filiformis=R. filiformis ...., Coasts of Norway, 30- ? | Sars (26).
| 40 fath.

C. catenata=Sareodictyon cate- British, deep water. | ........ Forbes (12).

nata.
O. colinabum=S. colinabum .... Scotland. ? | Forbes (2).
C. bathybius = Gymnosarca cas Cezimbra, Portugal, | ........ | Saville Kent (13).

bius.
C. modesta= Cornulariella mo- Gulf of St. Lawrence, | 6-18 mm. Verrill (31).

desta. | 80-100 fath.

To these must be added two species, Rhizoxvenia alba and Sympodium margaritaceum,
described by James Grieg from the coasts of Norway, both of them being undoubtedly
species of Clavularia :—

Name of Species. Locality and depth. Length of polypes. Authority.
Clavularia alba=Rhizoxenia alba) 68° N. long., 9° 44’ E. |... Grieg (6).
lat., 634 fath. |
| CO. margaritaceun=Sympodium | 63° N. long., 5° E. lat., |
|

|

5 | é A
margaritacenm. ee2zovemathe

The following species have been too imperfectly described to enter into any modern
system :—Anthelia rubra of delle Chiaje, Anthelia olivi and Anthelia domuncula of
de Blainville.

Genus Sympopium (Ehrenberg).

Some doubt has been thrown upon the stability of this genus by the recent investi-
gations of von Koch (18) upon Sympodium coralloides, who has shown that this form
is a true Alcyonid adapted to live on a Gorgonia stem, and proposes that its name should
be changed to Alcyoniwm coralloides. It is possible that, when they are properly
examined histologically, the other species attributed to Sympodiwm may turn out tc be
after all true Alcyonians with a modified habit. For the present, however, it may be
allowed to stand provisionally, and the reader may be referred to the excellent account of
the genus in the ‘Challenger’ volume for the species that have been hitherto described.

GENERA OF THE ALCYONARIA STOLONIFERA. 337

The genus includes those Clavulariide with a thick plate-like stolon into which the
polypes may be completely retracted.

Description of a new Genus (Stereosoma), and of new Species of Clavularia.

STEREOSOMA, gen. nov.
STEREOSOMA CELEBENSE. (Plate XLV.)

The only specimen of this interesting new genus known to me is one that I found
growing on the reefs of Talisse Island in North Celebes. It does not occur in great
abundance on any of the reefs that I visited; in fact the only specimen I found after
months of reef-wading in search of specimens was a small colony bearing five or six
polypes attached to a piece of water-logged wood.

The genus can be at once recognized by two important characters, the first being
that it shows no power of retracting either its body-wall or tentacles, and the second
that the pinne of the tentacles are separated from one another by very considerable
intervals.

The absence of contractility is a remarkable feature.

Many Alcyonaria are usually described as not contractile, but the description is
seldom perfectly accurate.

Polypes that possess a great number of densely-packed spicules take a long time to
contract, and they may be removed from their habitat, placed in ordinary spirit, and
be preserved without showing very much contraction. Again, many Alcyonarians
that do not exhibit any considerable power of contracting their body-walls may
contract their tentacles.

Now Stereosoma possesses no spicules, and the tentacles show no more power of
contracting than the body-wali.

The illustration given (Plate XLV.) is a faithful representation of the specimen
as it reached England on my return from Celebes, the colour alone having been added
from notes that I made at the time of its discovery.

On making a series of transverse sections through one of the polypes I found that
the ectoderm is remarkably thick, and presents a very vacuolated appearance.

Between the vacuoles and the ectoderm covering the body there may be seen a
number of isolated cells, islets of cells, and strings of cells (Plate L. fig. 1). ‘These are
undoubtedly derived from the ectoderm, and probably secrete the tough, vacuolated,
homogeneous substance that surrounds them and lies between them and the mesoglea.
I have had no means of ascertaining what is the precise chemical nature of this
substance, but it is undoubtedly of a horny consistency. It stains deeply in borax
carmine, and can be readily distinguished from the true mesoglcea which les below it.
It is a point of some importance, in comparing this genus with Cornularia, that in

538 DR. S. J. HICKSON—REVISION OF THE

Stereosoma the horny substance that is formed lies inside the ectoderm, whereas in
Cornularia it is outside it.

The mesenteries present well-marked muscle-ridges and muscular bands. The
muscles are used for producing the graceful swaying and bending movements that I
noticed in the living condition.

There is a small and not very well-defined siphonoglyphe (Plate L. fig. 2), and the
walls of the stumodeum in preserved specimens are slightly folded.

There are no spicules in the body-walls, tentacles, or stolon.

The stolon is a moderately thick plate-like structure containing numerous ramifying
canals.

The tentacles are long and delicate, and present the remarkable feature of possessing
only a few small teat-like pinne, separated from one another by considerable intervals.
In this feature Stereosoma presents a character that seems to separate it from all the
other Stolonifera. In all the species of Clavularia that I have examined the pinne
are exceedingly numerous and very closely set, so that the tentacle has a considerable
resemblance to the vexillum of a feather. ‘The tentacle of Stereosoma has no resem-
blance whatever to a feather.

The elongated slit-shape mouth is situated on the top of a prominent conical
hypostome.

There are no external ridges or other markings on the body-walls.

The genus may be defined as follows :—

Clavulariide forming small colonies, consisting of stiff non-retractile polypes situated
at considerable intervals from each other on a thick plate-like stolon. Tentacles
non-retractile. Pinnee few and widely separated. Spicules absent.

One species, Stercosoma celebense. Polypes 15 mm. long, 3 mm. in diameter ;
tentacles 10 mm. long, with from 5 to 10 pinnz on each side. Colour pale brown.

Locality. Shore reefs on southern part of Talisse Island, North Celebes.

Genus CLAVULARIA.

CLAVULARIA AUSTRALIENSIS, Hickson (11), Variety A. (Plate L. fig. 3.)

Specimen 1. Stolon thin and membranous, forming in some places sympodial plates,
in others broad and narrow strands.

Polypes partially retracted into the stolon, forming protuberances on its surface,
0°5 mm. in diameter, 1:0 to 1:5 mm. in height, and about 2 mm. apart.

Spicules numerous, simple, multituberculate, 0-14 to 0°18 mm. long (Plate L. fig. 4).

Colour in spirit white.

The specimen sent to me is parasitic on a piece of sponge.

An interesting point in connection with this specimen is the enormous number of
zooxanthelle in the intermesenterial spaces (Plate L. fig. 5).

GENERA OF THE ALCYONARIA STOLONIFERA. 539

I am convinced, from an examination of a large number of specimens of Alcyonarians,
that the number of the zooxanthelle in the ccelenteron cannot be used for purposes
of classification, since polypes of the same species, and even of the same colony, show
very great variation in this respect. Generally speaking, shallow-water polypes possess
more than those that live in deeper water, the zooxanthelle being probably dependent
for their growth and multiplication upon the intensity of the daylight. But whether
this is the only cause or not does not seem certain. At any rate we can say that it is
highly probable that the conditions favourable for the growth and multiplication of
the zooxanthelle are not precisely the same as those favourable for the growth and
multiplication of the Alcyonarian colonies, and thus the variations in the number of
these symbiotic algze in the polypes may be accounted for.

Specimen 2. The stolon is somewhat thicker in the central part of its area than in
Specimen 1. It is membranous, but becomes divided at the edges into broad and
narrow strands. The polypes are densely crowded on the central parts of the stolon,
but scattered at the edges.

The polypes are partially retracted, but not to such an extent as in Specimen 1, the
adpressed tentacles remaining visible in more than fifty per cent. of the polypes.

The average height of the partially retracted polypes is 3 mm., and their diameter
1:5 mm.

The walls of the polypes are thicker than they are in the first specimen, with dark
brown corrugated outer surfaces.

Spicules resemble those in Specimen 1, but the tubercles are slightly longer and less
numerous. Average length 0:15 mm. (Plate L. fig. 6).

The colony is parastic on a piece of sponge.

CLAVULARIA AUSTRALIENSIS, Variety B. (Plate L. fig. 7.)

The three specimens that I have grouped together as Variety B of this species are
distinguished from the others by the absence of spicules. This fact in itself might be
considered by some naturalists to be sufficient reason for the establishment of a new
species, or even a new genus, for their reception; but after a careful examination of the
anatomy of the specimens, their mode of growth, structure of the tentacles, and
general anatomy, I am convinced that we are not justified in separating into different
species those forms that differ mainly in the presence or absence of spicules. I believe
that it is quite possible that in some localities, where there is but little lime in
the water and an abundance of sand, the Clavularias do not develop spicules. This
is sufficient to constitute a separate local variety, but not a species.

Associated with the absence of spicules in this variety there may be noticed a
difference in the character of the ectoderm from that of Variety A (Plate L. fig. 8).

The ectoderm of Variety A is over the great part of ts surface smooth and columnar,
each cell being marked off from its neighbours by very definite cell-outlines. In

340 DR. S. J. HICKSON—REVISION OF THE

Variety B, on the other hand, the ectoderm is highly vacuolated and its surface
irregular. The cells have branched processes which anastomose with one another, and
it is impossible to determine, in most cases, where one cell ends or another begins.
I believe that this vacuolated ectoderm is considerably stiffer or firmer than the simple
columnar ectoderm of Variety A, and that it is formed for the support of the body-
wall in the absence of spicules.

In Specimen 1 of this variety the stolon is thin and membranous, dividing into
ribbon-like pieces at the edges. The polypes are partially retracted and densely
crowded on the parts of the colony with a membranous stolon, but fully expanded and
widely separated from one another at the edges.

The colour is brownish yellow, due to a considerable deposit of sand.

The colony is parasitic on an Ascidian test.

Specimens 2 and 3 are probably young examples of the above. ‘The stolon is very
thin and composed of a number of anastomosing bands or ribbons. Most of the polypes
are fully expanded. They are both parasitic on mussel-shells.

Locality. Coast of Victoria, shallow water.

CLavuLaRria RAMOSA, Hickson (11). (Plate L. figs. 9 & 10.)

Stolon composed of a number of thin branching strands clinging to the branches of
aseaweed. The strands of the stolon are usually about $ mm. in breadth, but never
exceed 1 mm.; they do not fuse to form membranous or plate-like expansions.

The polypes spring from the branches of the stolon singly at intervals of 3 mm.
The youngest polypes are found at the ends of the youngest branches.

New polypes apparently never arise between the older polypes, but each polype is
formed in succession at the end of the growing branch of the stolon.

The ramifications are formed by a simple bifurcation of the growing point of a
branch, and they are produced quite independently of the position of the youngest
polypes. Sometimes a polype may be seen springing from the angle of a bifurcation,
but more frequently there is no polype in this position.

Judging from spirit-specimens only, the polypes are not capable of complete retrac-
tion into the branches of the stolon. In the retracted condition they are funnel-
shaped. ‘The bread rounded distal extremity, 1‘5 mm. in diameter, contains the
retracted calyx. The narrow proximal extremity at the point of attachment to the
stolon is *O mm. in diameter.

The distal extremity is marked by eight deep furrows.

The spicules are numerous, both in the stolon and the polype-walls; they are
double clubs 0°1 to 0°15 mm. long. In the tentacles there are a few elongated lancet-
shaped spicules with irregular dentate projections (Plate L. fig. 11).

In spirit the specimens are dirty yellowish white in colour.

Locality. Coast of Victoria, shallow water.

GENERA OF THE ALCYONARIA STOLONIFERA. 341

CLAVULARIA FLAVA, Hickson (11). (Plate L. fig. 12.)

The stolon is thin and ribbon-like, not coalescing into membranous plates. There
are comparatively few polypes situated on the stolon at intervals of 4to6 mm. At
the edges of the stolon there are frequently to be seen considerable areas devoid of
polypes.

The polypes have in all cases their crowns retracted, and the tracts of insertion of the
mesenteries are not indicated externally by longitudinal grooves or lines. This feature,
connected probably with the denseness of the mesogloea and the great number of
spicules, is quite sufficient in itself to distinguish this species from C. australiensis.

The length of the polypes, as they are seen in spirit with their tentacles retracted,
is 4-6 mm., the diameter 1°5 mm.

The spicules are of a bright yellow colour, and form a dense armature for the
polypes and stolon. On slicing off a piece of the body-wall and examining it with the
microscope the spicules appear to be locked together to form a compact skeleton.

The spicules are 0-1 to 0:15 mm. in length, and are of three kinds :—(qa) short and
broad double cones with numerous blunt tubercles, found in great number in the
mesogloea of the body-walls of the polypes and the stolon; (6) elongated style-like
spicules, with very few short and pointed tubercles, found principally in the tentacles;
(c) a few spicules of irregular shape that I have never seen in situ (Plate L. fig. 13).

The colour of the specimens in spirit is orange. They are situated on fragments of
an old lamellibranch (oyster ?) shell.

Locality. Coast of Victoria, shallow water.

CLAVULARIA GARCIA, sp. nov. (Plate XLVI.)

The stolon is in the form of a thin membranous plate about 1 mm. in thickness.

The polypes are evenly distributed over the stolon, and separated from one another
by short intervals. When looking down upon a spirit-specimen it appears as if the
polypes were densely crowded, on account of the long tentacles and pinne, but on
carefully separating the polypes with needles it is clear that there are considerable
intervals between their bases.

The polypes have remarkably thin and transparent walls containing a number of
very small scattered (not crowded) spicules; the mouth is very small and situated at
the extremity of a teat-like papilla in the centre of the oral disk. Each polype is
from 9-10 mm. in length.

The tentacles are about 5 mm. in length, very thin-walled, and bear on each side
about 30 long hollow pinne. The great length and number of the pinne give the
species a very fluffy or downy appearance quite peculiar to it.

Neither the polypes nor the tentacles show any signs in the spirit-specimens of a
power of contraction.

VOL. XIII.—PART Ix. No. 3.—October, 1894. 3D

342 Dk. 8. J. HICKSON—REVISION OF THE

Every polype is fully expanded. ‘This is a noteworthy feature in a species with such
thin-walled tentacles and polypes. It must be noted that it is highly improbable that
in the natural condition the polypes can retract, for there are no spaces in the basal
stolon that could contain them.

Another very remarkable feature of the species is the minuteness of the spicules.
They are a great deal smaller than the spicules of any species of Clavularia I have
yet examined (Plate L. fig. 15). They are all of one kind, namely, rhombic in shape,
with the angles rounded off, and they show a number of extremely minute thorn-like
projections. Each spicule measured 0:05 mm. in length and 0-003 mm. in breadth.

The specimen now in my possession was kindly given to me by Mr. G. C. Bourne,
who found it in shallow water on the reefs of Diego Garcia, in the Chargos Archipelago.

CLAVULARIA REPTANS, sp. nov. (Plate XLVIL.)

This species of Clavularia is quite different in habit from any species yet described.
The stolon consists of thin strands creeping over pieces of dead branched coral, in
many cases stretching across the spaces between the branches, forming bridges on
which polypes may arise. The important point about this form of stolon, and one
upon which I was inclined to lay special stress, is the extraordinary area over which
each colony extends, and the absence of any special point of concentration. When
dredging off the coast of Talisse I often fished up bits of coral, much too large for my
collecting-jars, that had this species of Clavularia growing over it in a form that
reminded me of a very wide-meshed net. The whole colony grows, in fact, like a
Canariensis creeper, clinging to any projecting branch that may be in its vicinity.

The breadth of the stolons averages 1 mm., the diameter of the contracted polypes
2 mm., and the length of the expanded polypes 7 to 10 mm.

It should be noted here that it is very rarely the case in Clavularia that the
diameter of the retracted polype is actually greater than the average breadth of the
stolon from which the polype springs. This character, then, is one which helps us to
distinguish Clavularia reptans from other species of the genus.

The tentacles of this species are rather short and provided with numerous densely
packed pinne, resembling somewhat the tentacles of Clavularia garciew. Spicules
absent.

Locality. 1 have only found this species at depths of 5 to 20 fathoms in the Banka
Straits, North Celebes.

CLAVULARIA CELEBENSIS, sp. nov. (Plate XLVIII.)

I have established this species for a small specimen of Clavularia I found off Talisse
Island in 10 fathoms of water on an old water-worn branch of a madrepore.

In habit it is very similar to Clavularia viridis, but differs from it in several
important points of structure.

GENERA OF THE ALCYONARIA STOLONIFERA. 343

The stolon is composed of thin strands varying from 1 to 3 mm. across, which
coalesce at intervals to form small plate-like expansions. The polypes are of various
sizes, the largest I have measured being 8 mm. long, with a maximum of 2°5 mm. in
breadth. They have very thick walls, due to an extreme development of the mesoglea.
No grooves or lines of any kind mark externally the insertion of the mesenteries.
The tentacles are long and pointed in life, and provided with numerous densely
crowded pinne. The polypes are not capable of any very great contraction, but the
crown of tentacles can be introverted into the anterior part of the polype-walls.

As I have had at my disposal such a small specimen I have not made as complete a
study of the anatomy of this species as I should wish, but in the fragments of the
stolon and polypes I have examined with the microscope I have not been able to find
any trace of spicules. It is possible that they exist, for I find that it is never safe to
state that there are no spicules in any species unless several polypes and a large piece
of the stolon have been boiled in potash and the residue examined with the microscope.

The colour of the stolon and body-walls is the usual dull olive-brown, but the
pinnee of the tentacles are bright green.

When examined alive with the polypes expanded this species is one of the most
beautiful, delicate, and graceful Alcyonarians I have ever seen.

Locality. Talisse Island, N. Celebes ; shallow water.

CLAVULARIA VIRIDIS, Quoy & Gaimard. (Plate XLIX.)

I published, in the ‘Proceedings’ of the Royal Society, 1886 (10), a preliminary
account of some observations on this species that I made when I was resident in Celebes.
I then pointed out the existence of the remarkable tubes connecting the polypes, and
the similarity of the expanded polypes, both in form and colour, to those of Tubipora.

Since my return to England I have made a few more observations upon its anatomy.

The species may be found in abundance on most of the coral-reefs of North Celebes,
and probably occurs on the shores of nearly all the islands of the Malay Archipelago.

Quoy and Gaimard, who originally described the species, found it at Vanikoro, and
Wallace obtained some specimens, which are now in the British Museum, from the
Aru Islands.

Its usual habitat is, like Tuézpora, on the shore side of the reef, where it is left
exposed to the air at low water of spring tides. It occurs either in large clumps
five or six inches in height and over a foot in diameter, or in small creeping colonies
clinging to dead water-worn coral branches.

When dried in the sun it leaves a firm but brittle skeleton, composed of a plexus
of irregular branching fibres, which fuse into a continuous sheath in the lower parts.
This skeleton retains the original form of the contracted colony.

In colour the expanded polypes are either olive-brown or green, or any of the
intermediate colours between the two.

3D2

344 DR. S. J. HICKSON—REVISION OF THE

The length of the polypes varies according to the size of the colony and the mode
of growth. The longest tube I have measured is four inches, but the average length
is not more than one or two inches.

When the tide goes down the crown and neck of the polypes are slowly but
completely retracted within the firm walls of the lower part of the polype-tubes. In
this firmly retracted condition they retain a considerable quantity of sea-water in their
coelentera, and they are able, in consequence, to withstand exposure to ane 2 air and sun
for an hour or two.

The stolon consists of a network of tubes and strands clinging to the supporting
coral blocks. In some places these strands are somewhat expanded, but I have never
found any very extensive membranous plates in this species.

There are no spicules in the tentacles nor in the crown and neck, but in the lower
parts of the body-walls of the polypes there are a few very large calcareous spicules.
They are long spindles beset with numerous small spines and tubercles. Their average
size is 2°3 mm. long by 0°14 mm. broad (Plate L. fig. 16).

A series of sections through the polypes shows that the muscular Sitges on the
mesenteries are very numerous and long (fig. 17) ; in fact the mesenterial muscles of
Clavularia viridis are stronger than any I have met with in the Alcyonaria, excepting
perhaps Tubipora (9).

The mesoglcea is very thick for a Stoloniferan, both in the tentacles and body-wall.

The spicules are situated in the mesogloea, and in transverse sections of decalcified
specimens empty spaces may be seen, indicating the places that they formerly occupied.

The horny skeleton is formed by some modification of the mesogloea. It occurs in
the form of a number of very dense fibres, which are figured in transverse sections in
Plate L. fig. 18.f. They appear in the form of deeply stained cores situated in the
centres of wide lacunz in the homogeneous mesoglea.

It is difficult to determine the exact chemical nature of this horny skeleton, but it
is apparently closely related to keratin.

It is insoluble in weak and strong nitric and hydrochloric acid. It is partially
soluble in strong hot sulphuric acid. It is not digested by pepsin and ‘2 per cent.
hydrochloric acid, nor by solution of pancreatin.

On burning it gives a pungent and somewhat aromatic odour.

In origin it differs from the horny skeleton of Cornularia and Stereosoma in being
a product of the mesoglea. There is nothing that resembles it in any other species
of the genus.

List of the Literature referred to.

I. DE Buarnvitty.—Manuel d’Actinologie.
2. DELLE Cutase.—Descriz. e anotom. degli Anim. iny. della Sicilia Citer. t. v. p. 160, fig. 5.
3. Dana, J. D.—United States Exploring Expedition. Zoophytes. 1848.

29.

30.

GENERA OF THE ALCYONARIA STOLONIFERA. 346

. Ducnassaine and Micurtorri1.—Sur les Coralliaires des Antilles, Mem. della R. Accad. d.

Torin. il. tom. xix. 1860.

. Enrenpere, C. G.—Corallenthiere d. rothen Meeres. 1834.

Forses, vide JoHNSTON.

. Grine, James A.—To nye Cornularier fra den Norske kyst. Bergens Museum, No. 3, 1887.
. Herpman, W. A.—On the Structure of Sarcodictyon. Proceedings Roy. Soc. Edinb. viii.

p. dl.

. Hickson, 8. J.—On the Ciliated Groove (Siphonoglyphe) in the Stomodzum of the Alcyo-

narians. Phil. Tr. 1883.

. Hickson, S. J.—The Structure and Relations of Tubipora. Q. J. Micr. Sci. 1883.
. Hickson, S. J.—Preliminary Notes on certain Zoological Observations. P.R.S. 1886.
. Hickson, S. J.—Preliminary Report on a Collection of Alcyonaria and Zoantharia from

Port Phillip. P. R. Soc. Vict. 1890.

. Jounston, G.—History of British Zoophytes. 2nd ed. 1847.

. Kent, W. Savitte.—Two new Genera of Alcyonoid Corals. Q. J, Mier, Sci. vol. x. p. 397, pl. xxi.
. Kiunzincer, C. B.—Die Korallthiere des rothen Meeres. 1877,

. von Kocn, G.—Anatomie der Clavularia prolifera, un. sp., nebst einigen vergleichenden

Bemerkungen. Morph. Jahrb. vol. vii. p. 467. 1882.

. von Kocu, G.—Die Alcyonacea des Golfes von Neapel. Mittheil. a. d. zool. Stat. Neapel,

vol. ix. p. 652. 1891.

. von Kocu, G.—Kleinere Mittheilungen iiber Anthozoen. Morph. Jahrb. xvi. 1890.
. von Kocu, G.—Die systematische Stellung von Sympodium coralloides. Zool. Jahrb. v.

Heft 1, pp. 76-92.

. von Kocn, G.—Die Gorgoniden des Golfes von Neapel. 1887.
. Kotter, A.—Icones Histiologice. 1864.
. Koren and Danretssey.—Nye Alcyonider, Gorgonider og Pennatulider tilhérendes Norges

Fauna. Bergens Museum, 1883.

. Koren and Danretssen.—Norske Nordhavs-Expedition. Alcyonida. 1887.
. Kowatewsxy, A., and Marron, A. F.—Documents pour l’histoire embryogénique des Alcyo-

naires. Annales du Musée de Marseille, vol. i. Mémoire 4. 1883.

. Lamarcx.—Hist. des Animaux sans Vertébres. 1816.

. Quoy and Garmarpv.—Voyage de l’Astrolabe. 1834.

. Sars, M.—Fauna littoralis norvegice. Part II. 1856.

. Strurson, W.—Descriptions of some of the new Marine Invertebrata from the Chinese and

Japanese Seas. Proc. Acad. Philad. Nat. Sci., May and June, 1855.

. Sruprr, T.—Alcyonaria der Gazelle. Monats. d.k. preuss. Akad. d. Wiss. Berlin, October

1878, p. 682.

Sruprr, T.—Note préliminaire sur les Aleyonaires provenant des Campagnes du yacht
VHirondelle, 1886, 1887, 1888. Part 2. Aleyonacea and Pennatulacea. Mém. Soc. Zool.
iv. pt. 2, pp. 86-95.

Wrient, P., and Stuprr, T.—Report on the Alcyonaria collected by H.M.S. ‘ Challenger.’
Zoology, xxxi. 1889.

. Verritt, A. E.—Results of recent Dredging Expeditions on the Coast of New England. Am.

J. Sci. 1874, ser. 3, vol. vii. p. 40.

346 DR. 8. J. HICKSON—-REVISION OF THE

32. Vurritt, A. E.—Notes on Radiata. Trans. Connecticut Acad. vol. i. 1868, pt. 2, no. 6, p. 455.

33. Vurritt, A. H.—Critical Remarks on the Haleyonoid Polyps in the Museum of Yale College.
Am. J. Sci. 1868.

34. Verrity, A. H.—Radiata from the Coast of North Carolina. Am. J. Sci. 1872.

35. Vieurrr, C.—Un nouveau type d’Anthozoaire (Fuscicularia edwardsi). Arch. Zool. Expér.
2° série, vol. vi.

EXPLANATION OF THE PLATES.

PLATE XLV.

Stereosoma celebense, p. 357.

PLATE XLVI.

Clavularia garcie, p. 341.

PLATE XLVII.

Clavularia reptans, p. 842.

PLATE XLVIUTI.

Clavularia celebensis, p. 342.

PLATE XLIX.

Clavularia viridis, p. 343.

PLATE L.

Structure of Stereosoma and Clavularia.

Fig. 1. Transverse section through a portion of the body-wall and one mesentery of
Stereosoma celebense, showing the thick vacuolated ectoderm, consisting of
an outer layer of cells and a subjacent dense homogeneous substance con-
taining a number of isolated cells, rods of cells, and cell islets (Hct.'), as well
as the vacuoles or lacune. The mesogloea is sharply defined and is not
vacuolated. A considerable number of zooxanthellae may be seen adhering
to the endoderm.

Fig. 2. Outline drawing of a transverse section through Stereosoma celebense in the
region of the stomodzum, showing the small but prominent muscular ridges
and the siphonoglyphe.

Fig. 8. A small specimen of Clavularia australiensis, Variety A.

Fig. 4. Two forms of the spicules of Clavularia_australiensis, Variety A.

JO Oe

Fig. 8.

GENERA OF THE ALCYONARIA STOLONIFERA. 347

. Transverse section through a polype of Clavularia australiensis, Variety A,

showing the enormous number of zooxanthellee adhering to the endoderm,
and the simple columnar form of the ectoderm.

. Two forms of spicules found in another specimen of Clavularia australiensis,

Variety A.

A specimen of Clavularia australiensis, Variety B, showing the ribbon-like
character of the stolon at the edges. One of the polypes is fully expanded,
but all the others are in different stages of retraction.

Transverse section of a polype of Clavularia australiensis, Variety B, showing
that there are only a few zooxanthelle adhering to the endoderm (compare
C. australiensis, Variety A, fig. 5). ‘The ectoderm is thick and vacuolated,
the cells being irregular in shape. The siphonoglyphe is large and well-
defined.

. A small portion of a colony of Clavularia ramosa, growing on a ramifying

sponge.

. The growing point of a colony of Clavularia ramosa, showing two young

polypes.

. Three forms of spicules found in Clavularia ramosa: a, a spicule from the

body-wall ; 6, two spicules from the tentacles.

. A specimen of Clavularia flava growing on a piece of oyster-shell.
. Three forms of spicules found in Clavularia flava.
. Transverse section through a portion of the stolon of Clavularia flava, showing

four endodermic canals in section.

. Spicule of Clavularia garcie.
. A spicule of Clavularia viridis.
. Outline sketch of a transverse section through a polype of Clavularia viridis,

to show the large and deep muscular ridges on the mesenteries.

. Transverse section through a portion of the body-wall of Clavularia viridis,

showing the horny fibres, h.f., and the lacune left after the solution of the
calcareous spicules in the mesoglea.

Reference letters used in all the figures:—ct. Ectoderm; Hnd. Endoderm; Lae.
Lacune; Mes. Mesoglea; Musc. Muscular ridges; Siph. Siphonoglyphe ;
Stom. Stomodeum ; Zr. Zooxanthelle.

0 pon
ih eM
the)

any

ave y

mere ha
Ht

era

ii

= ASNHEGHTHO VNOSOUEALS

comm LEC gL

‘der wermeyy sey
; : yee burr at

AMY UN De PFs 7° Zp COGS

att
in b
i
|
iy ‘

L.BKing del.

West, Newman imp.

MP Parker chromo.

CLAVULARIA GARCIL..

in
aie oP

Oe

West,Newmiari imp.

MP. Parker chrome,
L.B.King del.

CLAVULARIA REPTANS,

" ite
i aA OE Oty
ra

i CEE. eC FEMI

TB Kang del.
MP. Parker chromo.

West, Newman inp.

CLAVYVULARIA CHLEBENSIS,

Sea hemes on

West, Nevnnan imp.

Cec

aol ee UO ET Oa

=
Ni ccrsinteetererecueiee ase ecpe-ar

Si

CLAVULARIA VIRIDIS.

ne HAA DAE

AWEHickson del.
MBP Parker chromo.

Frans. Lool So Vol smFY £.

S.J. del, West N :
M.PParker lith. est,Newman inp.

STRUCTURE OF STHREOSOMA AND CLAVULARIA.

otc nay:
“i Hi 4)
ID)

Cay,
CoN Maal

ian

eines
sie ip
Sata

nas

XIV. Descriptions of nine new Species of Amphipodous Crustaceans from the Tropical
Atlantic. By the Rev. Tuomas R. R. Srespine, M.A.

Received November Ist, 1892, read December 20th, 1892.
[Puates LI.-LV.]

THE specimens described in this paper were obtained by Mr. John Rattray, during
the expedition of the ‘Buccaneer,’ the telegraph-ship belonging to the Silvertown
Company, when engaged in surveying for the laying of cables on the West Coast
of Africa. The scientific investigations made during the expedition were arranged
for, and the expenses met by, Dr. John Murray, the Director of the ‘ Challenger’
Commission, and Mr. J. Y. Buchanan, the latter of whom accompanied the ship.

Tribe HY PERIDEA.

Family SCINIDA, Stebbing, 1888.

The head is small, of less width than the pereon. ‘The eyes are small. The first
antenne are large, straight, generally (perhaps always) three-jointed, attached at the
front corners of the head. ‘The second antenne are attached to the underside of the
head; they are rudimentary in the female, but in the male become long and slender,
after being at an early stage short and curved one over the other. ‘The mandibles are
without palp. Both pairs of maxilla are well developed. The maxillipeds have a
small inner plate, and two large outer plates which are distally narrowed. Both first
and second gnathopods are simple. Of the persopods the third are generally the
longest, the fifth always the shortest. The pleon is narrower than the pereeon. The
fifth and sixth segments are generally (perhaps always) coalesced. In the uropods
only the outer branch is free. The telson is small.

Definitions of this family have been given recently by Dr. Bovallius, Professor Chun,
and Professor Sars. With all of these the above substantially agrees. Chun includes
the character that the body is not compressed, which will not apply to the new species
Scina stenopus, and is rather vague in its application to other species. Sars speaks of
the first antenne as divergent, an epithet which is unsuitable, since, though capable of
great divergence, they can lie with the inner margins perfectly parallel, and one may
even suspect that this is their natural position when at rest. Both Bovallius and Sars
speak of the second antenne in the male as angularly bent. ‘This angulation, it may
be remarked, is distinct from the zigzag folding familiar in several other Hyperid
genera. As Streets has explained, it merely refers to a single bend at one point of the

VOL. XIl.—Part x. No. 1.—ebruary, 1899. 35

350 REV. T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

long slender flagellum, when that, not being in use, is laid for security beneath the
animal’s body. For the two species which Dana named Clydonia gracilis and Clydonia
longipes, Bovallius accepts from Dana a division between the fifth and six segments
of the pleon. Streets, in describing what he regarded as a specimen of Clydonia
longipes, says that the fifth and sixth segments in question are apparently consolidated.
Since in all the species which have been recently examined these two segments are
coalesced, it is most probable that their separation was not observed by Dana, but
taken for granted, contrary to the actual fact, although according to what is normal in

the Gammaridea, to which he assigned the genus Clydonia.

Genus Scina, Prestandrea, 1855.

1833. Scina, Prestandrea, Effemeridi scientifiche e letterarie per la Sicilia, t. vi. p. 10.

1840. Tyro, Milne-Edwarads, Hist. Nat. des Crustacés, t. 11. p. 80.

1849. Clydonia, Dana, American Journal Sci. and Arts, vol. vii. no. 22, p. 140.

1885. Tyro, Bovallius, On some forgotten Genera among Amph. Crust. p. 12.

1887. ,, Bovallius, Monogr. Amph. Hyperiidea, K. Svensk. Vet.-Akad. Handl. Bd. xxi.
no. 5, p. 9.

1888. Scina, Stebbing, Challenger Amphipoda, pp. 151, 1271, &e.

1889. Fortunata, Chun, Akad. der Wissenschaften zu Berlin, Math. u. naturwiss. Mitth. p. 342.

1889. Scina, Chun, Zoologischer Anzeiger, Jahrg. xii. no. 308, p. 286.

1890. ,, G.O. Sars, Crustacea of Norway, vol. i. pt. 1, p. 18.

The other references will be found in the Monograph of the Hyperidea by Bovallius
and in the Report on the ‘Challenger’ Amphipoda. The genus at present stands by
itself, and may therefore be content with the character of the family. Nevertheless,
Bovallius and Sars have drawn out separate generic descriptions. Bovallins in his
definition speaks of the third perzeopods as “transformed into jumping-legs.” For
such a function they do not seem particularly well suited. The long and strong second
joint is directed forward and upward, and is prolonged into a spine-like process at
the apex of the front margin. According to Professor Chun, by aid of these processes,
the animal attaches itself to a free-swimming hydrozoon, and floats about without
exertion after the fashion of Phronima and various other Hyperidea. Sars states that
the third pereeopods are the longest, which is indeed usually a conspicuous feature ; but
the Scina bovallii of Chun is said to have the fourth pereeopods somewhat longer than
the third, and Scina clausi (Bovallius) to have the fourth as long as or a little longer
than the third. Of the uropods Sars declares the first and second pairs to be “simple,
with the peduncle not defined.” Yet both in his own species, Scina borealis, and in
all the other known species, with one or two doubtful exceptions, the extent of the
peduncle is defined in all the uropods by the presence of a free outer branch, albeit
that branch is sometimes extremely small and spine-like. There are more or less
conspicuous gland-cells in the limbs of the perzon and the uropods. ‘The pleopods

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 301

are provided with a couple of serrate coupling-hooks on each peduncle, and a single
cleft spine on the first joint of the immer ramus. At least in general there are fewer
joints to the inner than to the outer ramus.

The species belonging to this genus are now numerous. Those earliest described
remain, and will probably for ever remain, involved in much obscurity. <Astacus
crassicornis, Fabricius, 1775, Tyro cornigera, Milne-Edwards, 1840, and Zyro sarsl,
Bovallius, 1885, all with large upper antenne, may very likely be one and the same
species, which I am disposed to call Scina cornigera, rejecting the earlier crassecornis
on the ground of the too uncertain identity. ‘The Scina ensicorne, Prestandrea, 1833,
from the Mediterranean, may yet be identified, but the Clydonia gracilis and Clydonia
longipes of Dana are so figured that, in the absence of the type specimens and with
our present knowledge of the genus, I do not think they will ever be reconciled with
any actual species. Professor Chun recognizes that his Scina lepisma stands near to
the Mediterranean Scina marginata of Bovallius, but considers it distinct because the
upper antenne are shorter, with strikingly strong armature of filaments, and because
it has four pairs of branchial vesicles instead of six. There does not, however, appear
to be really any difference in the armature of the antenne, and the difference in length
is not by any means considerable. As for the branchial vesicles, in Scina marginata
these are said to be found on all the limbs of the peraon from the second to the seventh,
but in Scina lepisma only from the fourth to the seventh. ‘This would constitute a very
important distinction, but the fact stands in need of contirmation. These vesicles are
very easily detached and lost in the handling of a specimen, and there is a great
improbability that a species should be without them on the second and third pairs ot
limbs, and yet have them on the seventh pair. Thus the validity of the species /episma
is left in some doubt. Of his other species, Scina bovallii, Professor Chun says that it
has four pairs of branchial vesicles, “ between the third to the seventh pairs of trank-
limbs ;” and this he considers one of its chief distinctions from Scina borealis, Sars, and
Scina clausi (Bovallius), both of which have the nermal set of branchial vesicles
extending from the second gnathopods to the fourth peropods. Sars is rather
inclined to regard Scina clausi as a synonym of Scina borealis, but voth claust and
bovallii may be distinguished from Jdorealis by different proportions in some of the
limbs.

Arithmetic shows that a pack of fifty-two cards may be dealt out in a bewildering
number of ways. It may be noticed, therefore, that in the genus Scina the animals
have on the pereon seven pairs of limbs, each limb having six free joints, and
that they have also a pair of antenne consisting of peduncle and flagellum, and
three pairs of uropods, each uropod having one branch free from the peduncle.
Thus there are fifty-two pieces to be played with, each of which may be relatively
long or short, broad or narrow, simple or variously armed. Relatively also to each
other these eleven pairs of appendages may go through any number of variations

3E 2

REV. T. R R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

oo
or
1)

of size. Without taking into account other features, such as the eyes and the
telson, or the general shape and armature of the body, it may be left to the
arithmetician to calculate, if he can, how many species may be framed from the given
conditions. Some praise should be allowed to the moderation of nature and of
naturalists, in that, with such facilities at command, they have been contented as yet
with creating only twenty-one species in the genus, including in that number all the
doubtful names and seven new species instituted in the present paper. Probably not
more than fifteen of the twenty-one can be sustained. Of these there are some which
can only be distinguished from one another by close comparison of various details, but
Scina marginata (Bovallius) is at once marked out by having the apex of the hands in
the gnathopods produced. No other species, unless Scina lepisma be distinct from Scina
marginata, shares this peculiarity. Scina acanthodes, n. sp., is unique in the dentate
armature of the pereon and pleon. Scina stenopus, n. sp., is unique in the enormous
elongation of the peduncles of the uropods, only the otherwise very different Scina
acanthodes making any approach to it in this respect. Bovallius considers the Scina
cornigera of Milne-Edwards uniquely devoid of outer branches to the third uropods ; but
in that case I take it for granted that they were present though not observed, such an
oversight easily occurring when the more striking features of the animal were attracting
attention by their novelty. Scina wneipes, n. sp., is unique in the blunt-ended finger of
its fifth peraopods, though it agrees more or less with Scina marginata in the unwonted
thickness of those limbs. ‘The species at large may be roughly divided into two groups
—one in which, as in Scina cornigera, the first antenne are of very great length, the
other in which, as in Seina borealis, they are of much more moderate extent. In
the determination of species it is useful also to note whether the second joint of the
third persopods is dentate on both sides or only on one, and whether the finger of the
fifth pereeopods is hooked or simple. The serrature or denticulation of the margins of
the uropods varies in different species, but the details are often microscopic. The
number of species is at present rather surprisingly large compared with the number of
specimens known. ‘They have been instituted on the supposition that the proportional
sizes of the antenne, of the joints of the limbs, and of the uropods are fairly constant
for each species. Should this supposition prove unsound, a further revision would
doubtless be required. As the list at present stands, if it be right to cancel for different
reascns the names crassicornis, gracilis, longipes, sarst, lepisma, and to leave ensicorne
in suspense, the species remaining will be cornigera, Milne-Edwards, borealis, atlantica,
clausi, marginata, tullbergi, and pacifica of Bovallius, bovallii of Chun, and acanthodes,
stenopus, edicarpus, rattrayi, concors, similis, and uncipes of the present paper.

ScINA ACANTHODES, n. sp. (Plate LI.)

The head in front is deeply emarginate, forming two blunt lobes, behind which it is
dorsally traversed by a curved line. ‘The peraeon increases in width to the fourth

CRUSTACEANS FROM THE TROPICAL ATLANTIC.

(SX)
Or
(Se)

segment. The boundary between the first and second segments is rather indistinct.
All the segments, except the first, have on the hind margin a strongly projecting
median tooth. The pleon is much narrower than the perzon, but similarly armed
with a median tooth on each of the first three segments, of which the postero-lateral
angles appear to be acute. The fourth segment is short, with a thin semicircular
shield arising from near the front margin and covering the chief part of the segment.
The coalesced fifth and sixth segments are together not longer than the fourth. The
small telson is broader than long, distally truncate, not narrowed, with a setule on
either side of the centre of the distal margin; folded under is a thin curved lobe about
half the length of the telson. That the fold is natural and not accidental is evident
from the uninterrupted double marginal lines running round the sides and end of the
other portion. No eyes were perceived.

The first antenne are longer than the head and perzon together. The long first
joint of the flagellum has serrate edges, with numerous hyaline bacilli along the whole
length and spines at intervals; the terminal joint is quite small.

The second antenne (in the female) are short, obscurely three-jointed, the first joint
being a broad tubercle, the other two joints linear.

The mandibles are of the usual simple character, ending in narrow, finely denticulate
cutting plates. Of the other mouth-organs as much as could be made out is shown in
the figures; they are not suggestive of anything exceptional, apart from the figure @,
which does not agree with anything hitherto described for this genus. Whether it
may be a part of the maxillipeds I have not been able to determine.

The first gnathopods. The side-plates are bluntly pointed in front; the second joint
has the edges almost parallel, with minute spinules along the front one; the short third
joint has a terminal spine; the fourth joint is very little longer than the third; it has
spinules on the hind margin and two apical spines; the fifth joint is a little longer than
the sixth, the two together being longer than the second; each carries a single spine
and a pair of spines at intervals on the hind margin; the sixth joint is slightly curved ;
the finger is straight and slender, nearly half the length of the sixth joint.

The second gnathopods nearly resemble the first, but the side-plates are larger, the
second, fifth, and sixth joints longer, and the finger seemingly shorter. The branchial
vesicles are broader than the second joint, and more than two-thirds as long. ‘The
marsupial plates of the specimen are smaller than the branchial vesicles, and successively
smaller to the last, the fourth, pair. Of the five pairs of branchial vesicles the third is
the largest.

The first and second pereopods are much longer than the gnathopods, with similar
but larger side-plates. The second joint is not wider and not greatly longer than the
fifth; the fourth joint is rather longer than the sixth. There are no strong marginal
spines. The finger is minute, clasping between two sharp forward-directed teeth at
the apex of the hind margin of the sixth joint. In all the limbs the muscles are short,

oo4 REY, T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

leaving plenty of space for the glandular secretion, which probably finds an exit just
above the finger-tip.

Third perwopois. The side-plates are produced in a spine-like manner both forward
and backward, the hinder processes being very prominent features in the appearance of
the animal, extending back considerably beyond the sixth segment. ‘The second joint
has five spines on the hind margin, and three spines on the front margin, the apical
tooth of which reaches almost as far as the small apical tooth of the short third joint.
The fourth joint carries eleven or twelve spines on each margin, and is considerably
longer than the long second joint. ‘The remainder of the limb was missing.

The fourth and fifth pereeopods are alike. The side-plates of the fourth are the
larger, with a backward-directed point, which is evanescent in the next segment. The
joints are nearly as in the second pereopods, but the sixth joint is longer and without
the apical teeth, and the finger is more produced.

The pleopods carry two very small coupling spines, and have five or six joints to the
inner branch, and six or seven to the outer branch, the large first joint of the inner
branch carrying a long cleft spine.

The uropods are narrow and elongate, the first a little longer than the second, and
the second than the third; the first has a minute serration of the outer margin, and
numerous small spines on the lower half of the peduncle; its coalesced inner branch is
about two sevenths of its length; the outer branch is imperfect, less than half the
length of the inner; the second pair have a few spines on the outer and several on
the inner margin, the coalesced branch a little damaged, but probably lke that in the
preceding pair, the free outer branch longer than in the preceding pair and probably
more than half as long as the inner; the third pair have three spines on the outer
margin, the free outer branch shorter than the inner, but both damaged at the apex.

From the front of the head to the extremity of the uropods the specimen scarcely
measured one-tenth of an inch. It could not be persuaded to lie flat, though it would
readily stand on its head. Neither body nor appendages showed any inclination to
break, notwithstanding pressure repeatedly applied, so that the texture must be
tolerably tough. That the third pereeopods were already broken was no doubt due
to their exceptional length.

Habitat. Atlantic. Lat. 7° 54’ N., long. 17° 25' W. ‘Taken with the tow-net from
a depth of five fathoms, between 7.20 and 8.20 P.M.

The specific name, from the Greek axavdadnc, refers to the spiny armature of the
body, a feature so uncommon among the Hyperidea.

SCINA STENOPUS, n. sp. (Plate LII. A.)

The head is slightly emarginate between the antenne. ‘The first four segments of
the pergon were to a certain extent twisted and telescoped, so that their relative
lengths could not be accurately ascertained. The last three segments of the perzon

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 55:

are equal in length to the first three of the pleon. These latter have the postero-
lateral angles well rounded. The next three segments of the pleon are short, together
scarcely longer than the third segment. ‘The fifth and sixth segments are apparently
coalesced, and the telson seen in profile seems to be narrow, about equal in length to
the sixth segment.

The eyes are small, consisting of about a dozen ocelli; the colour in spirit very pale.

The first antenne are longer than the whole body from head to telson, with the first
joint or peduncle stout, not very much longer than broad. In the elongate, straight,
tapering flagellum no division could be discerned even near the apex. ‘The inner
margin is fringed with numerous filaments, which when highly magnified are seen to
be broad and round-ended. The outer margin carries many spinules.

The second antenne (of the female) are minute, planted near the hind margin of the
head, and only extending forward as far as the eyes. A trace of division into two or
three joints could rather be imagined than seen.

The mouth-organs form a small, nearly circular, group, the pointed apex of the
epistome scarcely reaching halfway from the back to the front of the head.

First gnathopods. The side-plates of these and the following limbs are small and but
faintly distinguished. All the limbs of the pereon are exceedingly narrow, and the first
gnathopods are the shortest. Of its joints the second is the longest, the third the
shortest. The fourth, which is not twice as long as its breadth, is distinguished by an
armature of prickles. The fifth is longer than the sixth; they carry a few setules, and
are together longer than the second joint; the sixth has a small tooth at the apex.
The finger is very slender, a little over half the length of the preceding joint.

Second gnathopods. The branchial vesicles are small. The limbs are not very
different from the preceding pair, but with the second, fifth, and sixth joints longer, and
the fourth not prickly.

First and second perwopods. 'The branchial vesicles are slightly larger than those of
the preceding, and smaller than those of the following limbs, those of the large third
pereopods being the largest. These slender and unarmed limbs have all the joints,
except the third and seventh, longer than in the gnathopods, and the fourth joint
longer than the fifth. The finger is small.

Third pereopods. The elongate second joint is less than half the total length of the
limb, armed with spines on both margins, those on the front being the larger, and this
margin ending in a tooth which projects beyond the similar but smaller tooth of the
short third joint. The fourth joint is considerably longer than the fifth, the two
together being a little shorter than the second. The sixth is considerably shorter than
the fifth, and the finger is minute.

Fourth perwopods. The branchial vesicles, though smaller than the preceding pair,
are larger than any of the others. The limbs are unarmed, more slender than the
preceding pair and shorter, but not very greatly so. The proportions of the joints are

(Se)

56 REV. T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

not very dissimilar, but the fourth and fifth together are a little longer than the second.
The length of the limb is equal to that of the first five joints of the third pereopods.
The sixth joint is longer than in that pair.

Fifth perwopods. The slender, unarmed limbs are about two-thirds the length of the
fourth pereeopods, which they much resemble in the proportions of the joints. The
finger is minute, apparently hooked.

Pleopods. Vhe peduncles are strong. The two coupling-hooks are minute, with
three pairs of backward-directed points to each. The slender rami have seven or eight
joints with the usual seta, and the cleft spine on the first joint of the inner branch.

Uropods. These are subequal to one another and half the length of the first antenne.
No outer ramus could be distinguished in the first pair; the second have a very small
one, little more than a quarter the length of the coalesced inner branch, which is itself
not quite a fifth of the total length of the uropod. The third pair are slightly the
shortest. The outer branch is fully two-thirds the length of the inner, which is a fifth
of the total length of the uropod. ‘The first and third have spines on both margins,
the second appeared to have them only on the outer margin. The outer branch ot the
third pair has spines on the inner margin.

The length is half an inch, the first antenne measuring one-fifth of an inch, the body
nearly the same, and the uropods a tenth.

The specific name, from the Greek orevérove, meaning narrow-footed or narrow-
legged, speaks for itself.

Habitat. Atlantic. Lat. 7° 1’ 1" N., long. 15° 54’ W. ‘Taken in the daytime from
a depth of 100 fathoms.

Of species hitherto described not one makes any very near approach to the present,

unless possibly Dana’s Clydonia longipes from the Pacific, but even in that the uropods
are entirely different.

SCINA GDICARPUS, n. sp. (Plate LII. B.)

The head has the front shallowly emarginate. The pereon is dilated, with the last
two segments narrowing rather abruptly. The pleon is narrow, and has the fifth and
sixth segments coalesced. The telson is small; owing to its transparence its outline
could not be clearly discerned.

The eyes are small, and pale in spirit.

The first antenne are equal in length to the person and first three segments of the
pleon. They have the usual stout one-jointed peduncle and two-jointed flagellum, the
long first joint carrying minute spines on the outer margin and long filaments on
the inner.

The second antennee are folded in a curve across the underside of the head above the
mouth-organs. ‘The first three joints are short, the basal one partially soldered to the
head; the fourth joint is not twice as long as the third; the fifth joint is rather longer

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 357

than the preceding four together, and represents the flagelium in an early stage. The
condition of these antenne indicates that the specimen is a male not fully adult.

The first gnathopods have the fifth joint rather longer than the sixth. As in the
other limbs there is scarcely any armature, but the short fourth joint has a few spinules
and the two following joints a few slender sete.

The second gnathopods have the second joint rather longer than that in the first pair,
the fourth joint smooth, the fifth shorter than the sixth.

The first pereeopods have the fourth joint much longer than that of the gnathopods,
but much shorter than the fifth joint, which is dilated and almost entirely glandular ;
the following joint is slender.

The second perxopods scarcely differ from the first, except in having the fifth joint
not dilated.

The third pereopods have the second joint longer than all the rest of the limb, with
spine-like serrations on both margins, those on the front the stronger. The apical
tooth is produced much beyond the short third joint. The fourth joint is decidedly
longer than the fifth, which is not quite three times as long as the sixth. The finger
is minute and bent.

The fourth pereopods have the second joint little more than half the length of that
in the preceding pair; the fourth joint longer than the fifth, and these two together a
little longer than the second. The sixth joint is half as long as the fifth. The finger
is very small, strongly curved.

The fifth perseopods are slender, about half as long as the fourth, and a third as long
as the third pair. The fourth, fifth, and sixth joints are subequal one to another, and
two of them together are subequal to the second joint. The finger is minute, seemingly
retractile.

The first uropods are rather longer, while the other two pairs are rather shorter, than
half the upper antenne. In all, the peduncles are elongate, but shorter than the
coalesced inner ramus. The outer ramus in the first two pairs is like a small spine; in
the third pair it is slender, but rather longer than half the inner ramus. The marginal
armature in all three pairs is extremely minute, except for a single spine or process on
the inner margin of the first pair a little higher up than the outer ramus.

The total length a little exceeds a fifth of an inch, the measurement without the
antenne and the uropods being just one tenth of an inch.

Habitat. Atlantic. Lat. 7° 1! 1" N., long. 15° 54’ W. Taken in the daytime from a
depth of 100 fathoms.

Branchial vesicles were present on the first, second, third, and fourth pereopods,
but not on the second gnathopods. No stress, however, can be laid on this deficiency,
since on one side of the specimen the vesicles were absent also from the first and
second pereopods.

From Scina atlantica (Bovallius), to which the present species makes the nearest
VOL. XIII.—PART X. No. 2.— February, 1895. 35

358 REV. T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

approach, as well as from all other known species of the genus, it is distinguished by
the proportions of the third peropods, which are unique in having the second joint
longer than all the remainder of the limb. ‘The specific name, from the Greek owdeiv, to
swell, and capréc, wrist, refers to the wrist or fifth joint of the first perseopods, which is
more tumid than usual, though in this genus it is normally glandular, Dr. Bovallius has
noticed that in species of Rhabdosoma the females sometimes have the fourth and fifth
joints of the first and three following perzeopods “inflated and almost egg-shaped,
owing to a strongly developed glandular mass surrounding the axis of the joint for the
whole of its length.” He supposes the development of these powerful glands to be
periodical, and “to have some connection with the fixation of the eggs on the underside
of the body.” He had seen full-grown females of some species without the joints
inflated at all, but had never seen females of the same species with eggs or young ones
which had not at the same time those joints more or less inflated. He therefore
deems it “‘ probable that the development of these glands may be connected with the
maternal functions of the animal.” In the present species of Scina this glandular
development occurs, it will be observed, in a male specimen.

ScINA RATTRAYI, n. sp. (Plate LILI. A.)

The head is broad, dorsally smooth, with the front margin between the antenne
faintly concave or perhaps straight. The segments of the perzeon are broad, except
the last two. ‘The first four segments of the pleon are rather long, each subequal in
length to the coalesced fifth and sixth segments, of which the part furnished by the
fifth is abruptly narrower than the fifth. The telson is small and tongue-shaped. The
length of the body from the front of the head to the end of the perzon is the same as
that of the outstretched pleon to the end of the telson.

The eyes are minute, situated at the anterior corners of the head.

The first antennz are nearly equal in length to the head and pleon. ‘They are set
wide apart. The first joint of the flagellum carries numerous filaments; the small
and slender second joint is tipped with a long fine seta.

The second antenne of the specimen are short, two-jointed.

The first gnathopods have the second joint equal in length to the three following
together, and broader than the corresponding joint in any of the other limbs. The
fourth joint has some minute prickles on the hind margin, and a long spine and a short
one at its apex; the fifth is considerably longer and broader than the sixth, and is
armed with numerous sete or seta-like spines on the distal half of the front margin
and along the hind margin, which carries a rigid spine near its apex; the sixth joint is
straight and narrow, with flexible spines on both margins and on the surface. The
needle-like finger is about half the length of the preceding joint.

The second gnathopods have a few spines but no prickles on the hind margin of
the fourth joint, the fifth joint considerably shorter and very little broader than

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 309

the sixth; both these joints being armed with spines, but less densely than in the
first pair. The finger is slightly curved, and is much less than half the length of the
preceding joint.

The first peraopods have the second joint equal in length to the fourth and fifth
together, the fourth rather more than half the length of the fifth. The latter, which
is as usual specially glandular, is considerably broader and a little longer than the
sixth joint. The finger is minute.

The second pereopods scarcely differ from the first.

The third pereeopods have the second joint equal in length to the first antennz minus
the terminal joint. The front margin is produced into a short tooth, in advance of
which are two spine-like processes; on the hind margin there are twelve of these
processes. The third joint does not equal the breadth of the second ; the fourth is
slender, but in length remarkable, falling not far short of the second; the fifth is
about two-fifths of the length of the fourth; the sixth less than half the length of the
fifth. The finger minute.

The fourth pereopods are slender and very long, though shorter than the fifth ; the
second and fourth joints being much shorter than in the preceding pair, while the
fifth joint is fully as long, and the sixth between two and three times as long as in
that pair.

The fifth perzeopods are very slender and short, the total length scarcely equalling
that of the second joint in the preceding pair. ‘The minute finger has a bulbous base
and a slender hooked termination.

Branchial vesicles are attached to the first four pairs of pereopods, and apparently
also to the second gnathopods. Some of the vesicles exhibit a rather unusual appear-
ance, the centre seeming to be occupied by a series of little globules.

The pleopods have nine joints to the outer ramus and seven to the inner.

The uropods all have the peduncle considerably longer than the coalesced inner
branch. The first pair are the longest, but do not reach quite so far back as the
third; they have three spines spaced on the inner margin of the rather broad peduncle:
the outer branch is represented by a small spine; the inner is finely serrate on the
outer margin. The second pair have a similar outer branch, and the inner branch
slightly serrate on the inner margin, and reaching back as far as or a little beyond that
of the first pair. The third pair have two spines on the inner margin of the peduncle,
the outer margin of the inner ramus serrate with six or seven tiny spinules; the outer
ramus about four-fifths of the length of the inner, microscopically serrate, and carrying
minute spinules on its inner margin.

The length of the specimen, including antenne and uropods, is a quarter of
an inch.

Habitat. Atlantic. Lat. 1° 56’ 5" N., long. 5° 55’ 5" BE. Taken after 9 p.m. from a
depth of 360 fathoms.

360 REY. T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

The specific name is given in compliment to John Rattray, Esq., under whose
supervision the present collection of Amphipoda was made.

The only species which make any approach to the peculiar character of the third
pereeopods here exhibited are Scina tullbergi (Bovallius) and Seina pacifica (Bovallius),
the latter of which may, according to Dr. Bovallius himself, be only a variety of the
former. They do not, however, show so great an elongation of the fourth joint in the
limbs in question, and they have very different proportions in the fourth pereopods,
and exhibit many differences in the uropods. Of the pleopods in the two species above
named Bovallius writes that the exterior ramus has seven joints, the interior nine joints.
He also assigns fewer joints to the outer than to the inner ramus in the pleopoda of
Scina sarsit, Scina atlantica, and Scina marginata, giving more to the outer than to the
inner only in his species Scina clausi. In all the species of Scina which I have
examined the pleopods have uniformly had the smaller number of joints on the inner,
not on the outer ramus. It may be useful to mention that when the pleopod is
detached from the animal the presence of the cleft spine on the inner branch distin-
guishes it with certainty from the outer.

Scrva concors, n. sp. (Plate LIIT. B.)

The head is dorsally smooth, truncate between the antenne, this part of the margin
not at all concave, but by the rounding of its corners tending rather to become convex.
The peron is broadly ovate, together with the head equalling in length the pleon to
the end of the coalesced segments. ‘These latter, with the fourth pleon-segment, are
together scarcely equal to half the length of the first three segments of the pleon. The
fifth segment is much broader but not longer than the sixth, which is coalesced with
it. The small telson is broader than its length, with the apex truncate, as in Scina
acanthodes. ‘The shape and proportions are unusual in this genus, but there is nothing
in either case to indicate that the telson is either broken or abnormal.

The eyes are comparatively large, composed of some nine or more pairs of cones
arranged in loose order.

The first antenne are rather shorter than the pereon, with the first joint of the
flagellum broad, four-sided, armed below with fourteen teeth, and on the outer margin
with ten, the inner densely fringed with cylinders. The short second joint has four
cilia on the inner margin, and is tipped with a fine seta not so long as itself.

The second antenne are set well behind the first on the underside of the head, and
have behind them a prominence, apart from which the peduncle is four-jointed, with
the third and fourth joints longer than the first and second. The flagellum is of great
length, though consisting, perhaps, of not more than eight or nine joints. ‘The first of
these is nearly as long as the peduncle, or not quite half the length of the upper
antenne; it narrows from a broad base, and is again a little enlarged apically; its
upper or inner margin is fringed with decurrent cilia. The remaining joints are also

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 361

more or less ciliated, filiform, some longer, some shorter than the first, but the
boundaries not always easy to detect, unless where the apical dilatation is seen
broadside on.

The mouth-organs, as seen without dissection, show a broad and deep helmet-shaped
epistome, rather small mandibles with finely-toothed cutting-edges, the plates of the
maxillipeds elongate, with the apices not acute but rounded.

The first gnathopods have a short spine at the apex of the third joint, prickles along
the hind margin of the fourth joint, and two spines at the apex; the fifth joint longer
and much broader than the sixth, beset with several slender spines; the sixth joint
narrow, especially in the distal half, and beset with many spines; the finger slender,
nearly straight, about half the length of the sixth joint.

The second gnathopods have both the fifth and sixth joints slender, the sixth
the longer.

-The first and second pereopods have the sixth joint much shorter than the fifth,
subequal in length to the fourth, but much narrower, with three minute spinules
indenting the distal half of the hind margin. ‘The finger is curved, much shorter than
in the gnathopods.

The third perzeopods have the long second joint much broader than it is in any of
the other pairs, with its hind margin cut into about fourteen teeth, the front having
only the apical one, and that small; the fourth joint is about three-quarters the length
of the second, the fifth two-thirds the length of the fourth, and the very slender sixth
a little shorter than the fifth. The finger is minute.

The fourth perzeopods have the second joint three-quarters the length of that in the
preceding pair; the fourth, fifth, and sixth joints each subequal to the fifth in the
preceding pair and to each other, but the fifth slightly the longest of the three. The
finger is very small, yet longer than in the other pereopods.

The fifth pereeopods are small, but not strikingly slender. The second joint is
more than half the length of that in the fourth pair; the fourth and fifth joints
are subequal, together longer than the second; the sixth joint is much shorter and
narrower than the fifth. The finger is small, consisting of a bulbous base and a short
spine-like tip.

The branchial vesicles are rather large, showing the cell-structure rather con-
spicuously.

The pleopods have strong peduncles. ‘The branches are shorter, with ten joints to
the outer, and nine to the inner.

Lhe uropods have the peduncles longer than the coalesced inner ramus, considerably
in the second pair, less so in the first, and very slightly in the third. ‘he first pair
have the inner ramus serrate along much of the outer margin, the outer ramus
resembling a considerable spine, equal to a fourth of the length of the inner; the
second pair are serrate on the inner margin from a little above the commencement of

062 REV. T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

the ramus, the outer ramus spine-like, but longer than in the first pair, and equal to
two-fifths of the inner ramus; the third pair are serrate on the outer margin of the
inner, and the inner of the outer ramus, the latter being two-thirds as long as the
former.

The specimen measures a little more than a fifth of an inch, by inclusion of the first
antenn and the uropods.

Habitat. Atlantic. Lat. 4° 26'7" S., long. 10° 1’ 8" E. Taken in the daytime from
a depth of 135 fathoms.

From the closely allied species, Scina tullbergi and Seina pacifica of Bovallius, the
present is distinguished by shorter fingers to the enathopods and the first four pairs of
pereopods, by a longer sixth joint in the third pair, and especially by differences in the
uropods, the outer branch being here longer in the second than in the third pair instead
of vice versd, and the outer branch of the third pair being lanceolate and serrate instead
of linear and smooth, and also much longer than in the other species. The difference
in the shape of the telson might be considered to outweigh all the other distinctions,
but its very peculiarity arouses some suspicion that the shape may be accidental. ‘The
specific name refers to the near agreement of this with the species reported from Cape
Horn and from the Pacific, at Corinto, Nicaragua.

ScINA SIMILIS, n. sp. (Plate LIV. A.)

The head is dorsally smooth, truncate or even slightly convex between the antenne.
The length of the head and perzon equals that of the pleon to the tip of the telson.
The fifth and sixth segments of the pleon are coalesced. The telson is narrowly
triangular, about a third of the length of the peduncle of the third uropods. The
ovaries extend backwards to the end of the sixth pereeon-segment.

The eyes appear to be composed of nine ocelli, each ocellus consisting of four cones.
In the present condition of the specimen these four cones seem to be encircled by a
ring, and are separated one from another by a cross-shaped interval.

The first antenne are equal in length to the last five segments of the pereon. ‘The
first joint of the flagellum has seven spiniform processes along the outer margin and a
larger number on the inner, over which the long and numerous filaments project. The
slender second joint is tipped with a fine seta.

The gnathopods of both pairs agree closely with those of Scina concors, except that
the fingers are proportionally a little longer.

The first and second pereopods have the sixth ost longer than the fourth, and
nearly as long as the fifth. The finger is slender, curved, not a quarter of the length
of the hand.

The third pereeopods have thirteen teeth on the hind margin of the second joint, its
front margin smooth, with an apical tooth a little longer than the third joint. The
fourth joint is only a little shorter than the second apart from its apical tooth; the fifth

Qo

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 36%

isa little more than half the length of the fourth; the sixth is about three-quarters the
length of the fifth. The finger is small and slender.

The fourth pereopods have the fourth and sixth joints subequal to the fifth in the
preceding pair, but the fifth joint shorter than either of these and about half the length
of the second joint. Both the sixth joint and the finger are considerably longer than the
corresponding parts in the preceding pair.

The fifth pereeopods are slender, scarcely equal in length to the second joint in the
third pair. The sixth joint is slightly shorter than the fourth or fifth. The finger is
very slender, with only a small bulb at the base.

The first uropods have the peduncle equal in length to the coalesced inner branch,
the outer margin of the peduncle more faintly serrate than that of the branch, the
inner margin of both smooth; the outer branch is spine-like. The second uropods are
more slender than the first, with peduncle and inner branch subequal, serrate only
along the inner margin, and having a spine-like outer branch, which is a little thinner
and a very little longer than that of the first pair. The third pair have the peduncles
a little longer than the outer branch, but considerably shorter than the coalesced inner
branch; the outer is two-thirds of the length of the inner branch; the outer margin
of the inner and the inner of the outer branch are serrate.

The specimen, a female with setiferous marsupial plates, measures three-twentieths
of an inch, antenne and uropods included.

Habitat. Atlantic. Lat. 3° 0! 8" N., long. 7° 43’ W. Taken at noon from a depth
of 50 fathoms.

The specific name refers to the general likeness which this species presents to the
species tullbergi, pacifica, and concors. Yet it differs from all three in the relative
proportions of the joints of the fourth perzeopods, and by the different character of the
fifth perzeopods, especially in regard to the finger; moreover, the proportions of the
uropods are different, not to mention other marks of distinction which separate it from
one or other of the species in question.

SCINA UNCIPES, n. sp. (Plate LIV. B.)

The outline of the head was probably broadly truncate between the antenne, but
owing to an accidental damaging of the specimen this cannot be affirmed with certainty.
The telson is triangular, scarcely a third of the length of the peduncles of the third
uropods.

The eyes resemble those of Scina similis.

The first antenne are stout, equal in length to the hinder half of the pleon, from the
base of the fourth segment to the extremity of the uropods. The first joint of the
flagellum has twelve teeth on the outer margin and about sixteen on the inner, over
which project some setules, and a rather slender supply of filaments. The second joint
is only a small stump, without any sete ; but this is probably not its normal condition.

564 REV. T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

The second antenne have at the base two short stout joints. The third is longer
and more slender than either of these, and the fourth than the third, with a slight
dilatation near the middle. The fifth joint is more slender than the fourth, but quite
as long. ‘The four remaining joints are together somewhat longer than the preceding
joint. This form is no doubt a stage in the development of the antenne of the male.

The gnathopods differ little from those of Scina similis. ‘There are branchial vesicles
attached to the second gnathopods and to the first four pairs of pereeopods.

In the first and second pereopods the fifth joint is longer than the fourth or the
sixth. The finger is much curved.

The third pereeopods have twelve teeth on the hind margin and about the same
number on the front, together with an apical tooth longer than the third joint. The
fourth and fifth joints are subequal in length; together they are as long as the second
joint apart from its apical tooth. The sixth joint is rather shorter than the fifth. The
finger is minute, triangular, its apex forming a nail.

The fourth pereopods are shorter than the first or second. They have the fourth
and fifth joints equal, the sixth shorter than the fifth and a little shorter than the
corresponding joint in the preceding pair, while the finger is similar to that of the
third pair, but rather stouter.

The fifth peraopods are robust, with the fourth joint longer than the fifth, and the
fifth than the sixth. ‘The finger is curiously shaped, thick at the base, then narrowing,
and again widening, the widened extremity being closely beset by a pair of spines or
sort of double nail.

The. pleopods have nine joints to the outer and seven to the inner branch.

The first uropods reach beyond the second, nearly to the extremity of the third.
They have the peduncle rather shorter than the inner branch, of which the outer margin
is minutely serrate; there are teeth sparsely set on the inner margin of both peduncle and
branch. ‘The outer ramus is a small thick spine. ‘The second pair reach nearly as far
as the outer branch of the third pair; they have the peduncle rather shorter than the
inner branch, which is finely serrate on its inner margin; the outer branch is a little
longer than in the first pair. The third pair have the peduncle nearly as long as the
outer branch, which is fully four-fifths as long as the inner; the confronting margins
of the two branches are serrate.

The specimen measures three-tenths of an inch.

Habitat. Atlantic. Lat. 7° 54’ N., long. 17° 25’ W. Taken between 6 and 7 P.M.
from a depth of 50 fathoms.

The specific name refers to the peculiarly hooked finger of the fifth persopods.
From the more or less nearly related species, Scina marginata (Bovallius) and Scina
lepisma, Chun, it is separated not only by the finger in question, but by the hands of
the gnathopods, which are not apically produced into a sharp process.

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 365

Scina CoRNIGERA (Milne-Edwards).
1830. Hyperia cornigera, Milne-Edwards, Ann. des Sci. Nat. t. xx. p. 387.
1840. Tyro cornigera, Milne-Kdwards, Hist. Nat. des Crustacés, t. iii. p. 80.
1885. ,, sarsii, Bovallius, Some forgotten Genera of Amphipoda, p. 15, figs. 3, 3 a.
1887. ,, sarsi, Bovallius, Contributions to a Monograph of the Amph. Hyperiidea, p. 9, pl. 1.
figs. 1-17, pl. 2. figs. 1-10.
1888. Scina cornigera, Stebbing, ‘ Challenger’? Amphipoda, p. 1273, pl. 146.

The identity of Zyro sarsit, Bovallius, with Tyro cornigera, Milne-Edwards, cannot
be insisted on as more than probable. If the later name is upheld, the earlier one
might as well be cancelled.

A specimen more than half an inch in length was obtained by the ‘ Buccaneer’ near
the surface after dark, in the Atlantic, at lat. 5° 88’ N., long. 14° 20’ W., together with
two other specimens of the same species.

Scina ATLANTICA (Bovallius).

1885. Tyro atlantica, Bovallius, Some forgotten Genera of the Amphipoda, p. 14.

Wetevio 5 5 Bovallius, Contributions to a Monograph of the Amph. Hyperiidea, p. 13,
pl. 2. figs. 11-18.

1888. Scina atlantica, Stebbing, ‘ Challenger ? Amphipoda, p. 558.

1889. _,, 33 Chun, Zool. Anzeiger, Jahrg. xi. no. 308, p. 289.

A specimen, apparently belonging to this species, was taken in the daytime from a
depth of 100 fathoms, in the Atlantic, at lat. 7° 11! N., long. 15° 54’! W. It measured
fully a fifth of an inch. The first antenne are unusually wide apart. The third
pereopods in this specimen are unsymmetrical, so that a distinct species might be
made out of each longitudinal half of the animal. The pereopod on the right side
has the full number of joints, but all, except the third, are dwarfed. The preceding
limbs are affected in a similar manner, but to a smaller extent.

Scina PactFica (Bovallius).
1887. Tyro pacifica, Bovallius, Systematical List of the Amph. Hyperiidea, p. 4.
WetV/s pg a Bovallius, Contributions to a Monograph of the Amph. Hyperiidea, p. 25,
pl. 3. figs. 10-17.
1888. Scina pacifica, Stebbing, ‘ Challenger’ Amphipoda, p. 587.

The differences between this species and the earlier Tyro tullbergii, Bovallius, 1885,
are, as Dr. Bovallius himself remarks, very slight. A specimen, agreeing best with the
atlantica form, was taken by the ‘ Buccaneer’ with the tow-net at noon from a depth
of 235 fathoms in the Atlantic, in lat. 4° 26’ 7’ S., long. 10° 18’ EK. The specimen
measured three-twentieths of an inch, including as usual the antenne and uropods. It
contained about a dozen relatively large eggs.

It seems not a little remarkable that the only twelve specimens of the genus Scina
VOL. XIII.—PaRT x. No. 3.—February, 1895. 34

366 REV. T.'R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

which I have been able to discover in the collection should include ten distinct species,
of which no less than seven are new. One may without rashness forecast that, if
species continue to multiply at this rapid rate, before long it will be thought necessary
to relegate such forms as acanthodes, stenopus, and uncipes to so many separate genera.

Family RHABDOSOMID.

In his admirable work on the Oxycephalids, published in 1890, Dr. Bovallius
separated from the Oxycephalide a new family which he named Xiphocephalide,
distinguished by having the eyes planted between an outdrawn neck-like portion of
the head and a long needle-shaped rostrum, by having the telson articulated to the
preceding segment, by having the fifth pair of pereopods reduced to a single bladder-
like joint, and by the absence of marsupial plates from the ovigerous female.
Dr. Bovallius recognized that the ‘Challenger’ species Rhabdosoma brevicaudatum,
as described, formed an exception, by having its telson coalesced with the preceding
segment, and that his own genus Twlbergelia, with the fifth pereeopods four-jointed,
was in that respect intermediate between the other Oxycephalide, which have those
limbs fully jointed, and the Xiphocephalidz, in which he supposed them to be limited
to a single joint. After the publication of his book he came to England and inspected
the type specimen of Rhabdosoma brevicaudatum at the British Museum. He persuaded
himself that the telson was in fact articulated; but from a re-examination of the
specimen I am myself convinced that the original description was correct, and that any
appearance to the contrary is due to the transparency of the animal, making it some-
what difficult to perceive that the ventral suture is not dorsal. ‘There is now to be
described another short-tailed Rhabdosoma, in which likewise the telson is coalesced
with the preceding segment. No stress, therefore, can properly be laid on the
articulation of the telson as a family characteristic. In yet another new species of
Rhabdosoma the fifth pereeopods prove to be three-jointed, thus making a close
approach to what is found in the acknowledged Oxycephalid genus 7ulbergelia. Even
the neck-like constriction and the needle-shaped rostrum of the Rhabdosomid head are
met with, though in far less exaggerated form, among the Oxycephalide, so that it
seems inconsistent on the part of Dr. Bovallius to rely on these as points of distinction
between the two families, while, on the other hand, he introduces into the Oxycepha-
lide the very different-looking and blunt-snouted Simorhyncotus. The effect of the
latter innovation is at least very awkward for the received nomenclature, since the
family name, which signifies that the beak is sharp, is made to cover a genus the name
of which signifies that the beak is blunt. There still remains for the Xiphocephalide
the substantial distinction that the female is unprovided with marsupial lamelle, and
carries her eggs in the singular manner which Dr. Bovallius has described. The reason
for changing the name Xiphocephalide into Rhabdosomidz will be apparent from the

CRUSTACEANS FROM THE TROPICAL ATLANTIC. 367

discussion of the synonymy of the solitary genus by which the family is at present
represented. ‘The change of name should not deter the student from consulting the
exceedingly important morphological notes with which Dr. Bovallius has enriched his
elaborate account of this family.

Genus RHapposoma, Adams and White, 1847.
1847. Rhubdosoma, Adams & White, in White’s List of Crustacea in the British Museum, p. 138.
1848. _ Adams & White, Zoology of the Voyage of H.M.S. ‘ Samarang,’ p. 63.
1858. Macrocephalus, Spence Bate, Annals and Magazine of Natural History, ser. 3, vol. i. p. 361.
1862. Rhabdosoma, Spence Bate, British Museum Catalogue of Amphipodous Crustacea, p. 344.
1887. Rhabdonectes, Bovallius, Systematical List of the Amphipoda Hyperiidea, Bihang till Kongl.
Svenska Vet.-Akad. Handlingar, Bd. xi. no. 16, p. 39.
1888. Rhabdosoma, Stebbing, ‘ Challenger’ Amphipoda, p. 1606.
1890. Xiphocephalus, Bovallius, The Oxycephalids, Royal Society of Sciences of Upsala, pp.3, 116, &e.

For other references the two last-named works may be consulted. Dr. Bovallius
assigns the genus, with the name Xtphocephalus and the date 1841, to the well-known
zoologist Guérin-Méneville, on the authority of Kydoux and Souleyet’. He quotes
the passage in point from those authors, which is to the following effect :—“ M. Guérin-
Méneville, who has been kind enough to study this species with us, thinks that it will
have to be separated from the genus Oxycéphale to form a new generic division to
which might be given the name of Xyphicéphale, which expresses its principal
character; he bases his view on the fact that the true species of Oxycéphale have the
body shorter, of different shape, and on the fact that they have seven pairs of feet, of
which two pairs are didactyle and five ambulatory.”

It will be observed that here the genus is not instituted, but only an opinion given
that it will eventually have to be, and the reasons for that opinion are stated. A
suitable name for the genus thus foreshadowed is indicated by the French word
Xyphicéphale. It is, or was, a common custom with French authors to give zoological
names both in French and Latin. The Latin form is in this case not added, very likely
because the authors intended to leave to Guérin himself the privilege of technically
naming the genus, when a proper occasion should present itself. Of this privilege he
appears to have never availed himself. The short Latin account of the species under
discussion actually calls it ‘ Oxycephalus, corpore perangusto, elongatissimo,” &c.
Under these circumstances I cannot bring myself to believe that the French form
Xyphicéphale has any claim to be accepted as an authoritative name, and, since it is
a monstrosity in spelling, its rejection should inspire but little regret. Three changes
in it have to be made in order to polish it into Xiphocephalus, a name which Guérin-
Méneville might have given, but did not. By discarding Xyphicéphale we are brought

«Voyage autour du monde exécuté pendant les années 1836 et 1837 sur la corvette la ‘ Bonite,’”
Zoologie. Tome 1", pp. 267-271, pl. iv. figs. 18-32. Paris (1841 ?).

362

368 REV. T. R. R. STEBBING ON NEW SPECIES OF AMPHIPODOUS

to Rhabdosoma as the earliest name of the genus, and thus in a manner forced to
change the name of the family from Xiphocephalide to Rhabdosomide.

Dr. Bovallius has very acutely unravelled the synonymy, and explained the distinctions
between the species hitherto included in the genus Rhabdosoma. These are Khabdosoma
armatum (Milne-Edwards), 1840, Rhabdosoma whitei, Spence Bate, 1862 (the Xyphi-
céphale of Eydoux and Souleyet), Rhabdosoma lilljeborgt (Bovallius), 1890, - and
Rhabdosoma brevicaudatum, of the ‘Challenger’ Report, 1888. ‘Two new species,
kt. piratum and R. brachyteles, are now instituted.

RHABDOSOMA PIRATUM, n. sp. (Plate LV. A.)

‘The head is longer than the perzon and the first three segments of the pleon, as long
as the whole pleon to the extremity of the uropods. ‘The slender rostral part is nearly
double as long as the part behind the first antenne. Of the person-segments the
fourth, fifth, and sixth are the longest. The first three segments of the pleon have the
postero-lateral angles acute. The third is a little longer than either of the first two
or the fourth, but a little shorter than that which is formed by coalescence of the fifth
and sixth. The narrow tapering telson is considerably longer than any of the preceding
segments, and reaches a little beyond the first uropods.

The eyes are of the character usual in the genus.

The first antenne are very small, two-jointed, tipped with four short filaments.

No second antenne were apparent in the specimen.

The first gnathopods are as usual minute, the second joint a little sinuous, the
fourth scarcely larger than the third, the fifth sharply produced behind as far as the
minutely toothed apex of the hand, which is narrower than the basal part of the wrist,
but about as long. The finger is slender, very slightly curved, more than half the
length of the hand.

The second gnathopods differ little from the first, but have the process of the fifth
joint a little shorter, and the apical margin of the hand narrower.

First perwopods. ‘The second joint is rather wider but not longer than the fourth ;
the fifth is narrower and shorter than the fourth, to which the slightly curved sixth is
subequal in length though not in breadth. ‘The finger is slender, curved, half as long
as the preceding joint.

The second pereopods scarcely differ from the first, except in slightly greater length
of some of the joints.

Third pereopods. ‘The branchial vesicles are oval, more than half as long as the
second joint of the limb and much wider than it. No trace of branchial vesicles
could be perceived on any of the preceding limbs. In slenderness and general structure
the third are very like the first and second pereopods, but they have the second, fourth,
and sixth joints much longer, and the second decidedly longer than the fourth.

Fourth pereopods. ‘The branchial vesicles are a little longer than the preceding pair.

©RUSTACEANS FROM THE TROPICAL ATLANTIC. 369

The limbs differ very little from the third perzeopods, unless the second joint be a little
longer and the sixth a little shorter than in that pair.

Fifth perwopods. These are minute, with a shortly pear-shaped dilated portion that
may be taken to represent the second joint, which is immediately foilowed by a much
shorter and very much narrower piece carrying at its truncate apex the tiny finger,
broad at the base, and slender at the strongly bent tip.

The pleopods are not very robust, with very small coupling-spines. The outer
branch has four and the inner three or four joints.

The first uropods have the peduncle about four times as long as the branches, with
sixty or more spinules along the inner margin and many along the outer one. The
branches do not reach quite to the extremity of the telson: the outer is bordered
with nine long spine-like teeth on each side; the inner, which is slightly the shorter,
has eighteen teeth on the inner margin, small near the base and larger behind; the
outer margin has eight teeth, and its upper part unarmed. The second pair have the
peduncle subequal in length to the two coalesced segments of the pleon, the outer
branch a little more than half as long as the inner, carrying about six teeth on each
margin: the inner branch has six long teeth and one or two very small ones on each
margin ; it is coalesced with the peduncle, which has about fifteen small teeth on the
inner margin, and apparently a smaller number on the outer. ‘The third pair have the
peduncles intermediate in size between those of the first and second pairs, subequal in
length to the telson, not twice as long as the coalesced inner ramus, with many teeth
on two margins, the upper ones small and distant. The outer ramus is not half as
long as the inner; it carries six long teeth on the inner margin, and two or more on
the outer. The coalesced inner ramus has about fourteen teeth on each margin.

The specimen was.a little less than half an inch long.

Habitat. Atlantic. Lat. 0° 38' 6” N., long. 6° 25' 8” E., off St. Thomas’s Island.
Taken with the tow-net at noon from a depth of 10 fathoms.

The specific name is given in remembrance of the vessel by aid of which the present
collection of Amphipoda was made.

RHABDOSOMA BRACHYTELES, n. sp. (Plate LV. B.)

The head is considerably shorter than the pleon to the extremity of the uropods, and
has the rostral not greatly longer than the part behind the first antenne. Of the
pereon-segments the fourth, fifth, and sixth are the longest. ‘The first three pleon-
segments have the postero-lateral angles blunt, not produced backwards: these
segments are subequal in length; the fourth is slightly shorter; the coalesced remainder
of the pleon scarcely longer. ‘The telson is very bluntly pointed, not so long as the
combined fifth and sixth segments, with which it is itself coalesced.

The eyes and antenne are as in the previous species, and the same may be said of
the minute gnathopods.

370 REV. T. R. Rv STEBBING ON’ NEW SPECIES OF AMPHIPODOUS

The first and. second pereeopods have the second joint decidedly broader and. longer
than the fourth, which exceeds the fifth in length and is subequal to the sixth. . The
finger is slender, curved, fully half as long as the preceding joint.

‘The third and fourth pereopods have the second, fourth, and sixth joints longer
than in the two preceding pairs; they also have rather large oval branchial vesicles,
of which no trace could be detected on the preceding limbs.

The fifth perseopods are extremely minute, with the upper part only a little more
dilated than the hand, of which the articulation is very indistinct. The finger has
the usual broad base and slender hook.

The pleopods have the branches three-jointed.

The uropods are comparatively broad. The peduncles of the first pair are between
two and three times as long as the branches, reaching beyond the telson, and are
minutely denticulate on the inner margin; the branches are subequal, more or less
strongly denticulate on both margins. ‘The peduncles of the second pair extend but
little beyond the base of the telson, and the coalesced inner branch does not reach its
apex; the outer branch is more than half as long as the inner; both have denticu-
lations on the distal halves of both margins. ‘The peduncles of the third pair extend
a little beyond those of the first; the coalesced inner branch is about as much
longer as the free outer branch is shorter than the branches of the first pair ; both are
denticulate on the distal part of both margins.

The specimen was a rae of an inch in length from tip of rostrum to the extremity
of the uropods.

Habitat. Atlantic, near lat. 0° 19! 2" S., long. 7° 19' EB. It was taken February 2,
1886, at 8.45 p.m., near the surface.

The specific name, from Bpaytc, short, and sédoc, end, refers to the shortness of
the telson.

EXPLANATION OF THE PLATES.

PLATE LI.
Scina acanthodes (p. 852).

The full figure in dorsal view slightly inclined to the left, the line above it indicating
the actual length of the specimen.
C. Dorsal view of cephalon and pereon.
Pl. Dorsal view of pleon.
a.s., a. Upper and lower antenne.
m., mx.1, map. Mandible, first maxilla, maxilliped, a situ.
m. Mandible, separated from the other mouth-organs.
x. A doubtful portion of the oral apparatus.

CRUSTACEANS FROM THE TROPICAL ATLANTIC. Byiiall

gn. 1, gn. 2. The first and second gnathopods.
prp. 1, 2, 3, 4, 5. The first, second, third, fourth, and fifth pereeopods.
plp. Pleopod.
ur. 1, 2, 3. First, second, and third uropods.
T. Telson.
PLATE LII.

A. Scina stenopus (p. 354).

Lateral view of the animal without the head, the natural size being indicated above.
A separate ventral view is given of the head and antenne, on the same scale as the
lateral view of the rest of the animal.

For the separate figures the lettering is the same in all the Plates.
B. Scina edicarpus (p. 356).

The full figure is given in dorsal view, as much of the ventral ganglionic chain being

shown as could be seen through the transparent integument.
PLATE LIILI.
A. Scina rattrayi (p. 358).
The full figure is given in dorsal view.

B. Scina concors (p. 360).

The full figure is given in dorsal view.

PLATE LIV.
A. Scina similis (p. 362).
The full figure is given in dorsal view, the outline of the ovaries appearing as
seen through the transparent integument.
[B. Scina uncipes (p. 365).
The antenne and the limbs of one side are all drawn to the same scale. The
uropods are more highly magnified, being drawn on the same scale as the larger

figures of prp. 3 and prp. 5.
PLATE LV.
A. Khabdosoma piratum (p. 368).
The full figure is given in lateral view.
B. Rhabdosoma brachyteles (p. 369).

The full figure is given in lateral view.

ee

*
vin
ma

Suns, Loot, Goce Vel JW PEL

Del TRRStebbing,

SCINA) ACANTHODE(S.

Mi

o

j
Ma

the

if a *

sl

Sram Loot, foc. Vol, HILFE Li

A i

|

—=S
Se eee
x
SS a
¢ a ce ft AEE ;
af i - Cee
ena eer
eas peer ene SSS
: SSSSSITSs=
is LEI
< ee ing p
aie 5, ee Z
is :

ee
mee
YESS
of SS a
=

prp. I,

Del. TRRStebbing

A. SCINA STENOPUS. B. SCINA OEDICARPUS.

ST

ae iy
r

rt 4 ;
edhe ep

7 ie
yoy

Srand Loot oc. Vo MIL LM.

Del. T R.R.Stebbing.

B. SCINA CONCORS.

Po Aa SINAN CEVAG I EVACvaTe

ERY Why .

i "aN

iain
ay

oT
i‘

rans Look ee. VNU ILLN

Del. TR.R.Stebbing.

A. SCINA SIMILIS. B. SCINA UNCIPES.

SN) tat

Sromds Lok doc. Vol MM FE LIV

S

Del. TR RStebbing

RHABDOSOMA BRACHYTELES.

IB

A. RHABDOSOMA PIRATUM

he Unb ton

XV. On the Cramal Osteology, Classification, and Phylogeny of the Dinornithide. By
T. Jerrery Parker, D.Sc., F.RS., Professor of Biology in the University of
Otago, Dunedin, New Zealand.

Received December 5th, 1892, read February 14th, 1893.

[Puates LVI.—-LXIT.]

ConvEN's. Page

ieaplnr LOMUCLOMY Must cr ter ayaracckei che le(er sl spec, sees ol Vaca oeabevep icise ovailecape neuter custom tienen arcane Oe 373
2, Ibis OF SPHOCINEMS Games) coos¢oodcscsanbacvonoeosoonbbuEanedooaeduoes 374
3. A Comparative Account of the Skull in the Dinornithide .................... 380
4, A Comparison of the Skulls of the Dinornithide with those of the other Ratite ,, 398
5. Measurements of the Skulls of the Ratite ................ ccc ccc cece ces 405
6. Summary of the Cranial Characters of the Ratite ................ 0. cee eee ee 408
(emuhe) ClassiticationsotthemDinornithidce een eee eerie eee ee 413
8. Summary of Cranial Characters of the Subfamilies and Genera of the Dinornithidx. 417
Omelhesbhylorenysortheshatiteaynce erecta eos een ee ae 423
IhistvoiaVVorks reterredetopya steer caeelr et cre erie a ee eee 427
Hxplanationsomthemalatesipr tic scirs favs tetiotelcr error aces eiere ee ene 429
MistiolPAbbreviablonsy | Piessisyas me rrepeycrehee cuts teres thereat sy creer eewe es) eie as sboo oS 430

1. Intropvuctory.

A. FIRST glance at the magnificently illustrated series of memoirs by Sir Richard
Owen on the osteology of the Dinornithide gives the impression that the whole subject
has been exhausted ; but a more careful perusal, aided by a comparison with the recent
works of Lydekker (12) and Hutton (g), is enough to show that the material at Sir R.
Owen’s disposal was far from complete, that skulls were assigned to the skeletons of
certain species on purely conjectural grounds, and that some of the figures were even
made up of portions belonging to different species. ‘The reason of this confusion is
that it is extremely seldom that the bones of a single individual skeleton, or even the
parts of a single individual skull, are found associated together and apart from those
of other individuals.

It is to this circumstance that the chief difficulty of the present investigation was
due—the difficulty of assigning correct names to the various skulls examined. To
mention only the two most recent authorities: Lydekker describes four genera and
nineteen species; Hutton seven genera and twenty-six species: species associated in a
single genus by the one are widely separated by the other: and, most confusing of all,
skulls assigned by Lydekker to certain species are considered by Hutton to have
been wrongly associated with the leg-bones upon which the species were founded.
Moreover, while my enquiries fully confirm the view that the Dinornithide are divisible

VOL. X111.—PART XI. No. 1.— October, 1895. 5

374 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

into several genera, the generic groups, as deduced from a study of the skulls, do not
agree with those of either of the authors referred to. It seems reasonable, however, to
claim that, if constant and definable differences can be shown to occur in the skulls,
these should outweigh mere differences in the size and proportions of the limb-bones.

I have derived the greatest assistance throughout the investigation from frequent
correspondence with Captain Hutton, who has, with rare generosity, placed at my
disposal the wide knowledge of the whole Moa question gained during the course of
his extended, and still partly unpublished, researches. Sir James Hector has most
kindly lent me the entire collection of Moa skulls in the Colonial Museum, Wel-
lington, including the unique skull of Wesopteryx, species @, figured on Plate LX. figs. 20
& 21. Dy. H. O. Forbes has been good enough to lend me the large series of skulls
in the Canterbury Museum, recently collected by himself at Enfield, near Oamaru, as
well as the skulls from four skeletons in the Canterbury Museum, articulated under
the superintendence of the late Sir Julius von Haast. Mr. A. Hamilton, Registrar of
the University of Otago, has placed at my disposal the large collections made by
himself at the Te Aute swamp, near Napier, and at Castle Rocks, Oreti River, South-
land—the latter collection including several immature skulls which have been quite
invaluable for my purpose. I am also indebted to Mr. Hamilton for the drawings from
which figs. 59-64 are taken. Mr. R. J. Kingsley has lent me the skull of a fine
individual skeleton in his possession, the type of Dinornis torosus, Hutton. Mr. J.
Thomson, Lecturer on Applied Mechanics in this University, has devoted a great deal
of time and trouble to taking the photographs from which Plates LVI., LVIII., & LX.
are copied. And, lastly, during my visit to England I have received the kindest help
from Prof. Newton, Dr. Henry Woodward, Mr. A. Smith Woodward, and Mr. H.
M. Platnauer. ‘To all these gentlemen I beg to return my most sincere thanks.

2. List oF SPECIMENS EXAMINED.

As the nomenclature of many species is still doubtful, and as it is desirable to refer
to certain individual specimens in the various collections to which I have had access,
the following list is given in order to facilitate identification :—

Genus Dinoryis, Owen.
1. Dinornis Maximus, Owen.
a. The skull belonging to the large skeleton in the British Museum (Natural
History) and numbered 46050 (Lydekker, Cat. Foss. Birds, p. 232).
6. Portions of a skull in the same Collection, numbered 466313 (figured by Owen,
Extinct Birds of N. Z. pl. Ixii.).
The measurements of this last-named skull do not differ from those of
D. robustus }.

* Through the kindness of Captain Hutton, I have just examined a very fine skull of D. imaaxzmus belonging
to Mr. M*Ewen, of Christchurch. There is nothing to distinguish it from D. robustus—June 1895.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA, 375

2. DINORNIS ROBUSTUS, Owen.
a. The skull belonging to the very fine individual skeleton found at Tiger Hill, Otago,
and now in the Museum of the Philosophical Society, York.

This specimen is referred by Hutton to his D. potens, but I am more than
doubtful whether the distinction from robustus can be maintained, and prefer to
place it in the present species.

6. The skull belonging to an individual skeleton found at Highley Hill, Otago, the
property of Dr. T. M. Hocken. (Otago University Museum.)

c. An imperfect and partly restored skull from Hamilton Swamp, Otago. (Otago Univ.
Mus.)

d. Mandible belonging to an individual skeleton found at Shag Valley, Otago. (Otago
Univ. Mus.)

e. Associated premaxille, maxillo-jugal arch, quadrate, and mandible from Maun-
gatua, Otago. (Otago Univ. Mus.)

f. A cranium from Enfield, Otago. (Dr. H. O. Forbes’s collection.)

This species is figured by Owen, Trans. Zool. Soe. vol. v. pls. liii. & liv. (Ext.

Birds of N. Z. pls. lxii. & Ixiii.).
3. Dinornis Torosus, Hutton.
a. Skull belonging to a nearly perfect skeleton, the type of the species, found in the
Takaka district, Nelson. (Mr. R. J. Kingsley’s collection.)
6. A nearly complete skull from Hamilton Swamp. (Otago Univ. Mus.)

Captain Hutton considers that this skull is probably referable to D. struthioides,
but I can see no differences of any importance between it and the previous
specimen.

This species is figured by Owen as D. ingens, Trans. Zool. Soc. vol. vii. pl. xv.
(Ext. Birds of N. Z. pl. Ixxxii.); also by Jaeger, as Palapteryax ingens, Reise der
Novara, Palaontologie, pls. xxv. & xxvi.

4, Drvorvis, species a.
A damaged cranium of the same size as that of D. torosus, but differing from it in
having the temporal and lambdoidal ridges in contact. (Coll. H. O. Forbes.)

Genus Pacuyornis, Lydekker.
1. PacHYoRNIS ELEPHANTOPUS, Owen.

a. Six crania with separate premaxille, maxillo-jugal arches, quadrates, and mandibles,
from Enfield; in one instance (Plate LX. fig. 22) premaxille and maxillo-jugal
arches were found which exactly fitted a cranium, and I have no doubt that they
belonged to the same individual. (Coll. H. O. Forbes & Cant. Mus.)

6. Two crania, one with (associated) premaxille and mandible, locality unknown.

(Coll. Mus. Wellington.)
c. An imperfect cranium with associated premaxille and mandibie from Hamilton

Swamp. (Otago Univ. Mus.)
3H 2

376 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Hutton (9, p. 153) has brought forward evidence for considering that this skull
should be associated with the leg-bones upon which the species crassus was
founded; if this were the fact it would, according to the nomenclature I have
adopted, be placed in the genus Hmeus, which genus would then change places
with Pachyornis in my table of classification (p. 427). I learn, however, by
recent communication with Prof. Hutton that he is still somewhat uncertain upon
this point, and I think it will give rise to less confusion if I follow Owen and
Lydekker and assign the present skull to the leg-bones upon which the species
elephantopus was founded.

2. PACHYORNIS IMMANIS, Lyd.

Cast of a cranium in the Nat. Hist. Museum, numbered A. 201, and described by
Lydekker, Foss. Birds, p. 344. I doubt whether this specimen can be specifically
distinguished from elephantopus 1.

This species is figured by Owen, as Dinornis elephantopus, ‘Trans. Zool. Soc.
vol. vii. pl. x. (Ext. Birds of N. Z. pls. lix. & Ixxvi.). The best skull from Enfield
is figured from beneath on Plate LX. fig. 22.

3. PACHYORNIS, species a.

a. A cranium from Shag Point, differing from P. elephantopus in its greater breadth,
especially across the postorbital processes, and in having the temporal and
lambdoidal ridges in contact. (Coll. A. Hamilton.)

Figured in outline, Plate LXI. figs. 26 & 39, and Plate LXII. fig. 50.

6. A skull with greatly damaged cranium, from a skeleton named Dinornis struthioides

by Sir J. von Haast, is probably to be referred here. (Canterbury Museum.)

Both these skulls may belong to very muscular individuals of P. elephantopus ;
but I hardly think so, as in all the specimens undoubtedly referable to that species
the temporal and lambdoidal ridges are distinct. ‘They may ultimately be found
to belong to D. rheides, the skull of which is not known.

4, PACHYORNIS, species (3.
A single cranium from Glenmark, differing from P. elephantopus in being decidedly
narrower in proportion to its length.

5. PACHYORNIS, species y.

A single cranium, from Enfield, differing from P. elephantopus in its much smaller
size, but agreeing with it in other respects. It is very possibly a mere variety,
but I have found no intermediate sizes.

1 In the ‘Transactions of the New Zealand Institute,’ vol. xxvi. (1893), I have described a skull of

Pachyornis which is about 10 per cent. larger in nearly all dimensions than P. elephantopus, and have

referred it provisionally to P. immanits.—June 1895.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID#. 377

6. PAcHYORNIS, species a, Lydekker.
The cranium in the British Museum (Nat. Hist.), No. 32197, so named by Lydekker
(Foss. Birds, p. 320).

7. PacHyornis, species 0, Lydekker.
A cranium in the Nat. Hist. Museum, numbered 52205, and described by Lydekker,
lc. p. 345. It agrees in all measurements, except a narrower temporal fossa,
with P. elephantopus.

Genus Mrsoprmryx !, Hutton.

1. M&soPpTERYX CASUARINA, Owen.
. Twenty-nine crania, with several mandibles, premaxille, quadrates, and maxillo-jugal
arches, from Enfield’. (Cant. Mus. and coll. H, O. Forbes.)
6. A cranium from Hamilton Swamp. (Otago Univ. Mus.)
¢. A cranium from Glenmark Swamp, Canterbury. (Cant. Mus.)
d. A cranium from Castle Rocks. (Coll. A. Hamilton.)

The examination of this large series of skulls has convinced me that didina,
Owen, and huttonii, Owen, are synonyms of caswarina; there is a perfect gra-
dation between the larger skulls (caswarina) and the smaller (didina). The same
gradation is found by Prof. Hutton in the case of the leg-bones from Enfield, but
as there are considerable differences between the two ends of the series he prefers

SY)

to keep the species distinct.
e. A skull in the Natural History Museum, numbered 32210, A typical but unusually

large specimen.
f. A skull in the same collection, numbered 32214, 32199.
This species is figured by Owen as Dinornis rheides (with mandible of
Pachyornis), Trans. Zool. Soc. vol. vii. pl. xii. (Ext. Birds of N. Z. pl. Ixxv.).
The best of the Enfield skulls is figured in Plate LX. fig. 19.

" Reichenbach’s name Syornis has priority for this genus, having been applied to the type species casuarina ;
but, as Lydekker has pointed out (11, p. 254), this name clashes with Synornis, Hodgson, and I have therefore
adopted Hutton’s name Mesopteryx, the type species of which (J. didina) is probably only a small form of
casuarina. [Captain Hutton (Trans. N. Z. Inst. vol. xxvii. 1894) now considers that Meionornis, Haast, is
the correct name of this genus.—June 1895. ]

* According to Mr. Forbes (2) the Moa-remains from Enfield belong chiefly to the species elephantopus,
ingens, and rheides. This is hardly correct; the vast majority of the skulls belong to the species now under
discussion, a few to Pachyornis elephantopus, and one or two to Dinornis, sp., and Hmeus, sp. My deter-
minations are confirmed by Prof. Hutton’s measurements of 351 metatarsi from this abundant deposit sent to
the Canterbury Museum. He finds that 181, or more than half, are referable to three doubtful species—casuarina,
didina, and huttonii, all of which I include under casuarina; of the rest he assigns 40 to rhetdes, 38 to crassus,

40 to various species of Dinornis, and 52 to species of Pachyornis.

378 PROF, T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

2. MESOPTERYX, species a.
A cranium and mandible from Te Aute, near Napier. (Coll. A. Hamilton.)
This skull was referred by Prof. Hutton to Cela geranoides, but the evidence for
the determination appears to me to be insufficient, and I think it best to leave the
name in abeyance !.

Figured in outline, Plate LXI. figs. 28 & 41, Plate LXII. fig. 52.

3. MESOPTERYX, species (3.
An entire skull found by Mr. A. M‘Kay in a limestone fissure on Salisbury Tableland,
Nelson. (Col. Mus., Wellington.)

This skull, which was found associated with the cervical vertebre, is one of the
most perfect ever discovered. It does not correspond with any of the figures or
descriptions I have met with and appears to belong to a species the skull of which
has not hitherto been described.

Figured on Plate LX. figs. 20 & 21.

4, MESOPTERYX, species y.
The skull on a mounted skeleton of WM. didina, from Hamilton Swamp. (Otago Univ.
Mus.)
This skull differs from the Enfield specimens of IZ. casuarina in the form of the
orbit, which is right-angled as in Dinornis.
Figured in outline, Plate LXI. figs. 50 & 44, and Plate LXII. fig. 54.

Genus ANOMALOPTERYX, Reichenbach.

1. ANOMALOPTERYX DIDIFORMIS, Owen.

a. Three perfect crania with premaxille, maxillo-jugal arches, quadrates, and mandibles;
all found in the same cave (with the skeletons) at Castle Rocks, Southland.
(Mr. A. Hamilton’s collection.)

6. One perfect and several imperfect immature crania, from the same locality. (Coll.
A. Hamilton.)

ce. A cranium with (tassociated) premaxille ; locality unknown. (Colonial Museum,
Wellington.)

These skulls are assigned to this species in accordance with Prof. Hutton’s
researches (g, p. 125), confirmed by Mr. Hamilton’s discoveries at Castle Rocks.
The skull referred by Lydekker (12, p. 275) to this species is apparently that of
Mesopterya didina.

The most perfect of the immature crania referred to above is figured on
Plate LVIII. figs. 12 & 13,

* Since writing the above, Captain Hutton has lent me for examination a cranium which he considers to
be that of Megalapteryx tenuipes, since it was found in a cave associated with bones of that species. The
skull was a good deal damaged, but appears to agree very closely with that of Mesopterya, species a.—
June 1895.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID2. 379

2. ANOMALOPTERYX PARVA, Owen.
Skull of the very fine individual skeleton in the Natural History Museum, numbered
A. 3, and figured and described by Owen, Trans. Zool. Soc. vol. xi. pl. lii.

It is an open question whether this species is distinct from didiformis; I am
inclined to think it is merely a small variety, and only keep it separate in order to
give measurements of the type specimen.

Figured by Owen as Dinornis parvus, Trans. Zool. Soc. vol. xi. pl. lii.

Genus Emeus, Reichenbach !.
1. Emeus crassus, Owen.

a. Two nearly perfect skulls from Shag Point, Otago; in both, the premaxille and
bones of the palate were fixed in sité by the collector Mr. R. 8. Booth?. (Otago
Univ. Mus.)

A similar skull, from the Maniototo Plains, Otago. (Coll. H. O. Forbes.)

. Four less perfect skulls from Shag Point. (Otago Univ. Mus.)

. About fifty crania belonging to this or the next species, from Shag Point. (Otago

Univ. Mus. and coll. A. Hamilton.)

This species is figured by Owen as D. crassus, Trans. Zool. Soc. vol. x. pl. xxxi.
(Ext. Birds of N. Z, pl. ecxiv.); the beak and mandible of the skull figured as
D. crassus in Trans. Zool. Soc. vol. vii. pl. xi. (Ext. Birds, pl. Ixxvii.) belongs to this
species, and probably the skull called D. gravis on pl. xiv. (Ext. Birds, pl. Ixxxi.).

Sy SS

2. EMEUS, species a.
a. An entire skull found by Mr. R. 8. Booth at Shag Point. (Otago Univ. Mus.)
With the unimportant exception of a slight injury to the left antorbital, this
skull is absolutely perfect and is probably on the whole the best Moa skull ever
found. (See Plate LVI.)
6. 'Two less perfect skulls, also from Shag Point. (Otago Univ. Mus.)
c. About fifty crania belonging either to this or to the preceding species. (Otago Univ.
Mus. and coll. A. Hamilton.)
This species is easily distinguished from elephantopus by the shorter and
narrower beak, but I can find no constant difference between the crania.
Figured on Plate LVI. and Plate LVIII. figs. 9 & 10.

3. EMEUS, species .
The skull on the skeleton of gravis in the Canterbury Museum, so named by Sir J. von
Haast. (Cant. Mus.)
This skull undoubtedly belongs to the present genus, but appears to exhibit well-
marked differences from the two preceding species.
* Captain Hutton (Trans. N. Z. Inst. vol. xxvii. 1894) considers that this name should give way to
Euryapteryx, Haast—June 1895.
* A similar skull was found at the same place by Prof. Hutton and was presented by him to the British

Museum ; it is figured by Owen in Trans. Zool. Soc. vol. x. pl. xxxi., as D. crassus. Lydekker describes it an
Emeus, species a.

380 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

4, Emnus, species y.
The cranium on a skeleton from Hamilton Swamp, named D. gravis by Prof. Hutton.
(Otago Univ. Mus.)
This skull agrees in general characters with the present genus, but has right-
angled orbits—a character I have never observed in any of the preceding species.
Figured in outline, Plate LXI. figs. 54 & 40, and Plate LXII. fig. 58.

3. A CoMPARATIVE ACCOUNT OF THE SKULL IN THE DINORNITHID.

The skulls of the Moas are usually found in a more or less fragmentary condition,
and it will be advisable, for the sake of convenience, to take this circumstance into
consideration and to describe the skull under the following heads :—

. Cranium.

. Premaxilla.

. Maxillo-jugal arch.

. Vomer, palatine, and pterygoid.
. Quadrate.

. Mandible.

. Hyoid.

The cranium is the part most commonly found’ in large deposits of Moa-bones
crania may occur in hundreds, whilst other portions of the skull are of comparatively
rare occurrence. Premaxille, quadrates, and lower jaws are not uncommon, while a
complete maxillo-jugal arch is rare, and very few skulls have been found with the

ea Seasae,

palatines, pterygoids, and vomer uninjured.

a. The Cranium.
In the cranium the following regions may be distinguished :—

i. The occipital region, including the whole posterior portion of the skull; it contains
the occipital condyles and foramen and is produced at the sides into the large
and prominent paroccipital processes.

ii. The cranial roof, continuous behind with the occipital region.

iti. The base of the skull, also continuous behind with the occipital region ; posteriorly
it is raised into a prominent squarish elevation, the basitemporal platform ;
anteriorly it is continued into a more or less cylindrical rod of bone, the
rostrum, which forms the axis of the beak.

iv. The lateral surface of the cranium, presenting three well-marked depressions—the
orbit in front; the temporal fossa immediately behind the orbit and separated
from it by a downward projection of the skull-roof, the postorbital process; and
the tympanic cavity, bounded above and separated from the temporal fossa by
an outstanding mass of bone, the sguamosal pronvinence, and bounded behind
by the paroccipital process.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA. 381

v. The ethinoidal region, lying anterior to the orbits, enclosing the olfactory chambers,
and produced forwards into a median vertical partition, the mesethmoid, anky-
losed below with the rostrum. |

vi. The cranial cavity.

i. The occipital region. (Plate LVIII. fig. 9; Plate LIX. fig. 14; Plate LXII.
figs. 47-58.)

The occipital condyle (oc. con.) is distinguished by its more or less pedunculate
character; its dorsal surface is usually somewhat flattened, and its hemispherical
posterior face often presents a well-marked dimple-like depression indicating the
position of the notochord. Its median portion is formed by the basioccipital, its
lateral portions by the exoccipitals (fig. 14); the latter bones converge from below
upwards, so that a very narrow strip of the dorsal surface of the condyle is furnished
by the basioccipital (fig. 15).

The occipital foramen varies in form from subcircular to squarish ; in Dinornis the
plane of the foramen is nearly at right angles to that of the basitemporal platform,
while in the other genera it is distinctly inclined backwards; in the former case the
occipital condyle projects beyond the dorsal margin of the foramen, a unique peculiarity
pointed out by Owen. In some instances the condyle projects beyond the level of the
paroccipital processes, but as a rule the reverse is the case.

Immediately above the occipital foramen is a median vertical ridge (fig. 9), the
occipital crest (oc.cr.; median vertical ridge, Owen), which is connected ventrally with
another ridge lying immediately above the foramen, at first close to its dorsal border
and afterwards diverging laterad and becoming lost on the paroccipital process. I
propose to call this the supraforaminal ridge (sup.for.r.; lower transverse supra-
occipital ridge, Owen) ; it is well developed in all the species, while the crest, although
usually well marked, is often obscure and sometimes absent in Dinornis.

The occipital crest passes anteriorly into a transverse ridge which extends laterally
on each side to the base of the paroccipital process; the median portion of this
lambdoidal ridge (lamb.r.; transverse occipital ridge, Owen) is frequently double
(figs. 1, 9, & 12), a transversely elongated lozenge-shaped area being enclosed between.
its two divisions; we may thus distinguish an anterior lambdoidal ridge (ant.lamb.r.),
which is dentated and serves as the chief line of insertion of the neck-muscles, from a
posterior lambdoidal ridge (post.lamb.r.). The distinction between the two varies
greatly in different species, being best marked in Anomalopteryx, and hardly distin-
guishable in Dinornis (Plate LXI. figs. 25-34). On each side of the occipital crest,
close to its junction with the posterior lambdoidal ridge, is a more or less well-marked
depression, the supraoccipital fossa (fig. 9, s.0c.fos.).

The precise relation of the lambdoidal suture to the ridges of the same name is
difficult to make out, as the ridges are obscure in young specimens, but it appears to be
a little in front of the posterior ridge.

VoL. XIll.—Part x1. No. 2.—October, 1895.

O39
4

382 PROF. T, JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

The paroccipital processes (par.oc.pr.) are large and prominent; in Dinornis they are
comparatively flat, and their ventral edges are evenly curved and not greatly produced
downwards (Plate LXII. fig. 47); in Mesopteryx casuarina (fig. 51) they are very
convex backwards, and their ventral borders are produced into bluntly pointed
processes which extend downwards to the level of the mamillar tuberosities. Other
species show intermediate conditions (figs. 47-58). In Hmeus the supraforaminal
ridge stops short at the base of the paroccipital process; in most other species it is
continued on to its lower angle.

Externally to the supraoccipital fossa there is frequently a foramen, usually
continued into a groove, for one of the cerebral veins. ‘The line of junction between
the supra- and exoccipitals is probably situated mesiad of this foramen, but I have not
seen a skull in which those two bones are distinct ; the paroccipital process is no doubt
constituted entirely by the exoccipital.

ii. The cranial roof. (Plate LVI. fig. 1; Plate LVII. fig. 5; Plate LVIII. fig. 12;
Plate LIX. fig. 17; Plate LX. fig. 20; Plate LXI. figs. 23-34.)

The roof of the cranium is formed mainly by the parietals (pa.) and frontals (7),
the coronal suture passing transversely about halfway between the posterior lambdoidal
ridge and a line drawn between the posterior margins of the postorbital processes
(figs. 5 & 12). It is usually evenly arched both from before backwards and from side
to side. The lateral curvature is least in Dinornis (fig. 47), m which also, taking
the basitemporal platform as horizontal, the anterior or frontal region is considerably
deflected (fig. 35). In Mesopteryx casuarina (fig. 27), Mesopteryx, species « (fig. 30),
and Emeus, species y (fig. 84), the roof is swollen on each side of the middle line,
producing a double tumidity; in the other species of Emeus (figs. 38, 42, & 46)
there are slight unpaired elevations in the anterior and posterior frontal regions, the
intervening portion being flat or depressed. The roof is continuous behind with the
occipital region and with its squamosal prominences, narrowed between the temporal
fossee, and immediately in front of these produced into the large postorbital processes
(post.orb.pr.), which pass outwards and downwards, forming the posterior boundary of
the orbit and almost meeting the maxillo-jugal arch (figs. 3 & 7). In most cases the
postorbital process forms an even curve; in Dinornis and Pachyornis it is divisible
into horizontal and descending portions, the former gently inclined downwards, the
latter vertical.

Anteriorly the postorbital process passes into the swpraorbital ledge, which is itself
continued into a short preorbital process (pre.orb.pr.) formed by_a part of the ankylosed
lacrymal (figs. 5, 7, & 8). The margin of the orbit thus constituted exhibits certain
well-marked differences in the various genera and species. In most cases the whole
orbital margin is eveuly curved or slightly sinuous; in Dinornis the supraorbital ledge
is at right angles to the postorbital process (figs. 23 & 35); in Anomalopteryx the
postorbital angle thus formed is slightly obtuse (fig. 86). The doubtful crania referred

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA, 383-

to Mesopteryx, species y (figs. 30 & 44), and Emeus, species y (figs. 34 & 40), differ from
the other species of the genera in which I have provisionally placed them in possessing
a distinctly right-angled orbit.

The anterior part of the cranial roof, between the preorbital processes, is formed
partly by the frontals, partly by the nasals. ‘The junction between the two bones is
well shown in the young skull of Anomalopterya didiformis (figs. 12 & 17), in which,
the nasals being lost, each frontal is seen to present a deep triangular notch for the
lateral portion of the corresponding nasal, while between the notches the frontals end
in a straight transverse border, with which both the mesial portions of the nasals and
the nasal process of the premaxilla articulate.

The nasals are in contact by their ventral surfaces with the ethmoid, exposed by
their removal in figs. 12 & 17, and unite with one another in the middle line beneath
the premaxilla, for the reception of which the mesial portion of the conjoined bones is
excavated in the form of a shallow, parallel-sided groove, the premazillary fossa
(figs. 23 & 34). ‘Thus the mesial portion of the nasals is hidden in the entire skull,
and the roof of the olfactory chamber is formed in this region of a triple layer of
bone—premaxilla externally, ethmoid internally, and nasals between. Laterad of the
premaxillary fossa the nasal appears on the surface as a triangular bone the curved
base of which forms the posterior boundary of the nostril (figs. 5, 7, & 8), its external
angle being produced, in a young specimen of Anomalopteryx didiformis, into a short
maxillary process. In Dinornis torosus this process is continued as a slender bar of bone
which passes vertically downwards, in close contact with the anterior border of the
lacrymal (preorbital process), completing the lacrymal foramen externally, and articu-
lating by its lower or distal end with the maxilla. In Hmeus, species a, and E. crassus
this process is represented by a distinct ossification which I propose to call the
maxillo-nasal (figs. 7 & 8, ma.na.). I have been unable to ascertain the precise
condition of these parts in the other genera, but Iam inclined to think there is a
distinct maxillo-nasal in Anomalopteryx, while in Pachyornis there appears to be a
maxillary process as in Dinoriis.

In the skull referred to Mesopteryx, species 3, the anterior portion of the skull-roof
is marked with numerous shallow pits arranged more or less regularly in lines
radiating backwards and slightly outwards from the edges of the premaxillary fossa
(Plate LX. fig. 20). They appear to indicate the presence of a crest of specially strong
feathers, which must have consisted in the present case of paired tufts, since the pits
are absent in the middle line. In the type specimen of Dinornis torosus, and in the
specimen of D. robustus (D. potens, Hutton) in York Museum, similar pits occur,
extending across the middle line and on to the preorbital processes. Similar but less
distinct pits occur in two skulls of Anomalopteryx didiformis. Since in the two last-
named species the pits are absent in certain skulls, it seems probable that the crest

was possessed only by the male.
312

384 PROF, T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

iii. The base of the skull. (Plate LVI. fig. 2; Plate LVII. fig. 6; Plate LX. figs. 19,
21, & 22.)

The basitemporal platform (b.temp.pl.) is a prominent flattened elevation on the base
of the skull, having a nearly vertical posterior face separated by a deep groove, the
precondylar fossa (fig. 2, pr.con.fos.), from the occipital condyle ; in the fossa two pits,
probably venous, occur in Dinornis. The middle region of the posterior edge of the
platform is usually somewhat excavated, the excavation being bounded on either side
by a larger or smaller prominence, the mamillar tuberosity (Owen: basioccipital process,
Lydekker, mam.tub.), which forms the postero-lateral angle of the platform. Owen
considers these tuberosities to occur at the junction of the basisphenoid and basi-
occipital, but the examination of a young skull shows them to lie at the junction of
the basioccipital, exoccipital, and pro-otic (figs. 6,7, & 14). About the posterior third
of the platform is formed by the basioccipital, the rest by the basitemporal underlying
the basisphenoid; whether the basitemporal extends so far back as I have shown it in
fig. 6 is uncertain.

Each mamillar tuberosity is separated from the corresponding paroccipital process by
a deep paroccipital notch, immediately behind which occur one large and several small
foramina. ‘The larger of these is the vagus foramen (precondyloid foramen, vagal
foramen, Owen, vag.for.) for the ninth and tenth nerves; the smaller holes or condyloid
foramina (con,for.) give exit to the twelfth nerve. The notch between the paroccipital
process and the mamillar tuberosity is sometimes converted into a foramen by a slender
bar of bone bridging it over. Immediately in front of it is the carotid foramen
(fig. 6, car.for.), leading into the canal by which the carotid artery passes to the
pituitary fossa. In young skulls the greater part of the carotid canal is an open groove.

The antero-lateral angles of the basitemporal platform are formed by a pair of broad
projections, flattened from above downwards, the basipterygoid processes (b.pt.pr.) ;
they articulate by their distal ends with the pterygoid bones. Between the basi-
pterygoid process and the mamillar tuberosity, the lateral border of the platform is
obliquely furrowed by a deep groove for the Eustachian tube (figs. 2 & 6, eus.gr.).
The edges of the groove are often produced into roughened ridges, but these never
meet below so as to convert the groove into a tube; in this respect the adult Moa
resembles the Kiwi at the time of hatching.

In the centre of the basitemporal platform and between the inner ends of the
Eustachian grooves is usually to be found a more or less well-marked depression of
variable form—sometimes short and wide, sometimes long and narrow. At its posterior
end there frequently occurs a deep fossa, which in one young specimen was represented
by a foramen opening into the cranial cavity a short distance posterior to the pituitary
fossa. The aperture (fig. 6) is the posterior basicranial fontanelle (median venous
foramen, basisphenoidal mid-ventral canal, Owen, p.b.cr,fon.), marking the incomplete
concrescence of the parachordal cartilages of the embryo (24); the depression apparently
marks the incomplete extension mesiad of the basitemporal.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDZ. 385

Another interesting embryonic character is the occasional presence of a small pit or
foramen situated in the middle line at the junction of the rostrum with the basitem-
poral platform, and at the level of the anterior borders of the basipterygoid processes ;
when patent it leads into the pituitary fossa. This is a remnant of the anterior
basicranial fontanelle (a.b.cr,fon.) through which the pituitary pedicle passes from the
pharynx to the cranial cavity. Its retention in the adult is somewhat remarkable, as
in Apteryx (24) it is completely covered by the rostrum, and only visible when, in the
young skull, the latter is stripped from the underlying cartilage.

The rostrum as seen from below has the appearance of a more or less cylindrical rod
of bone; actually it is doubtless crescentic in section, as in other birds. It is frequently
slightly constricted towards its junction with the basitemporal platform, its middle
region is often much compressed and keel-like, and its anterior end is pointed. It will
be further considered in connection with the ethmoidal region, but one remark must be
made here as to its position. It usually lies nearly or quite in the same horizontal plane
as the basitemporal ; but in Dinornis the two are set at an obtuse angle (about 150°)
with one another, thus giving rise to the deflected beak characteristic of that genus.

iv. The lateral surface of the cranium. (Plate LVI. fig. 3; Plate LVII. fig. 7;
Plate LVIII. fig. 13; Plate LIX. fig. 18; Plate LXI. figs. 35-46).

The tympanic cavity is a deep depression bounded behind by the paroccipital
process, above by the squamosal prominence, and mesially by a greatly pitted surface
furnished by the exoccipital and pro-otic bones. It is continued downwards and
forwards into a sort of pocket, the anterior tympanic recess, bounded in front by a
thin, oblique, quarter-cylinder of bone, the pretemporal wing (figs. 6 & 13, pr.temp.w.).
A young skull shows this process to be part of the combined basisphenoid and
basitemporal ; probably, as in other birds, it is formed by the latter. In some cases
the anterior tympanic recess becomes largely filled up with spongy bone and thus
reduced to a quite insignificant cavity; it is best marked in young skulls of
Anomalopteryx didiformis, but is also large and conspicuous in Mesopteryx casuarina
and some other species.

The squamosal prominence (sq.prm.) is a thick outstanding mass, convex above,
deeply concave below, and formed by the squamosal bone (mastoid, Owen), which
articulates mesially with the parietal, exoccipital, alisphenoid, and pro-otic. It is
preduced downwards into the zygomatic process (mastoid process, Owen, zyq.pr.),
which, in the entire skull, lies immediately external to the quadrate, is directed
slightly forwards, and is sometimes bifid (fig. 3); it varies considerably in length. It
is continued upwards on to the lateral surface of the squamosal by a slightly wavy,
subvertical posterior temporal ridge (post.temp.r.), the dorsal end of which joins the
inferior temporal ridge described below (figs. 3 and 13).

The margin of the tympanic cavity varies in form: in Dinornis (fig. 35) its dorsal
edge, furnished by the squamosal, forms an even curve with its posterior edge, furnished

586 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

by the paroccipital process, the latter being directed slightly backwards. In Emeus
(figs. 38, 42, and 46) the two form a slightly acute angle, the paroccipital process
being directed forwards ; in HL. crassus there is often a kind of step at the junction of
the two, the margin of the cavity being therefore sinuous. In Anomalopterysx (fig. 36)
the paroccipital is still more sharply inclined forwards, the dorsal and posterior edges
of the tympanic cavity meeting at a very acute angle, and the width of the cavity from
before backwards being greatly reduced.

On the roof of the tympanic cavity is the large obliquely-transverse articular facet
for the head of the quadrate ; its outer end is immediately mesiad of the posterior
temporal fossa described below, and is bounded behind by the zygomatic process; from
this point it passes inwards and backwards, its inner end being exactly opposite the
Kustachian groove. According to my observations the inclination of the quadrate facet
(see Lydekker, 12, p. 298) is a point of no classificatory value; a straight line drawn
through the axis of the facet makes an angle with the sagittal plane, which rarely sinks
below 130° or rises above 140°, and which is, moreover, very variable, sometimes
differing considerably on the two sides of the same skull. The outer two-thirds of the
facet is furnished by the squamosal, the inner third in about equal proportions by
the exoccipital and pro-otic (Plate LXII. fig. 70).

The zygomatic process is, as it were, bent round the quadrate facet so as to bound its
outer end posteriorly as well as externally, and its flattened posterior face is continued
backwards into a horizontal tympanic ledge (fig. 2), the inner border of which gives
attachment to the tympanic membrane, while externally it is produced into a rough
horizontal supratympanic ridge.

Immediately posterior to the quadrate facet is a large pneumatic foramen leading
into the diploé of the pro-otic and squamosal. Mesiad of the inner end of the facet
are usually three more or less well-marked depressions in the roof of the tympanic
cavity; of these, the one nearest to the quadrate facet is a pneumatic foramen; the
hindermost of the other two is a simple depression in the bone, while the foremost is
the fenestral recess, containing the fenestra ovalis and fenestra rotunda. Just anterior
to the recess, and separated from it by a vertical bar of bone, is the small aperture for
the facial nerve, immediately dorsad of which is a large pneumatic pit. There is
another small but deep pneumatic foramen encroaching on the anterior border of the
quadrate facet, and the whole wall of the tympanic cavity is honeycombed with less
constant depressions of various sizes.

The temporal fossa isa deep depression between the postorbital process in front
and the squamosal prominence behind. It is limited above by a strongly marked
temporal ridge (linea semicircularis, temp.r.) which marks the origin of the temporal
muscle; the ridge forms an even curve, passing from the posterior edge of the post-
orbital process, at first backwards and upwards, and then curving round and passing
downwards and forwards (figs. 3 and 13), so that the whole temporal fossa has a strong
backward slope; at the base of the zygomatic process it joins the posterior temporal

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA. 387

ridge already mentioned (p. 385), and finally ends in a short, blunt pretympanic process,
situated immediately in advance of the outer end of the quadrate facet. The ventral
portion of the posterior limb of the tympanic ridge, between its junction with the
posterior tympanic ridge and its termination in the pretympanic process, is conveniently
distinguished as the inferior tympanic ridge (inf.tymp.r.). Between the inferior and
posterior ridges thus defined, a more or less triangular space is enclosed which serves
for the origin of the second portion of the temporal muscle, and may be distinguished
as the posterior tympanic fossa (post.tymp,fos.). Lastly, the main temporal fossa is
imperfectly divided, in some species, by a vertical mid-temporal ridge (figs. 12 & 13,
m.temp.r.) at about the junction of its anterior and middle thirds and just over the
lateral portion of the coronal suture.

The temporal fossa varies considerably in width in the different species, being
narrowest proportionally in Mesopteryx casuarina, widest in Anomalopterya didiformis.
But its most striking feature is the different degree of its extension on to the cranial
roof; in most species the distance between the right and left ridges is but little less
than the diameter of the cranium in the temporal region, while in Anomalopteryx
didiformis it is hardly more than half that dimension. Moreover, there is usually a
flat area between the temporal and lambdoidal ridges (see Plate LXI.); but in
Anomalopteryx didiformis (fig. 24), Dinornis, species «, and Pachyornis, species «
(fig. 26), in all of which the temporal fossa is unusually large, the two ridges are in
contact. In Pachyorms elephantopus, in which there is a wide area between the two
lambdoidal ridges, the anterior one often touches the temporal before uniting with the
posterior ridge.

The portion of the dorsal region of the temporal fossa lying behind the mid-temporal
ridge is formed, as already stated, by the parietal, the part in front of the ridge by the
frontal (figs. 7 & 18); its ventral region is furnished by the alisphenoid, which, in the
young, joins the parietal by a straight horizontal suture situated at about the level of
the roof of the tympanic cavity. Below, the alisphenoid is separated from the basi-
sphenoid, in the youngest cranium I have seen, by a gently curved suture (fig. 18}
extending forwards and downwards from the trigeminal foramen.

The trigeminal foramen (foramen ovale, Owen, trig.for.), for the second and third
divisions of the fifth nerve, is situated in the side-wall of the cranium below the
temporal fossa, and in the same transverse plane as the posterior edge of the
basipterygoid process (figs. 3 & 15); it is bounded above by the alisphenoid, below by
the pretemporal wing. Sometimes it is partially or completely divided into two
passages by a horizontal bar of bone; this duplication of the foramen seems to be
constant in Mesopteryx casuarina. About 1 em. below and in front of the trigeminal
foramen is a small (qu. arterial?) foramen situated just above the base of the basi-
pterygoid process as in Apterya.

The orbits are smaller than in the majority of birds, and, as Owen pointed out, there
can hardly be said to be an interorbital septum, owing to the backward extension of

388 PROF, T, JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

the olfactory chambers. There the Dinornithide show, to a less degree, the state of
things which reaches its maximum among birds in Apterya.

As already remarked, the upper margin of the orbit, formed by the supraorbital
ledge, may be either evenly curved, sinuous, or right-angled. The roof of the cavity
formed by the projecting ledge is almost horizontal from within outwards; its mesial
wall slopes from the supraorbital ledge downwards and inwards to the optic and
lacerate foramina, the slope being far steeper in the narrow-skulled genera than in the
wide-skulled Dinornis.

Both roof and mesial wall of the orbit are pierced with vascular foramina, but the
first feature of importance met with in this region of the skull is the optic foramen
(op.for.). This lies in the postero-ventral region of the orbit, slightly in advance
of the base of the rostrum. In Hmeus it is a nearly circular aperture directed from
the cranial cavity outwards, forwards, and upwards, and its lower margin is separated
from that of its fellow of the opposite side by a distance of about 9 or 10 mm. In
Dinornis it is proportionally smaller and oval in outline; its lower margin is formed
by a thin obliquely horizontal shelf of the presphenoid, represented in the other genera
by a mere ridge, and the distance between the two foramina is about 30 mm.

The greater part of the inner wall of the orbit is furnished by the orbital plate of
the frontal (fig. 18), which, in the young skull, articulates below with the inferior
aliethmoid (inf.al.eth.) and the orbitosphenoid (or.sph.); the suture passes about
2mm. above the optic foramen, which is therefore bounded above by the orbito-
sphenoid, below by the presphenoid.

Bounded above by the shelf or ridge forming the lower margin of the optic foramen,
and below by the rostrum, is a depression, the presphenoid fossa (fig. 18, pr.sph.),
very obvious in Dinornis, Pachyornis, Anomalopterya, rather obscure in Mesopteryx
and Hmeus. ‘The vertical plate of bone separating the fosse of the right and left
sides is very thin, and is the only indication of an interorbital septum, thus showing
an interesting approximation to the typical avian structure of the presphenoid.

Immediately posterior to the optic foramen is the aperture called by Owen the
prelacerate foramen or foramen rotundum, but more conveniently named the lacerate
fossa, as it isnot a simple foramen, but a pit including three perfectly distinct foramina.
One of these (Plate LXII. fig. 76, v!), placed postero-dorsally, is the opening of a
tunnel-like excavation in the alisphenoid, and transmits the orbitonasal nerve. The
second (iii), below and in front of the first, is bounded behind by the alisphenoid and
in front by the basisphenoid, and transmits the oculomotor nerve. ‘The third (vi & a),
situated below and in front of the second, perforates the alisphenoid and probably
transmits the internal ophthalmic artery ; it has in its hinder margin the aperture of a
canal through which the sixth nerve enters the orbit. The fourth nerve (iv) enters
through a very small foramen placed just above that for the third nerve, and quite
outside the lacerate fossa.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA. 389

v. The ethmotdal region. (Plate LVL. figs. 3 & 4; Plate LVII. figs. 7 & 8; Plate
LVIII. figs. 11-13; Plate LIX. figs. 16-18.)

The mesethmoid (figs. 16 & 18, meseth.) is a vertical plate of bone ankylosed in
the adult by its whole ventral border to the rostrum, continuous behind with the
presphenoid, and having its dorsal border sloping from behind forwards immediately
beneath the nasal process of the premaxilla. It is perforated posteriorly by an
irregular fenestra of very variable size (fig. 16), which, in the dried skull, places the
two olfactory chambers in communication with one another.

The olfactory chambers thus connected form a spacious cavity lying beneath and in
front of the anterior portion of the cranial roof and extending nearly as far back as the
optic foramen. Postero-dorsully they are perforated, on each side of the middle line,
by one or sometimes two rather irregular apertures which serve for the passage of the
olfactory nerves. Their roof and side-walls are formed by the lateral or ectoethmoids,
which are divisible into three principal parts. The first, or superior aliethmoid
(figs. 17 & 18, sup.al.eth.) extends almost horizontally outwards on each side from
about the posterior half of the dorsal border of the mesethmoid, underlying the
nasals and the anterior part of the frontals (fig. 16) and forming the roof of the
olfactory chambers. The second portion, or inferior aliethmoid (fig. 18, inf.al.eth.),
is a concavo-convex plate forming the side-wall of the posterior part of the chamber,
continuous behind with the pre- and orbitosphenoids, and in the young skull articu-
lating by suture with the orbital plate of the frontal.

These two are the only parts of the ectoethmoid retained in the majority of Moa
skulls, but in good specimens there is to be seen more or less of the third portion of
the ectoethmoid known as the antorbital plate (prefrontal, Huxley), and corresponding
with the dilated part of what I have called the fifth portion of the ectoethmoid in
Apteryx (24). This consists of an upright plate of thin bone (figs. 3 & 7, a.ord.)
placed somewhat obliquely in the front part of the orbit; its mesial border is
continuous with the inferior aliethmoid, and its lateral border is ankylosed to the
lacrymal. It is continued forwards (in Hmeus, species «) beyond its ankylosis with
the lacrymal by a thin scroll-like bone, the alinasal (figs. 3, 4, 7, & 8, al.n.), which
nearly meets above with the superior aliethmoid and below with the triangular process
described on page 390. The ossified posterior turbinal (figs. 6 & 8, turd.) is an
ingrowth from the posterior end of the antorbital.

The lacrymal (figs. 7 & 8, dac.) is completely ankylosed with the lateral border of
the antorbital, is deeply notched, or in Anomalopteryx didiformis perforated, for the
lacrymal duct, and presents below an oblique facet for articulation with the maxilla.
The maxillary process of the nasal, or maxillo-nasal of Hmeus, lies immediately in front
of and parallel with the lacrymal, converting the notch into a foramen (lac.for.).

In Dinornis the antorbital is a much stouter bone than in the other genera, and is,
therefore, more frequently found intact. It is much narrower than in the remaining
forms, owing to the great size of the supraorbital fenestra (see p. 390), and is sloped

VOL. XIII.—PART XI. No. 3.—October, 1895. 3K

390 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

obliquely forwards from its lower edge; it is firmly ankylosed to the inferior aliethmoid
as well as to the lacrymal. Its ventral border, instead of being thin and delicate,
takes the form of a stout transverse ridge or bar of bone forming the posterior boundary
of the aperture between the olfactory chamber and the posterior nasal passage. This
postchoanal bar, as it may be called, is a very obvious structure when the skull is
viewed from below (see Owen, 19, pl. 56, or Ext. Birds of N. Z. pl. 65. fig. 1, 9),
and constitutes a striking difference between Dinornis and the smaller genera.

A final ectoethmoidal structure, very characteristic of the whole family, is the trian-
gular process (Plate LV1. figs. 2 & 4, tri.pr.), a horizontal plate springing by a broad
base from the mesethmoid (fig. 18) and extending outwards at the level of the lacrymal
foramen. It forms the anterior boundary of the aperture between the olfactory
chamber and the posterior nasal passage, and its posterior border is connected by a
horizontal ridge with the postchoanal bar. The bar, ridge, and triangular process
are together called by Owen the girdle or cingulum olfactorium.

The antorbital plate is perforated dorsally by the supraorbital fenestra. ‘This, in
Dinornis torosus, is a single large aperture, more than 1 cm. in length, and continued
dorsad into a shallow fossa on the underside of the lacrymal (preorbital process).
Apparently the Harderian gland fitted into both the fossa and the fenestra, the orbito-
nasal nerve passing through the anterior part of the latter in its passage from the orbit
to the olfactory cavity. Above the posterior end of the dorsal margin of the fenestra
is a small foramen also leading into the olfactory chamber.

In Emeus, species « (fig. 7), the supraorbital fenestra (swp.orb.fen.) is quite small
and the Harderian pit (Hard.fos.) above it very deep; posterior: to these is a
foramen (a) leading into the olfactory chamber, and beneath it a fossa (4), some-
times perforated. In Anomalopteryx didifornis an intermediate arrangement is
found; the supraorbital fenestra is of moderate size and is connected by a wide
groove with a fossa in the posterodorsal region of the antorbital—this last answers
to fossa 6 in Hmeus, and to the posterior moiety of the single fenestra in Dinornis;
above it is a small foramen answering to the similarly placed hole (2) in Pachyornis
and Dinornis.

In good specimens a coiled posterior turbinal (fig. 6) is seen to spring from the inner
surface of the posterior end of the antorbital.

vi. The cranial cavity. (Plate LVIII. figs. 10 & 11; Plate LIX. figs. 15 & 16.)
The cavity for the brain has the usual avian form ; its length and breadth are nearly
equal, the greatest breadth being in the temporal region. ‘The entire cavity may be
divided into the following regions :—
a. ‘I'he metencephalic fossa for the medulla oblongata.
8. The cerebellar or epencephalic fossa for the cerebellum.
y. The mesencephalic fossz for the optic lobes.
6. The pituitary fossa and the optic platform for the pituitary body and the
optic chiasma.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDZ. 391

e. The cerebral or prosencephalic fossee for the cerebral hemispheres.
¢. The olfactory or rhinencephalic fosse for the olfactory lobes.

The metencephalic fossa is furnished by the post-pituitary part of the basis cranii; it
is gently concave both from before backwards and from side to side, and is terminated
in front by the dorsum sellz, while behind it is continued on to the flattened dorsal
surface of the occipital condyle. Near its posterior end is seen on each side the large
vagus foramen (va.for.) and mesiocaudad of this the two small condyloid foramina
(con. for.).

In front of the vagus foramen and separated from it by a vertical bar of bone about
2 mm. wide is the internal auditory meatus (fig. 16, int.aud.m.), a shallow pit
containing the foramina for the facial and auditory nerves. The minute adbducent
foramen for the sixth nerve lies on a line joining the internal auditory meatus with
the middle of the dorsum sell (fig. 15, abd.for.), and about 5 or 6 mm. from the
latter. Between the abducent foramina a median aperture of considerable size is
sometimes found, in one instance communicating with the pituitary fossa; this is
the internal opening of the posterior basicranial fontanelle.

The cerebellar fossa (fig. 11, cer.fos.) is in the roof of the cranium and is bounded
mainly by the supraoccipital and parietals above and by the pro-otics laterally. It is
bounded behind by the dorsal edge of the occipital foramen, and in front by the
median portion of the tentorial ridge, while ventrally it passes insensibly on each side
into the metencephalic fossa. It is marked by a series of transverse grooves corres-
ponding with the gyri of the cerebellum. On each side, about 3 or 4 mm. above the
internal auditory meatus, is the small /occular fossa, which varies considerably in size
in different individuals.

The mesencephalic fossa (mesen.fos.) lies in the alisphenoid, laterad of the dorsum
sellee (fig. 10) and is bounded externally by the ventral portion of the tentorial ridge.
It is largely occupied by a shallow depression for the root of the trigeminal nerve,
and from this depression the trigeminal foramen proceeds directly outwards (fig. 16,
trig.for.), and the tunnel-like orbitonasal foramen (orb.na.for.) forwards and slightly
outwards to the lacerate fossa (vide supra, p. 388).

The pituitary fossa or sella turcica (pit.fos.) is an almost spherical depression in the
middle of the cranial floor. It is bounded behind by a ridge, the dorsum selle,
which curves forwards on each side and ends at the small oculomotor foramen
(oc,for.). In front the pituitary fossa is bounded by a transverse prepituitary ridge
(fig. 16, pr.pit.r.), anterior to which is a wide ledge, the optic platform, terminating
on either side in the optic foramen (op.for.). Almost vertically above the prepituitary
ridge is a similar transverse prominence, the preoptic ridge (fig. 16, pr.op.7.), forming
the upper boundary of the optic platform and passing laterad into the tentorial ridge,

The optic foramen is bounded behind by a vertical bar of bone separating it from
the apertures for the third nerve and the internal ophthalmic artery (fig. 16). The
oculomotor foramen (0c.for.), which is the uppermost of the two, is continued back-
wards by & groove placed sometimes just above, sometimes just below the continuation

3K 2

392 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

of the dorsum sella: the internal ophthalmic foramen (int.op.for.) is also continued
by a groove which extends downwards and backwards to the entrance of the internal
carotid artery (car,for.). The internal opening of the minute pathetic foramen (path.
for.) is placed about 1 mm. dorso-laterad of the oculomotor. The internal carotid
canals (car.for.) open by paired apertures placed close together in the posterior wall
of the pituitary fossa. And lastly, in occasional instances, the fossa is perforated
anteriorly by the anterior basicranial fontanelle.

The cerebral fosse lie altogether in front of the cerebellar fossa, there being no
overlapping of the cerebellum by the hemispheres such as occurs in Apteryx. They
are separated from the cerebellar fossa by the prominent tentorial ridge, which, starting
in the centre of the skull-roof, almost exactly above the dorsum sellz, sweeps at first
backwards, outwards, and downwards, and then forwards, downwards, and slightly
inwards, and finely comes to an end near the extremity of the preoptic ridge. From
the middle of the tentorial ridge a somewhat less prominent median elevation, the
bony falx, passes forwards and ends just over the crista galli, marking the separation
of the hemispheres dorsally.

The line of separation between the hemispheres on the ventral surface is similarly
indicated by a low median ridge which extends from the crista galli backwards to the
preoptic ridge. For a distance of about 5 mm. (in Hmeus) on each side of this
ridge the cerebral fosse are floored by the presphenoid (fig. 15, pr.sph.); from it the
orbito-sphenoids (or.sph.) extend outwards and backwards, forming the dorsal
boundaries of the optic foramina. The resemblance of this portion of the Moa’s
skull to that of an embryo Kiwi is very striking; in the adult Apterya the
presphenoid undergoes a remarkable shortening.

The anterior moiety of the roof and side-walls of the cerebral fossa is furnished by
the frontal, the posterior portion by the parietal and alisphenoid. A low horizontal
ridge running a short distance above the parieto-alisphenoid suture indicates the
presence of a similarly-placed sulcus on the cerebrum.

The olfactory fossw are paired oval pits placed vertically at the anterior end of the
cerebral fossee. ‘They are separated from one another by a narrow bony ridge, the
crista galli, and the floor of each is perforated by a variable number of somewhat
irregular apertures for the branches of the olfactory nerve.

b. The Premazilla.

The premaxilla is a triradiate bone which may be described in general terms as
consisting of a thickened body which forms the end of the beak, a median dorsal nasal
process, paired ventro-lateral maxillary processes, and thin ventral palatine processes
connected in front with the body and externally with the maxillary processes.

The anterior extremity of the body, forming the apex of the beak, is rounded in
Dinornis and Emeus, bluntly pointed in Pachyornis, Mesopteryx, and Anomalopteryx.

Qs

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDZ. 39

Its gently curved alveolar margins are marked, especially in Dinornis, with a broad,
shallow groove, and are continued horizontally inwards into a flattened ventral plate ;
from this is given off a strong ascending keel, formed posteriorly of paired plates, but
solid in front and gradually diminishing in height towards the tip of the beak. The
dorsal edge of the triangular vertical keel thus produced is expanded to form a pro-
minent median ridge passing behind into the nasal process and in front continued
more or less distinctly to the apex of the beak; this ridge is best developed in Dinornis,
in Lmeus it is almost obsolete. The portion of the vertical keel included between the
ridge above and the ventral plate below has a triangular form, and is, as already stated,
bilaminar posteriorly ; it constitutes the prenarial septwm (Owen), and is especially
well-marked in Dinornis ; the extremity of the ankylosed rostrum and mesethmoid
fits between its lamine.

From the ventral plate of the body the thin, horizontal, palatine processes pass
backwards, diverging slightly so as to enclose a median palatine notch, through which
in the perfect skull (figs. 2 & 6) the anterior end of the vomer and rostrum are seen.
In Hmeus the posterior end of the inner margin of the palatine plate gives off a some-
what pedate vomerine process (vo.pr.) which passes inwards and clamps the vomer ;
this process is small or absent in the other genera. The proportional length of the
palatine process and of the body of the premaxilla differs greatly in the various genera.

From the posterior end of the alveolar margin of the body is given off the irregular
horizontally-flattened maxillary process, which extends backwards (figs. 3 & 7, max.pr.)
dorsad of and in close contact with the anterior end of the maxilla, and ends close to
the base of the maxillo-palatine and immediately below the ventral end of the maxillo-
nasal.

The nasal process is a flat plate extending backwards and upwards from the body
to the fronto-nasal suture, its posterior end lying, as already stated, in a shallow,
parallel-sided groove, the premaxillary fossa, furnished by the mesial portion of the
nasals. Its anterior or proximal end, where it joins the body of the bone, is slightly
thickened in Dinornis, considerably thickened and somewhat triangular in section in
Anomalopteryx, Pachyornis, and Mesopteryx, greatly thickened and almost cylindrical
in Hmeus.

ce. The Maxillo-jugal Arch.

This consists of the usual three bones, the maxilla, jugal, and quadrato-jugal.

The mavilla consists of two parts, a slender external and posterior portion, the
maxilla proper, and an irregular expanded antero-mesial portion, the maaillo-palatine
process (Plate LVII. figs. 6,7, & 8, max.pal.; Plate LXII. figs. 63 & 64). The
maxilla proper is a slender rod presenting a flat ventral surface, an oblique lateral
surface, a very narrow, smooth mesial surface, and a dorsal surface, to the whole length
of which the jugal is applied.

The maxillo-palatine, in Dinornis, Pachyornis, Mesopteryx, and Anomalopterys, is a
very irreguiar shell of thin bone, with a spacious cavity, the antrum, opening poste-

394 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

riorly by an aperture placed in the angle between the maxilla and the palatine (fig. 63).
Its ventral surface forms a nearly flat plate, roughly triangular in form ; its lateral
border and postero-lateral angle pass insensibly into the maxilla proper; its postero-
mesial angle is produced into a short recurrent process which articulates with the
palatine; the hinder half of its medial border also articulates with the palatine, and its
anterior angle and the fore half of its mesial border with the premaxilla. From this
flat ventral plate an irregular hollow mass of bone extends dorsad and fits into a space
left between the other facial bones at the base of the beak. Its anterior surface
appears in the ventro-lateral corner of the external nostril (figs. 7 & 8), articulating
with the palatine process of the premaxilla below, with the rostrum mesiad, and with
the maxillary processes of the premaxilla and nasal externally. Its dorsal region
articulates with the anterior border of the triangular process of the mesethmoid (fig. 8).
Its posterior surface, which is smooth and concave, forms part of the anterior wall of
the posterior nasal aperture.

In Emeus the structure of the maxillo-palatine is strikingly different ; its ascending
portion, instead of being a hollow shell, is an irregular flattened plate (fig. 64), either
quite solid or presenting a mere vestige of the antrum in the form of a very small pit
on the posterior surface. By this peculiarity Hmeus is sharply distinguished—at least
so far as my enquiries go—from the remaining Dinornithidee

The jugal is a slender bone forming the greater part of the dorsal surface of the
maxillo-jugal arch (figs. 6 & 7). It articulates by about the anterior half of its ventral
surface with the maxilla, by the posterior half with the quadrato-jugal. Near its
posterior end the dorsal edge of the bone is raised into a low triangular process
which extends upwards towards the postorbital process of the frontal.

The quadrato-jugal (figs. 6 & 7) articulates by more than the anterior half of its
outer surface with the jugal and maxilla; thus the whole length of the bone is exposed
on the mesial side of the maxillo-jugal arch, while less than half appears on the
lateral surface. In the greater part of its extent it is flattened from side to side, but
posteriorly it is much thickened and presents on its inner surface an oval facet for
articulation with the quadrate.

d. The Vomer, Palatine, and Pterygoid.

The vomer in fully adult specimens of Hmeus, species a, and E. crassus is a delicate
bone (figs. 2 & 6, vo.) formed of paired laminz united in front but diverging behind,
and enclosing an acute dihedral angle. In its whole length it embraces the rostrum ;

nteriorly it passes dorsad of the vomerine processes of the premaxilla and articulates
with the maxillo-palatines ; posteriorly each lamina turns outwards and fits into a
groove between the palatine and pterygoid, its lateral border articulating with the
mesial border of the palatine, and its posterior extremity being covered ventrally by the
vomerine process (vo.pr.) of that bone.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDZ. 399

In Mesopteryx, species (3 (Plate LX. fig. 21) there are distinct paired vomers ; a similar
condition appears to obtain in Anomalopteryx and in young skulls of Hmeus crassus.
Owen’s figure of Dinornis torosus (20, pl. 15; Ext. Birds of N. Z. pl. 82) also shows
paired vomers ; probably their concrescence is a sign of age.

The palatine is a delicate, twisted plate of bone, passing from the anterior end of the
maxilla in front to the pterygoid behind (fig. 6). Its anterior end is somewhat fan-
shaped and underlaps the maxillo-palatine; immediately posterior to this expanded
portion it is notched for articulation with the recurrent process of the maxilla. The
whole bone is twisted, the inner border of its anterior end becoming ventral in the
middle and finally external at the posterior end; at the same time the dorsal border
turns mesiad, reaching to within a short distance of the rostrum. ‘The posterior end
is obliquely truncated and articulated by all but its ventral extremity with the vomer,
becoming ankylosed with it in the adult. The ventral extremity of the posterior
border is produced into a thickened squarish vomerine process (vo.pr.) which underlies
the posterior end of the vomer, and laterad of this process the bone presents a
short pointed end which underlies the pterygoid.

The pterygoid (pt.) is a stout irregular bone with a bluntly-pointed anterior and a
thickened posterior end. By about the anterior half of its ventral border it articulates
with the combined palatine and vomer; by its inner surface it articulates with the
basipterygoid process; its outer surface lies parallel to and in close contact with the
orbital process of the quadrate ; and its posterior end expands into a somewhat pedate
surface for articulation with the quadrate.

e. The Quadrate.

The quadrate consists, as usual, of a body bearing the condyle for articulation with
the mandible, an upwardly-directed otic process, terminating in the head for articu-
lation with the tympanic cavity, and a forwardly-directed orbital process.

The articular surface on the head is somewhat wider at its outer than at its inner
end, and presents no trace of the double facet found in Apterye. The otic process is
subtrihedral, presenting a lateral border running upwards from the quadrato-jugal
facet, a mesial border from the outer condyle, and an anterior border from the orbital
process ; a posterior surface between the mesial and lateral borders, a mesial surface
between the anterior and mesial borders, and a lateral surface between the anterior
and lateral borders. On the mesial surface, just where the otic process merges into
the body of the bone, is a pneumatic foramen which varies greatly in size in the
different species and even in different individuals of the same species; speaking gener-
ally, it appears to be large in Dinornis, Anomalopteryx, and Pachyornis as well as in
Mesopteryx, species 3; smaller in Emeus, and smallest of all in Mesopteryax casuarina.
In Anomalopterya didiformis there is sometimes a second pneumatic foramen on the

396 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

posterior surface of the bone at a slightly lower level than the mesial one, and in one
specimen of Dinornis robustus there are two additional foramina on the posterior
surface, one at the level of the mesial aperture, the other close to the head. On the
whole the characters of the quadrate appear to be too variable to be of much use for
systematic purposes.

The orbital process is laterally compressed and bluntly pointed. On its mesio-ventral
border is a small facet (pterapophysial facet, Owen) for the pterygoid, and at the base
of the same border and extending downwards on to the body of the bone is a larger
facet with which the posterior end of the pterygoid articulates.

The ventral face of the body bears the usual two condyles for articulation with the
mandible. The inner condyle is placed transversely and is separated by a narrow
interspace from the outer condyle, which is set at an angle of about 40° with the sagittal
plane. Immediately above the outer extremity of the external condyle is the deep
hemispherical fossa for articulation with the quadrato-jugal.

f. The Mandible.

The lower jaw consists of two gently curved rami ankylosed with one another in front
in a wide symphysis.

Fach ramus is expanded at its posterior end to form a cup-like articular surface for
the quadrate, the outer border of which projects but slightly beyond the level of the
ramus, while internally it is much produced and ends im a triangular internal angular
process. ‘The surface for the outer condyle of the quadrate forms a long narrow facet
running parallel with the outer and posterior edge of the cup: that for the inner
condyle is an oval surface, deeply concave from within outwards, situated on the anterior
margin of the cup, just mesiad of its junction with the ramus. At the base of the inner
surface of the internal angular process is a nearly circular pneumatic foramen, which,
however, is sometimes absent in Hmeus and Mesopteryx. At the posterior end of the
surface for the outer condyle the bone is produced into a posterior angular process,
which is small in Dinornis, large and prominent in the other genera. Both internal and
posterior angular processes are continued on to the ventral surface of the articular cup
by ridges which meet each other below: they are especially prominent in Hmeus.

The ventral edge of each ramus has a sigmoid curve in all genera but Anomalopteryz,
being convex downwards in its posterior, concave in its anterior half, and the latter
being more or less deflected. The curvature of the dorsal follows to some extent that
of the ventral border, but is less regular. The deflection is most marked in Dinornis, in
which, when the mandible is placed upside down on a horizontal surface, the tip of the
beak is raised 35-45 mm. above the horizontal. In Pachyornis, Mesopteryx, and Emeus
this distance does not exceed 10-15 mm., and in Anomalopteryx the jaw is nearly
straight. The general plane of the ramus is vertical or nearly so in Pachyornis,

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA. 397

Anomalopteryx, Emeus, and Mesopteryx; in Dinornis it is inclined, sloping outwards
from its lower border. The anterior end of the dorsal edge presents a distinct alveolar
groove, like that on the premaxilla; the posterior end is produced into a rough
irregular coronoid process for the insertion of the temporal muscle.

The form of the symphysis naturally follows that of the premaxilla, being broad in
Dinornis and Emeus, narrow in Pachyornis, Mesopteryx, and Anomalopterya. The result
of this is that the entire mandible is U-shaped in the two first-named genera, V-shaped
in the others. The symphysis itself is almost horizontal, and shows considerable
variation in its proportional length. Its ventral or outer surface is marked with a broad
ridge, like that on the premaxilla, very well marked in Dinornis, less so in Pachyornis,
Mesopteryx, and Anomalopteryx, and only just raised above the surface in Emeus.

On the inner surface of each ramus, 1—2 cm. in advance of the articular cup, is the
dental foramen for the mandibular nerve ; it perforates the bone, appearing externally
at the hinder end of a deep groove along which the nerve runs, entering the ramus,
between the two lamine of the dentary, at its anterior end. ‘Two small foramina lie in
the groove, one near its lower border, the other towards the anterior end of its dorsal
border; they perforate the bone and probably transmit nerves to the tissues on the
inner surface of the jaw.

Another pair of foramina, apparently for the symphysial branches of the mandibular
nerve, lie in the posterior edge of the symphysis, usually just in the re-entering angle
between the rami. ‘They generally lie side by side, occasionally one above another ; in
some cases they are united at their origin into a single foramen, and in one instance
have moved forwards to near the anterior end of the symphysis.

Emeus has, of all genera, the stoutest and most coarsely built jaw, Anomalopteryx
coming nearest to it in this respect. The mandible of Dinornis, in spite of its strong
symphysis, has comparatively weak rami, but the most delicate lower jaw of all is that
of Mesopteryx casuarina.

In young specimens the mandible readily divides into three parts: a symphysial
portion, containing the ankylosed dentaries and the splenials, which latter are separate
in still younger skulls ; and the posterior portions of the two rami, each containing the
articular, angular, supra-angular, and coronary (fig. 7). None of the specimens I have
seen show the latter group of bones in the separate condition, but in a young
mandible of Anomalopterya didiformis the outlines of the angular and supra-angular
can be traced where they overlap the articular. The splenial is well shown in a
mandible of Hmeus crassus; it extends from about the level of the dental foramen
forwards to within 4 mm. of the symphysis. Lastly, in the type specimen of Dinornis
torosus, in which the splenial is absent, the articular is continued forwards into a
cylindrical tube of bone lying immediately mesiad of the dental foramen and evidently
representing the superficially ossified proximal end of Meckel’s cartilage.

VOL. XIUI.—ParT x1. No. 4.—October, 1895.

)

OL

398 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

eg. The Hyoid.

In the skull of Emeus, species «, figured on Plate LVI., the posterior cornua of the
hyoid—probably the only ossified parts—are present, as well as the larynx and anterior
end of the trachea. ‘They consist of gently curved rods of bone, 57 mm. long, 2mm.
in diameter, and expanded at both ends.

Mr. Booth, who found this skull, told Professor Hutton that he saw in the ear a
delicate hair-like bone which he was unable to preserve; no doubt this was the
columella auris.

4, A COMPARISON OF THE SKULLS OF THE DINORNITHID WITH THOSE
OF THE OTHER RATITA.

The occipital region is usually less clearly marked off from the skull in the other Ratite,
the lambdoidal ridge being comparatively faint : the fully adult Apteryx australis forms,
however, an exception ; in it the lambdoidal ridge is very strongly marked, and there is
a distinct angulation between the roof and the hinder wall of the cranium. ‘There
is no indication of the anterior lambdoidal ridge, and although the median occipital
region is swollen over the cerebellar fossa, the occipital crest is generally poorly
developed or absent; it is most distinct in Apteryx australis. The supraforaminal ridge is
obvious in all, and is continued to the angle of the paroccipital process; in Apteryx it
is interrupted, just at the margin of the foramen magnum, by a notch. <Apterya is the
only genus which resembles the Moas in its pedunculate occipital condyle, as also in
the great breadth—in relation to height—of the entire occipital region. In having
the plane of the occipital foramen vertical or nearly so the Dinornithide stand alone.

The roof of the cranium is more rounded in the other Ratitz, and in all but Apteryx
and Casuarius the parietal region slopes backwards instead of being nearly flat. Asa
consequence of this, the temporal fossa has a much stronger backward inclination in
Struthio, Rhea, and Dromeus, and the postorbital process is nearly in the same trans-
verse plane as the zygomatic instead of being well in advance of it. In this respect the
Cassowary approaches very closely to the Moas.

Another very marked difference in the roof of the cranium is due to the relatively
small size of the eyes in the Dinornithide. In the Ostrich, Emu, and Rhea, the
width of the orbit from the postfrontal process to the lacrymal is about equal to the
width of the cranium at the paroccipital processes ; in Casuarius galeatus it is about four-
fifths of the width, in the Moas not much more than half. Moreover, the interorbital
region of the skull-roof in the Ostrich, Rhea, Emu, and Cassowary is narrow, while its
preorbital region broadens out suddenly owing to the presence of large wing-like
orbital processes to the lacrymals. In the Moas this projection is only represented by
the comparatively small body of the lacrymal. In Apteryx there are no post or pre-
orbital processes, and the skull-roof narrows gradually from the occipital to the nasal

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID. 399

region. Struthio is peculiar in having a forwardly-directed process, given off from the
supraorbital ridge, which meets the lacrymal, enclosing a notch or foramen.

The nasals of the Dinornithide differ from those of all the other Ratite in the
junction with one another of their posterior ends above the ethmoid, so that none of
the latter appears on the surface. In the Ostrich, Rhea, Emu, and Kiwi a lozenge-
shaped area of the ethmoid appears between the bases of the nasals; in the Cassowary
the same arrangement obtains in the young birds (25) at the time of hatching, but in
the adult the actual condition of things is hidden by the development of the crest.

The maxillary process of the nasal is well developed in Struthio and Apterys,
extremely small and delicate in Dromwus, absent in Rhea and (?) Casuarius. It is
somewhat remarkable that the absence or vestigial nature of this process should be
given as a general character of the Ratite by Garrod (4) and Fiirbringer (3).

On the base of the skull of the other Ratitee the basitemporal platform is not raised
beyond the general level of the skull-floor to anything like the same extent as in the
Moas, and the precondylar fossa is therefore comparatively shallow. It is best defined
in Apteryx, in which also the mamillar processes are even longer proportionally than in
the Dinornithids, while in the other genera they are considerably less developed,
being fairly prominent in Caswarius, small in Dromeus and Struthio, and obsolete in
Rhea.

The paroccipital notch may or may not be bridged over by bone. ‘The vagus foramen
lies in the notch in Apteryx, Dromeus, and Casuarius, mesiad of it—as in the Moas—
in Struthio and Rhea. In the Australian genera the ventral edges of the Eustachian
tubes are prominent and sometimes meet, converting the groove into a canal. In the
Ostrich, Rhea, and Kiwi there are actual tubes, the closure of the groove being more
complete in the two first-named genera than in Apéeryx.

Another variable point is the extent to which the carotid canal is closed by bone
and the resulting position of its external aperture. In the Ostrich, as in the Moa, the
carotid foramen lies in or slightly in front of the paroccipital notch; in the Emu and
Cassowary it is on the lateral surface of the basitemporal platform, immediately caudad
of the Eustachian groove and laterad of the mamillar tuberosity ; in Rhea it is ina
similar position, but slightly further forward ; and in Apteryx it is altogether in front
of the mamillar tuberosity and fully visible from below.

The basipterygoid processes are most dinornithic in Apterya and Dromeus; in the
other three genera they are proportionately longer and more slender. Between their
bases the minute anterior basicranial fontanelle sometimes occurs in Struthio, Rhea,
Casuarius, and Dromeus, but I have never seen any trace of it in the adult Apterya.
Amongst my specimens a young /hea is the only one showing any sign of the posterior
basicranial fontanelle.

‘The rostrum of the Ostrich is rounded below; that of all the other genera is keeled,
except at the posterior end, the carination being most marked in Apterya.

312

400 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Apterya is the only genus besides the Dinornithide which has a well-marked
tympanic ledge ; in all the others the cavity is bounded laterally only by the sharp-
edged supratympanic ridge.

The quadrate facet on the roof of the tympanic cavity presents some interesting
variations in the various genera (Plate LXII. figs. 65-70). In the Dinornithide, as we
have seen, it is almost regularly oval, its inner third furnished by the pro-otic and
exoccipital, its outer two-thirds by the squamosal (fig. 70). In Struthio (fig. 65), at
about the middle of its anterior or mesial border, it is somewhat deeply notched ; as
all the specimens at my disposal were adults, the boundaries of the constituent bones
could not be made out. A similar notch occurs in Dromus (fig. 67) and Casuarius
(fig. 68), and the facet is divisible into an inner or prootic-exoccipital and an outer or
squamosal region; moreover, in the Cassowary the anterior edge of the external
region of the facet is encroached upon by the alisphenoid, which thus takes its share in
furnishing the articular surface. In Rhea (fig. 66) there is no notch, and the pro-otic
portion of the facet appears in an adolescent skull as a wedge of bone near the posterior
end of its anterior or mesial margin, and divided into a larger anterior and a smaller
posterior portion; the posterior end of the facet is therefore formed mainly by the
exoccipital: the alisphenoid enters into the facet as in Caswarius. Lastly, in Apterya
the facet (fig. 69) is distinctly divided into three portions—a mesial furnished partly by
the pro-otic, partly by the exoccipital1, a postero-lateral by the squamosal, and an
antero-lateral by the alisphenoid.

The anterior tympanic recess is large in all genera but Apteryx. The Ostrich differs
from the others in having a large (gu. venous?) foramen excavated in the postero-ventral
region of the pretemporal wing immediately cephalad of the Eustachian groove.

The zygomatic process is more slender in the other Ratite than in the Moas, and is
directed outwards as well as forwards and downwards. The squamosal prominence
is always obscure, and there is never more than the merest trace of the posterior
tympanic fossa and ridge. In the Emu there is a distinct facet on the inner surface of
the distal end of the zygomatic process for articulation with the quadrate. In Rhea the
squamosal sends off, posterior to the zygoma, a process which passes forwards and is
nearly met by a similar process from the pretemporal wing, the two together forming
an almost complete ring round the posterior projection of the head of the quadrate.

The temporal fossa and ridge are far less strongly marked in the Ostrich, Rhea, Emu,
and Cassowary than in the Moas, and in none of them is there any trace of the mid-
temporal ridge. In Apterya, on the other hand, the temporal fossa is very wide from

1 In my paper on Apteryw (24) I incorrectly described this facet as being furnished exclusively by the
pro-otic, but a renewed examination of stage M shows a distinct suture passing vertically across ib and
dividing off a posterior part furnished by the truncated end of a bar-like process of the exoccipital just above
the fenestral recess.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA, 401

before backwards, reaching well in advance of the optic foramen ; the temporal ridge
is strong, and there is a mid-temporal ridge at about the junction of the middle and
posterior thirds of the fossa.

The comparative size of the orbit constitutes, as already remarked, one of the most
striking differences between the Moas and Kiwis and the other Ratite. In the latter
the orbits are separated from one another by a median vertical plate of bone, the
interorbital septum, represented in the Dinornithide only by the wall between the
presphenoid fossee and wholly absent in Apteryx. In an adolescent skull of Rhea the
posterior third of the septum is formed by the presphenoid, the anterior two thirds by
the mesethmoid. There is thus produced an almost straight mesial wall to the orbit,
passing above and behind into the arched upper and posterior wall, which is formed
above by the orbital plate of the frontal, below by the alisphenoid. In the Dinor-
nithide, there is no clear distinction between the mesial and the postero-dorsal wall of
the orbit, the two passing insensibly into one another; the anterior part of the mesial
wall is formed by the inferior aliethmoid, owing to the backward extension of the
olfactory capsules between the eyes, and the presphenoid is limited to a small area
below the optic foramen. In Apterya the last trace of the interorbital septum dis-
appears, the swollen aliethmoids reaching back to the optic foramen.

The optic foramina are close together in all the more typical Ratite, being separated
from one another in front by the edge of the interorbital septum, which is very thin in
all but Struthio; the adjacent foramina show, however, considerable variations. In
the Ostrich (fig. 71) the oculomotor foramen (iii & vi) lies immediately behind the
optic and is continued into the interior of the skull bya groove, into the floor of which
the canal for the sixth nerve opens, the oculomotor and abducent nerves having there-
fore a common entrance into the orbit. The internal ophthalmic artery apparently
goes out separately by the foramen marked a. The small foramen for the fourth nerve
(iv) lies just above that for the third, and the orbitonasal foramen (v!) is an oblique
passage just behind it. In a young specimen the foramina for the third, fourth, and-
sixth nerves are represented by an oblique cleft communicating with the optic foramen.
The trigeminal foramen for the second and third divisions of the fifth nerve may be
divided by a narrow vertical bony bar.

In Rhea (fig. 72) the oculomotor foramen (iii and vi) is immediately behind and below
the optic, and the canal for the sixth opens just within its margin ; below and in front
of it is an equally large aperture (a), which probably transmits the internal ophthalmic
artery. The oculomotor and pathetic nerves enter through very oblique foramina (iv, v')
in the usual positions. In Dromeus (fig. 75) the sixth nerve (vi) hasa special foramen
below the oculomotor (iii). In Casuarius (fig. 74) the third nerve makes its exit
through a notch (iii) in the posterior margin of the optic foramen, the sixth (vi)
through a special foramen; a notch (a) in the ventral border of the optic foramen
possibly transmits the internal ophthalmic artery. In Apteryx (fig. 75) there are

402 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

distinct orbitonasal, oculomotor, and abducent foramina, but the fourth nerve goes out
through the optic foramen. ‘Thus all the Australasian genera agree in having a
distinct foramen for the sixth nerve ; the Moas are peculiar in having the oculomotor,
orbitonasal, and abducent foramina sunk in a fossa (fig. 76).

In the structure of the ethmoid Apterya stands at one end of the series, the Dinorni-
thide in an intermediate position, and the remaining Ratite at the other end. In the
Moas, as we have seen, the lateral ethmoid extends backwards almost to the optic
foramen, its posterior part consisting of the gently sloping inferior aliethmoid, which is
continued in front into the obliquely set antorbital. In Apteryx the place of both these
portions is taken by the shell-like aliethmoid, which extends from the optic foramen to
the lacrymal, bulging outwards in its whole extent and undergoing a special dilatation
immediately caudad of the lacrymal. In the remaining Ratite the olfactory cavity is
not continued backwards behind the antorbital, the latter springing directly from the
mesethmoid and passing outwards and forwards to the lacrymal. In Struthio and
hea only the mesial portion of the antorbital is ossified, so that in the dry skull a
considerable space is left between its outer edge and the lacrymal. In Dromeus and
Casuwarius the ossification extends to the lacrymal, and the dorsal portion of the bone
is hollowed by a deep pit for the lacrymal gland. In Struthio, Rhea, and Casuarius
the descending process of the lacrymal bone is merely notched for the lacrymal duct ;
in Dromeus and Apterys it is perforated.

The postchoanal bar formed by the ventral border of the antorbital is most distinct
in the Ostrich and Kmu. The mesethmoid is perforated posteriorly, so as to place the
olfactory chambers in communication, in the Ostrich, Rhea, and Emu, but not in the
Cassowary and Kiwi. In the possession of the triangular process of the mesethmoid
the Moas stand alone.

Leaving Apterys aside, the chief difference between the Dinornithide and the other
Ratite as regards the cranial cavity is the greater size of the mesencephalic fosse in
the latter. The pituitary fossa of the Moas is most nearly approached by that of Struthio,
in which the thickness of the presphenoid gives rise to a broad optic platform, poorly
marked in all the other genera. The cerebral fosse are more pointed anteriorly in the
other genera and are continued forwards into deep conical olfactory fossee; in this point
also it is the Ostrich which approaches most nearly to the Moas; its hemispheres are
blunter and its olfactory lobes shorter than in the American and Austro-Malayan forms.

Apteryx agrees with the Dinornithide in the small size of the mesencephalic fosse,
but is quite peculiar in the great proportional size of the cerebral and olfactory fosse.
The hemispheres extend backwards over the cerebellum, the cerebellar fossa being
therefore pushed backwards and the tentorial ridge made almost horizontal. Owing
to the great size of the olfactory fossee, the crista galli is nearly as long as the basis
cranii from the dorsum sellee to the occipital condyle.

The premaxilla has the usual structure in all. In the Ostrich alone the palatine

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID£. 405

plates are quite vestigial!; in the other genera they extend backwards and articulate
with the vomer: there is no vomerine process as in some of the Moas. In Séruthio,
Rhea, and Dromeus the body of the bone is very flat, but in Caswarius its median
region is elevated into a strong arched keel, from which the nasal process proceeds, and
there is a distinct prenarial septum. In Apteryx the form of the body is essentially
similar: its height is equal to its breadth, and there is a short, thick prenarial septum.
The unique form of the beak in this genus is due to the shortening of the body of the
premaxilla and to the elongation of the region between the prenarial septum in front
and the turbinals behind.

The maxilla shows a wide range of variation. In the Ostrich (Plate LXII. fig. 59) it
is a flat bone divided posteriorly into palatine and jugal processes, and sending off from
its mesial border an axe-head-shaped maxillo-palatine, which articulates by its thickened
inner edge with a facet on the side of the vomer. The lateral half of this process is
double, consisting of dorsal and ventral lamin so arranged as to enclose a wedge-shaped
cavity, the antrum, open behind. In Rhea (fig. 60) the maxillo-palatine is a broad
flat plate which gives off from its dorsal surface a nearly vertical, slender, ascending
process, which is attached by an outer crus to the maxilla proper and by a long inner
crus to the maxillo-palatine. Between the two crura is a small cavity opening behind,
apparently the vestige of an antrum. In the Emu (fig. 61) the maxilla is narrow,
except at its anterior end, where it broadens out into a maxillo-palatine having the form
of a pocket, wide from side to side, narrow from above downwards, and opening behind
along its whole width; this cavity is obviously the antrum, resembling pretty closely
that of the majority of the Moas (fig. 63), but situated farther forwards. In Caswarius
galeatus (fig. 62) the maxilla proper is still narrower, but the maxillo-palatine has the
form of a long conical pouch, like a jelly-bag, its point directed forwards, dorsad of the
palatine process of the premaxilla, and its base widely open behind. Lastly, in Apterya
the maxilla is a long flat bone and the maxillo-palatine is represented only by a narrow
seam-like projection of its mesial border: there is no dorsal prolongation of the
maxillo-palatine, the walls of the antrum being entirely membranous.

It is obvious that in the structure of the maxillo-palatine, upon which Huxley largely
founded his classification of birds (10), the Dinornithide, with the exception of meus,
approach most nearly to the Australian Ratite. Prof. Huxley makes no mention of
the antrum, describing the maxillo-palatine of the Emu, Cassowary, Moa, and Kiwi as
flat imperforate plates.

In the structure of the vomer it is Apterya which comes nearest to the Dinornithide,
that genus having a single vomer, deeply cleft posteriorly and ankylosed with the
palatines and pterygoids. The maxillo-palatines touch it by a part of their thin mesial
edges, but do not articulate with it. In Caswarius it has the same general form and

1 In the skulls in the Otago University Museum, as also in Huxley’s (ro) and Selenka’s (27) figures ; but

in my father’s figures of advanced embryos (25) large palatine processes are shown.

404 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

relations, but is proportionally much longer and more slender ; it articulates laterally
with the maxillo-palatines. In Dromeus it is also forked behind, but its anterior end is
expanded into a thin, flat plate, which articulates with the maxillo-palatine and is under-
laid by the palatine process of the premaxilla. In Rhea it is deeply forked in front,
each limb of the fork being underlaid by the corresponding palatine process of the
premaxilla. In Struthio it is flattened anteriorly, and on each side presents a facet for
articulation with the maxillo-palatine ; although shorter than in the other Ratite, it is
considerably longer in the specimens in the Otago Museum than in Huxley's figure,
reaching as far forward as the anterior end of the nasals, and its forked posterior
end being connected on each side by ligament with the corresponding palatine.

In the characters of the palatines also Apterysx is the nearest ally of the Dinornithide.
In it they are short bones, about the same length as the vomer, and having much the
same twist as was described above in the Moas. ‘They are also expanded posteriorly,
where they unite with the vomer and pterygoids, and are overlaid in front by the
maxillo-palatines. Each is, however, clamped along its whole lateral edge by the long
palatine plate of the maxilla, and does not reach so far forward as the palatine plate of
the premaxilla. In Struthio the palatines are long, flat rods, their posterior ends
expanded for union with the pterygoids, and their anterior ends passing ventrad of the
maxillo-palatines and reaching nearly as far forward as the anterior end of the vomer.
In Rhea, Dromeus, and Casuarius the palatines are short, thin plates, more or less
curved, which pass from the pterygoids behind outwards and forwards to the palatine
processes of the maxille; they are not ankylosed with either the vomer or the
pterygoids in the Rhea and Emus in the Otago Museum, but are firmly united to
both in Casuarius galeatus.

The pterygoid is a rod-shaped bone in all but Sétrwthio, in which it is expanded in
front. Without the opportunity of examining good specimens of adolescent skulls, it
is impossible to enter into a detailed account of the modifications of this bone in the
various genera.

In all the other Ratite except Apteryx, the head of the quadrate resembles that of
the Moas, bearing an elongated oval articular surface broader at its lateral than at its

‘mesial end, and showing no trace of division into two facets. In Apteryx, as I have
pointed out elsewhere (24), the quadrate is practically double-headed ; the details of its
form are, however, subject to considerable variation. In an adult A. oweni, the mesial
end of the head bears a very distinct, nearly circular surface for articulation with the
facet furnished by the pro-otic and exoccipital (Plate LXII. fig. 69) ; passing outwards
from this it narrows considerably and at its lateral end is greatly expanded, forming a
surface, very convex from before backwards, for articulation with the concaye surface
furnished by the squamosal and alisphenoid. In an adult A. australis there is a
perfectly distinct facet on the anterior surface of the outer end of the head for articu-
lation with the alisphenoid ; in another specimen of the same species the usually distinct

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID. 405

facets are so close together that the head is virtually single. In Dromeus alone
there is a facet on the base of the otic process, just anterior to the lateral ridge,
for articulation with the surface already mentioned on the zygomatic process.

It was pointed out that in all the Moas there is a pneumatic foramen on the mesial
surface of the otic process, while a second foramen occurs on the posterior surface in
certain instances. ‘The only other case in which I have found the mesial foramen is
that of Struthio, in which there is an extremely small aperture in the corresponding
position, looking, however, more like a vascular or nervous than a pneumatic foramen.
The posterior foramen is of moderate size in the Emu and Cassowary, and is situated
near the base of the otic process; in the Ostrich it is higher up and proportionally
smaller; in hea it is on the posterior margin of the head, immediately below the
articular surface. In the adult Kiwi there are several small foramina in the same
position, but at the time of hatching there is a single large foramen in the same position
as in Anomalopteryx. ‘The otic process in Apterya is peculiar for sending off small
mesial and lateral processes just below the head.

In the orbital process the Cassowary presents the closest resemblance to the Moas,
but is more compressed and blunter at the apex than in the latter. In the Emu it is
blunt and thick, and takes a more nearly horizontal direction than usual. In the Ostrich
it is also compressed and the apex truncated. In khea it is extremely short and blunt.
In Apteryx it is unusually long and strongly compressed, so as to have the form of a
thin vertical plate with a thickened ventral rim.

In the mandible the internal angular process and ridge are best developed in Apteryx
and to a less degree in Rhea, but are also well marked in Dromeus and Casuarius.
The posterior angular process and ridge are large in the Ostrich, Emu, and Cassowary,
obscure in the Rhea and Kiwi. In the general form of the jaw the Moas come nearest
to the Cassowary, in which the rami are moderately stout and slightly deflected distally.
‘The comparative weakness of the mandible in the other large Ratite is very marked,
and in Apteryx the immense length of the symphysis separates it at once from all
the other genera; in the strength and solidity of the jaw and in the size of the
coronoid process Apteryx australis is, however, the only form which approaches the
Dinornithide.

5. MmASUREMENTS OF THE SKULLS OF THE RATITA,

Measurements of the skulls of Moas are given by Owen, Haast (5), and Hutton (9).
It is, however, desirable for purposes of comparison that a more complete set of
measurements should be given, and that the precise way in which they are taken should
be accurately defined.

Owing to the early ankylosis of the bones of the cranium, it is impossible to determine

VOL. Xtt.—ParT x1. No. 5.—October, 1895. 3M

406 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

such points as the bregma, lambda, pterion, &c., and as the main object of the measure-
ments is to serve as a means of identification, it is important to select standard points
which can be readily made use of in any fairly well preserved skull. Moreover, as a
matter of convenience it is desirable that the measurements should be made with some
common and readily procurable instrument. I have therefore taken them all with
callipers, so that in the case of curved surfaces, such as the roof of the skull, the chord
is given, not the arc. All measurements are given in millimetres ; the standard being
so small, fractions are neglected.

The measurements given are defined as follows :—

1. Total length of skull: from centre of occipital condyle to anterior extremity of
premaxilla.

2. Length of cranio-facial axis: from centre of occipital condyle to tip of rostrum.

3. Length of basis cranii: from centre of occipital condyle to base of rostrum, 7. é.
to the centre of a line joining the anterior ends of the bases of the basi-
pterygoid processes.

It will be seen that the expression basis cranii is here used in a special sense;
the dimension chosen nearly corresponds with the length of the base of the skull
as measured from the condyle to the preoptic ridge. The entire basicranial
axis, 7. é. from condyle to junction of presphenoid and ethmoid, can only be got
at by bisecting the skull, and even then, owing to the ankylosis of the bones
named, cannot be determined with precision.

4. Length of roof of cranium: from the centre of the supraforaminal ridge to the
middle of a line joining the anterior borders of the lacrymals (preorbital
processes).

Hutton takes the naso-frontal suture as the anterior boundary of this line, but
in old skulls the suture is obliterated, and the posterior end of the premaxillary
groove which coincides with it is frequently obscure.

5. Width of cranium at paroccipital processes: the length of a straight line joining
the dorsal ends of the paroccipital processes, immediately below the supra-
tympanic ridges. Occasionally there is a sort of step where the supratympanic
ridge passes into the paroccipital process; the measurement should then be
taken below the step.

6. Width of cranium at squamosal prominences: length of a straight line joining
the most projecting portions of the squamosals.

7. Width of cranium at temporal fosse : length of the longest straight line joining
the right and left temporal fosse.

This measurement gives an indication of the width of the brain-case, the walls
of the skull being much thinner at the temporal region than elsewhere.

8. Width of cranium at postorbital processes: length of a straight line joining the
most prominent parts of the right and left postorbital processes.

9:

10.

fal:

12.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA, 407

Width of cranium at preorbital processes: length of a straight line joining the
posterior borders of the lacrymals.

Distance between temporal ridges: length of the shortest straight line joining
the right and left temporal ridges.

Height of cranium: length of a perpendicular from the highest part of the
cranial roof to the basitemporal platform.

Width of tympanic cavity: length of the longest horizontal straight line joining
the posterior temporal ridge and the edge of the paroccipital process.

. Width of temporal fossa: length of the longest horizontal straight line joining

the anterior and posterior limbs of the temporal ridge, above the junction with
the latter of the posterior temporal ridge.

. Width of orbit: length of the longest horizontal straight line joining the posterior

border of the lacrymal and the anterior border of the postorbital process.

. Distance between optic foramina: length of a straight line joining the lower

borders of the foramina.

. Greatest length of premaailla: from apex of body in a straight line to posterior

end of nasal process.

. Length of premaxilla to end of maxillary process: from apex of body in a straight

line.

. Length of body of premasilla: length of a straight line between apex and

posterior border of prenarial septum.

. Width of body of premazilla: length of a straight line between the right and

left alveolar borders at the level of the posterior edge of the prenarial
septum.

. Length of maxilloyjugal arch: the greatest length in a straight line.
. Length of vomer: the greatest length in a straight line.

22.
. Length of pterygoid: the greatest length in a straight line.

. Length of quadrate: length of a straight line from the articular surface of the

Length of palatine: the greatest length in a straight line.

head to the most prominent part of the internal condyle.

. Length of mandibular ramus: from middle of anterior border of symphysis to

extremity of posterior angular process.

. Greatest height of mandible: length of a perpendicular from the coronoid process

to the ventral border.

. Least height of mandible: a similar measurement of the anterior part of the

ramus a short distance posterior to the symphysis.

. Length of mandibular symphysis: length of a straight line passing from the

middle of the anterior to the middle of the posterior border of the symphysis.

. Width of mandibular symphysis: length of a straight line between the right and

left alveolar borders at the level of the posterior edge of the symphysis.
3M2

408 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Absolute measurements, as defined on pp. 406-407, of the various species examined,
are given in Table A.

But, as an aid to the determination and definition of the genera, the proportions of
the skull are more important than absolute measurements, a fact recognized by Hutton,
who gives (9, p. 107) the proportion between length and breadth, and between breadth
and height, in his eight genera. I have come to the conclusion, however, that a more
convenient method is to take asa standard the length of the basis cranii as defined
above (p. 406, § 3), and to express various other important dimensions as percentages.
In this way a number of indices are obtained, many of which are of great importance
in the definition of the genera: they are given in Table B.

Table C gives what may be called the temporal index; 7. e. the width of the
cranium at the temporal fossa as compared with the distance between the right and
left temporal ridges.

6. SUMMARY OF THE CRANIAL CHARACTERS OF THE RATITA.
STRUTHIO.

Occipital plane inclined backwards ; occipital condyle sessile; a broad occipital crest.

Length of cranial roof nearly three times that of basis cranii.

Width at paroccipital processes about one and a half times length of basis cranii.

Width at squamosals about double length of basis cranii.

Height of cranium nearly double length of basis cranii.

Temporal fossee continued mesiad on to parietal region; distances between temporal
ridges about one third less than width of cranium at temporal fosse.

Zygomatic process outstanding ; no squamosal prominence.

Width of orbit about equal to breadth of cranium at paroccipital processes, or one
and a half times length of basis cranii; interorbital septum present ; a projecting
supraorbital ledge notched in front and produced behind into a short blunt, post-
orbital process.

Lacrymal produced into a broad, backwardly-directed orbital process, and a strong
descending process passing mesiad of the lacrymal duct.

Nasal has a well-developed maxillary process, and is separated posteriorly from its
fellow by an interval in which the mesethmoid appears.

Ossified portion of antorbital does not extend outwards to lacrymal.

Premaxilla weak ; body flat, having no prenarial septum ; no palatine processes ;
width of body about equal to length of basis cranii.

Maxilla narrow; extends forwards to body of premaxilla; maxillo-palatine a
hatchet-shaped process given off from its mesial border; antrum small,

Trans. Zoo. Sqaillimetres). [To face p. 408.
kYX | Emrus | AprERYX | Drostzus Casuartus | Srrurnio Rura
parva. | species aut oe a. | — crassus. Species y. i eee nove- galeatus. || : camelus. americana.
pecimen. | 1 specimen. | 1 specimen. | hollandie.

ase seein ee | | oe a lesa

1. Total lengthy 93 | | 122 13410 ye. | 1860 yy ne 50 170 eS ||,

2. Length of et 969 | oa 110 102-112 | ae | 90 | 76 295 | 163 122)

pa Henschi of bi Si | 9235 37 37-41 40 18 29 29 33 30

4. Length of er 73 74 73-85 80-94 75 49 80 285 98 88

5. Width of eri 54 54 54-59 61-72 ? 67 39 43 53 50 48

6. Width of ers 63 | a 70-73 76-80 ||. “eB 57 60 65 53

7. Width of erj44 47 45-51 50-52 52 30 51 54 53 45

SyWadeh offerd73| | 80 79-85 84-86 80 a a il 7 85 62

9. Width oferd4, || 39 37-43 | 40-45 47 16 34 31 59 35

MO, Diseamcobshigg; 2) 42-49 41-50 50 a) BO 54 41 32
11. Height of ery 44 | 43 45-50 46-50 48 26 47 255 62 AT
12. Width of ty: 15 | 20 19-24 | Won 20 9 21 24 22 25
13. Width of tenog | 99 1623 16-28 17 15 TS wlle20 20 || 21
14. Width of ortog |, 30 30-81 29-838 32 eh ea ep | 46
15. Distance bet) 9 | 12 9-11 10 12 4 2 2 9 | 2
| 16. Greatest leng gg | ae, 80-88 Ae 144 80 2:90 109 | 95
17. Length of pri 5¢ . 54 60-66 a 2120 rae Noe 115 82
| 18. Length of boj9g oh 28-31 27-33 ae 5 23 24 28 29
19. Width of bodon = |, 28-31 37-41 se 4 | 25 | 14 35 | 24
20. Length of mazg9 | 67-72 | 69-73 * 277 STi es 139 «| 92
21. Length of vo || 46 49 33 36 Taher Waess 76 «|| @2
22. Length of pal _ e fe | ie i 231 2 il pail 287 nS
23. Length of pte ee 21-25 | 23-26 i 22) || RB | 232 2 40 | 33
24. Length of quizo9 | i 35 | 36-40 Ne | 12 | 25 26 | 31 | 25
25. Length of mi99 | ‘ VW | 129-197 183 | 141 | 165 184 | 173
| 26. Greatest heidgg | som | 90-22 toca agile FCAT wv | ww
P27 Wenstheisht 5) ill) enn NME were | 12-13 wel if ee ve 8 8 | 9
| | | \|
28. Length of mays, > EP aiming Wenicacn be Oe ee Sse || ay
| 29, Width of matyy | oe | 28-31 | 35-38 | 12 | 26 io | 94 35

ithouny f :
it mM } , y Nir,
wi oni,

Dil neat

Tine t\\ oA

anne

yi ts

pa ‘ pei a

Whos

ST ioe

‘cles

———ee

Trans. Zor, Soc. vol. xiii.) Table A.—Axsonure Measurements or THe SkuLrs or Ravirm (in millimetres). [bo face p
l | ie. Sook
Disornis | Pacnyonxis Mrsorrenys, | ANOMALOPTREYS. | ee Eweus Apreayx: Dnoweus Oasranius Stavrnio. Ri

seats iawn Gas | Rarantenl nes Species a. al camarina, | Species. || Species f. | Reese |) Cera | area } Bocieal By) Specicalay erasstis. Species y.[!/ australis, Nau galeatus. camelus. |) americana. |

1 specimen. | 4 specimens. | 2 specimens. | Sespecimens. | | specimen. 1 specimen. | 7 specimens. | 1 specimen. | 1 specimen. | 1 specimen. || 3 speciniens, | 1 specimen. 1 specimen. 1 specimen. jollandia. }
1. Total length of skull ............0.20.00e0005 296 195-205 | 2165-175 | 156 | 182 - 141 + 135-138 | 128 |e ere) 134-140 | | 186 150 170 188 173
2, Length of cranio-fucial axis ............-- 0... Mle ies: 2135 2 | 2 me 109 2 a 11 a 103-110 WO || oe 110 102-112 | 90 | 76 295 163 122
Biilienfeatoh basisternat| 4y 46 38-40 | BT-AL 241 38 27-83 226 36 33 3235 | 31 235 37 37-41 40 18 29 29 33 30
4, Length of cranial roof ...................--- 118 105 94-97 | 90-98 94 85 || 70-81 59 83 72 | 85-87 | 73 | 74 73-85 80-94 75 49 80 285 93 88
5, Width of cranium at paroccipital processes ...... 93 90-105 | si-s3 | 62-78 64 57 | 45-55 44 59 49 | 59-62 | 54 Odi) 5459 61-72 267 30 43 53 50 48
6. Width of cranium at squamosal prominences ....| 118 104-115 4 | 82-92 I ce = 52 76) Ge 70-74 | 63 | 70 70-73 76-80 75 || 38 57 60 65 538
7. Width of cranium at temporal fossw. 4) ag 65-82 59-63) ||| 52-57 50 50 | 48-50 41 48 a7) 42d | 44 | 47 45-51 50-52 52 30 51 54 53 45

| i} lk

8. Width of cranium at postorbital processes ...... 139 120-136 | 105-110 | 92-96 102 oo | BEE 61 90 69) 77-82 | 72 80 79-85 | 84-86 80 ee 66 70 85 62
9. Width of cranium at preorbital processes .... . .. 60 84 80 | 56-67 zon | 54 || ara 38 53 | 40 | a4) 39 a743 | 40-45 47 16 34 3L 59 35
10. Distance between temporal ridges..,........... | 58 46-59 40-46 37-44 35° | 35 49-49 37 i) aa 22-93 26 | 43 42-49 41-50 50 30 50 54 41 32
11. Height of cranium ... ea 49 50-61 48 || 51-55 752) | 51 || dics 235 47 41 || 44-46 | 44 | 43, 45-50 46-50 48 26 47 255 62 47
12. Width of tympanic cavity ..........2...0.... | 25 25-29 22-25 | 23-30 23 eo) || 7-21 18 21 17 16 5 || 20 19-24 17-21 20 9 21 24 22 25
13. Width of temporal fossa...................... | 32 28-37 23-31 26-32 | 38 29 13-17 16 23 13 28-30 26 20 16-23 | 16-28 17 1a 13 a || 20 21
TES WAL GR ORT, 5545 0.0099 Goacnvea90d0bubeS Was 35 34-36 Seay |i Si ug 25-31 24 BB || li Bal te 30 30-31 29-33 32 me 4 | 68 | 46
15. Distance between optic foramina . . 30-35 26-30 | 15-21 14 13 | 8-11 9 14 9 9-10 9 12 9-11 10 12 il 2 ) 9 2
16. Greatest leugth of premaxilla ................ | 157 110-129 | 108-110 81-90 | .. 09 Be 81 | 20 73-76 68 | ae 72 144 30 290 109 95
17. Length of premaxilla to end of maxillary process. .| Se 97 285-93 | 68-82 aa co 2 63 | 20 60-62 56 | 54 2120 71 93 115 82
18. Length of body of premaxilla ................ 82 60 249-62 || a743 |. ” 25 | 33) | : 31-35 | 23-31 | 5 23 24 28 29
19. Width of body of premaxilla 7 63-71 55-57 | 30-35 ‘a | 24 see i ee ec 30-35 2 |. 28-31 | oes 14 35 24
20, Length of maxillo-jugal arch ....... .......... oe 113 oe Hl O8 ne eS 70 ace 73 E 78 70? Be 67-72 | 60-78 on 27. 87 127 139 92
Eteeengtiiofvomers sete nese rane a: eee says Fe 40 eee i cs OS BR i 5 oo || a 3 ss z sf 46 | 49 a 36 74 36 76 «|| 62
2) Uengthiofipalating|..es.-.e.tee eves sees se | 8 = eS | oli ee | ae 80 oo 42 | a 43 a a | 3 231 | so | 287 | 29
PSacengthiofpterygoid! .....2<seve-eee. cesses) ee ES 3 s. Bs Wil ae S of) | jj satel see 2125 23-96 Bs 220 232 || 282 240 | 8B
24) Tengthiofiquadrate..-.....-...-.----+--++-0s ae 53 41 41 5 me 35 | at | bh |) BRR es 35 36-40 fs 12 | 25 26 31 25
25. Length of mandibular ramus.................. | on 179-208 | 163-164 | 141-155 1 ee os 117 88 | 125) : 127-135 in Il oe 11 122-197 ca (i) 1S} | ui | a5 184 173
26, Greatest height of mandible .................. 24 25-28 20-23 | 202216 |e mee (evo) | TG Wy +. i] 19-21 20 .. | 20-21 | 20-22 " 9 | 9 12 W7 12
27. Least height of mandible .... 19 16-20 13.49 | 11-12 a ah Hl 8B | a 10 | Pee | eiee) 12 . | was | 12-13 a 7 BH 8 8 9
28, Length of mandibular symphysis... ........ 29 27-31 | 17-20 a5 i ne I |] bo 16-20 23? Pe 14-19 16-20 ai i) 22 u 27
29, Width of mandibular symphysis .............. 52 47-52 35-42 | 24-25 28 fia a 2 eae | ol ? 23-25 Mi Hy se | BELG | 35-38 12 | 26 1z 24 25

AA

mec
i ee

ole
Avi

ntages of length of basis cranii).

[To face p. 408.

EUS APTERYX Drovaus Casvartus | Srruriio Ruea
F crassus. australis nOve- || galeatus. | camelus. iA americana.
| hollandie. | |
| | — | i—
100 100 100 100 | 100 | 100 |
| 204-247 272 275 ? 293 281 293
148-180 216 148 132) ee 1510 en elG0 |
185-216 183 196 206 196 176
121-137 166 175 | (186 160 150
121-132 || 144 162 | Rise) fA 156
51-70 83 44 68 60 70
78-89 155 141 || 160 153
200-237 800 275 2310 330 316
67-89 27 79 (| 82 | 84 96
100-110 22 86 48 106 80
132 200 255 296 230 206
121 yal 111 2141 263 96
309-335 1016 486 || 568 557 576
92-100 66 gl ve ee | 83
[ a
BUS APTERYX Dromxus Casuarius Srrurnro | Ruea
CrASSUS. iatratn NOve- galeatus. camelus, americanus.
hollandie.
100 100 100 100 100 100
100-120 100 100 110 130 140

eee
on

ie
res

an
Nannies

Trans. Zor. Soc. vol. xiii.) Table B.—PrincivaL Ditensions or Skunis or Ravine (expressed as percentages of length of basis cranii). [Lo face p. 408.

] I | | | ! | |
| Drxonsis | Pacuvonsis | Mesorrenyx || Axouatowrmays|| Eurvs Armays | Duowavs || Carvanros |) Stavmuo Beat
: SS SS 2 ee | ee es |= =| | IL
robustis. | torosts. elephantopus. || casuarina, Spevies 8. || didiformis. Species a. | crassits, australis nove- galeatus, || camelus. americana,
| i | | | fotlandia. | |
| | ] “| | _] _| Sm. || | |
—— —-- a la | | | | |
| F
A a 00 100 100 100 | 10) | 100 1 | 0
Length of basis cranii .........+-+.+-+-.05 | 100 100 100, 100 | 100 106 | | | 0, 00 | 10
i 228 4-245 || 230-25 2B 231-265 |) 211-290 | 204-247 272 275 2293, 2 29; |
Length of cranial roof ........2-2020 20200 228 | 234-245 || 280-260 230 | 231-265 11-229 | 47 7 |} 275) | | 28 3
ae 900-228 | 13 iB SIs | 145-159 CE 216 |} | 82 H
Width of craninm at parocoipital processes .... 200-228 | 151-195 || 148-166 163 | 181-184 145-15 148-180 16 | 148 | 1 151 160
Width of cranium at squamosal prominences ... 239-250 | |) 221-280 |) 200-214 | 211 208-218 | 189-197 | 185-216 SB) |) ao 205 aeuee 176
| | | | _ I) - | | }
Width of cranium at temporal fosse ........ , 14178 | 155-157 | 139-142 | 142-166 133 | 120-137 121-137 | 121-137 166 |) 175 | 186 | 160 150 |
499 20-19 9 97-15 ‘ 21-185 | 121-132 | ) c | o 4
Height of cranium........-+2+-00e200e0eee 117-132 | 120-128 | 129-137 || 127-151 130 | | 121-135 | 121-132 144 162 7189 187 156
ii 57. | 70-80 || 939-69 5p } 2 51~ 8¢ 7
Width of temporal fossa ..... -........... 71-80 57-81 70-80 39-62 63 | || 43-62 51-70 83 44 68 | 60 | 70
Width of orbit 76 $504 80-90 | 87-93 88 || s1-84 78-89 155 141 | ~=-160 153 |
Aa DURE SCOR ROONEY ABACRaeS |
Greatest length of premaxilla .............. 240-2802) 270-289 | 209-225 | 221 225} | 193. | 200-287 300 | 275 2310 | 330 316
ee ee ps \| Il oss an - |
Length of body of premaxilla .............. 130-147 ? 155 92-107 | 78 91 | || 75-84 67-89 | 79 82 84 I 96 |
Width of body of premaxilla............. Ge 137-150 | 75-87 | 75 || 7 | 7 | 75-84 } 100-110 22 | 86 | 48] 06 80
| |
- | || p | 49 || E | 3
Length of vomer . 5 oe se J oe co ee | Be || 124) 182 | 200 || 255 | 296 230 206
| | |
Length of palatine ..............--...- i re Re es: 6 | 134 ini || Tey Ww | im 2141 263° |) 96
|| | ] | adil | 2 | e —
Length of mandibular ramus .............. 390-452 | 407431 | 352-387 | 305 | BT 380-421 300 309-335 | 1016 436 568 || 557 || 576
| }
} = | ah AG = | = - PA 9 ry Q | 6 |
Width of mandibular symphysis ........ --..) 102-113 71 | 60-62 65 58 { 71 75-84 92-100 66 | 89 41 | 12 i 83 |
——— a —— E UL ae os a
‘Table C.—Trmporat Inpex or Raine.
z — —— ———— =
Dixonsts | Pacuyonsts Mesorrenyx Axes Exnus | Armamnx ]/ Duowses | Casvanws | Sravrmo | Ruma
robustus. torosus. | clephantopus. || casuarina. Species B. didiformis. | Speci Species a, crassils, | australis, | nove- galeatus, | camelus, americanus.
| } } hollandia.
te — = — If —— oe — —j | y
a i
Distance between temporal ridges ....-...-+ 100 100 | 100 100 100 100 | 100 100 | 100 I 100 100

I
Width of craninm at temporal fossw .....--- 140-150 140 130-140 100-110 110 180-200 100 100-120 |) 100 100 110 130 | 140

Sat Aen
hia Le hea
PN J

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID#. 409

Vomer more than double length of basis cranii; flattened in front, trough-like and
forked behind; articulates with mavxillo-palatines; connected with palatines by
ligament.

Palatine slender ; expanded posteriorly and ankylosed to pterygoid; about two and
a half times length of basis cranii.

Mandible weak ; symphysis flat, with very obscure median ventral ridge.

RuaEA.

Occipital plane inclined backwards; occipital condyle sessile; occipital crest
obscure.

Length of cranial roof nearly three times that of basis cranii.

Width at paroccipital processes about one and a half times length of basis cranii.

Width at squamosals less than double length of basis cranii.

Height of cranium about one and a half times length of basis cranil.

Temporal fossz continued on to parietal region; distance between temporal ridges
about one third less than width of cranium at temporal fosse.

Zygomatic process slender, outstanding ; no squamosal prominence.

Width of orbit about equal to width of cranium at paroccipital processes, or one and
a half times length of basis cranii; interorbital septum present; supraorbital
ledge narrow, not notched in front, produced behind into short blunt post-
orbital processes.

Lacrymal produced into pointed, backwardly-directed orbital process and curved
descending process which passes mesiad of lacrymal duct.

No maxillary process to nasal; mesethmoid appears on dorsal surface between
posterior ends of nasals.

Antorbital quite unossified in the skulls examined.

Premaxilla weak; body flat, with no prenarial septum ; long palatine processes;
width of body rather less than length of basis cranii.

Maxilla broad ; anterior end does not reach to body of premaxilla; maxillo-palatine
a broad, thin plate, perforated by small and variable apertures, and produced
dorsad into a narrow vertical plate, at the base of which is a vestige of the
antrum. é

Vomer about double length of basis cranii; flattened and deeply forked in front,
carinate and deeply forked behind; not ankylosed to palatines.

Palatine a broad, thin, curved plate, not ankylosed to pterygoid, and shorter than
basis cranil.

Mandible weak ; symphysis flat, with a median ventral ridge.

410 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

DroMavs.

Occipital plane inclined backwards ; occipital condyle sessile ; occipital crest obscure.

Length of cranial roof less than three times length of basis cranii.

Width at paroccipital processes less than one and a half times length of basis cranil.

Width at squamosals nearly twice length of basis cranii.

Height of cranium about one and a half times length of basis cranii.

Temporal fossa nearly vertical; distance between temporal ridges equal to width of
cranium at temporal fossz.

Zygomatic process strong and outstanding; bears facet for quadrate on the inner
surface of its distal end ; no squamosal prominence.

Width of orbit rather greater than width of cranium at paroccipital processes, and
about one and a half times length of basis cranii; interorbital septum present ;
supraorbital ledge narrow, not notched in front, produced behind into a short
postorbital process.

Lacrymal produced into a pointed, backwardly-directed orbital process, and a slender
descending process, which is perforated for the lacrymal duct.

Antorbital well ossified and ankylosed laterad with the descending process of the
lacrymal ; a deep Harderian fossa.

Maxillary process of nasal vestigial; mesethmoid appears dorsally between posterior
ends of nasals.

Premaxilla weak ; body flat, with a very short prenarial septum ; palatine processes
long, width of body rather less than length of basis cranii.

Maxilla rather broad ; anterior end does not reach to body of premaxilla; maxillo-
palatine thin and pocket-like, containing a wide antrum.

Vomer nearly two and a half times length of basis cranii; broad, flat, and longitu-
dinally grooved in front, narrow and deeply forked behind; ankylosed to ptery-
goids.

Palatine a thin, oblique plate, somewhat longer than basis cranii, articulating in
front with maxilla and behind with pterygoid.

Mandible weak ; symphysis flat with a narrow ventral ridge.

CASUARIUS.

Occipital plane inclined backwards; occipital condyle sessile; occipital crest
obscure.

Cranial roof produced into a large crest ; nearly three times length of basis cranii.

Width at paroccipital processes nearly double length of basis cranii.

Width at squamosals double length of basis cranii.

Height of cranium, not counting crest, nearly double length of basis cranii.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA., Al]

Temporal fossa nearly vertical; distance between temporal ridges about equal to
width of cranium at temporal fosse.

Zygomatic process narrow and outstanding ; no squamosal prominence.

Width of orbit considerably less than width of cranium at paroccipital processes,
and about one and a half times length of basis cranii; interorbital septum
present ; supraorbital ledge continued into cranial crest and produced behind into
a broad, nearly vertical, postorbital process.

Orbital process of lacrymal united with cranial crest, only its posterior end projecting
as a short, backwardly-directed process ; descending process notched for lacrymal
duct and sending off, ventrad of the latter, a short upwardly-directed process from
its lateral margin, which partly converts the notch into a foramen.

Antorbital well ossified and ankylosed laterad with descending process of lacrymal ;
a deep Harderian fossa.

No descending process to nasal; the specimen examined does not show whether the
mesethmoid is covered by the nasals, but it appears to be so’.

Premaxilla strong ; body high and strongly ridged, with a distinct prenarial septum ;
narrow palatine processes; width of body about half length of basis cranii.

Maxilla reaches nearly as far forward as prenarial septum; maxillo-palatine an
elongated cone with the apex directed forwards, and contains a spacious antrum
opening behind by a wide aperture.

Vomer nearly three times length of basis cranii; flattened and obscurely grooved in
front, deeply forked behind; ankylosed to palatines and pterygoids.

Palatine a short oblique plate, nearly one and a half times length of basis cranii,
articulating anteriorly with the maxilla and ankylosed posteriorly with the vomer
and pterygoid.

Mandible weak ; symphysis narrow, pointed, and keeled below.

APTERYX.

Occipital plane inclined at an angle of 45° to the basis cranii; occipital condyle
pedunculated ; occipital crest variable.

Length of cranial roof less than three times that of basis cranii.

Mamillar processes very large and prominent.

Width at paroccipital processes about double length of basis cranii.

Width at squamosals nearly double length of basis cranii.

Height of cranium less than one and a half times length of basis cranii.

Temporal fossa very wide and nearly vertical; distance between temporal ridges
equal to width of cranium at temporal fossz.

* See, however, 25, p. 428.

412 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Zygomatic process directed forwards and strongly compressed; no squamosal
prominence.

Orbit small and ill-defined; no interorbital septum, supraorbital ledge, or post-
orbital process.

Lacrymal ankylosed to aliethmoid and perforated for lacrymal duct; no orbital
process.

Antorbital not clearly marked off from rest of aliethmoid, which is strongly convex,
completely ossified, and extends backwards to the optic foramen.

Nasal has maxillary process ankylosed to lacrymal; mesethmoid appears dorsally
between posterior ends of nasals, but complete ankylosis of the bones in this
region takes place in the adult.

Premaxilla strong and greatly elongated; body short but high, with a very short
prenarial septum ; narrow palatine processes ; width of body not more than one-
fifth length of basis cranii.

Maxilla broad and flat; articulates in front with maxillary and palatine processes of
premaxilla ; maxillo-palatine represented by the thin mesial border of the bone ;
no bony antrum.

Vomer about twice length of basis cranii; flattened and narrow in front; broad,
keeled, and deeply cleft behind; ankylosed to palatines and pterygoids.

Palatine a thin, twisted plate with pedate posterior end; about one and three
quarters length of basis cranii; ankylosed in front and by its entire outer border
to maxilla and behind to vomer and pterygoid.

Mandible strong; symphysis occupies more than half its length, narrow, strongly
keeled below.

DINoRNITHID&.

Occipital plane vertical or very slightly inclined backwards or forwards; occipital
condyle pedunculate ; occipital crest variable.

Length of cranial roof from two to two and a half times length of basis cranii.

Mamillar tuberosities usually prominent ; basitemporal platform always well defined
and separated from occipital condyle by a deep precondylar fossa.

Width at paroccipital processes from less than one anda half to more than twice
length of basis cranii.

Width at squamosals from about one and three quarters to two and a half times
length of basis cranii.

Height of cranium about one and a quarter times length of basis cranii.

Temporal fossa extends mesiad to a greater or less extent on to parietal region ;
distance between temporal ridges varies from about width of cranium at temporal
fossee to half that width.

9

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID. 413

Zygomatic process short, pointed, and nearly parallel to median plane; auditory
region of skull produced into a strong squamosal prominence.

Width of orbit about half width of cranium at paroccipital processes, and almost
invariably less than length of basis cranii; interorbital septum absent or greatly
reduced ; a broad supraorbital ledge, produced behind into a strong, broad, post-
orbital process.

Lacrymal ankylosed with frontal, forming a preorbital process; no orbital process;
a descending process ankylosed with outer border of antorbital and notched or
perforated for lacrymal duct.

Mesethmoid produced into paired horizontal triangular processes.

Antorbital well ossified; ankylosed to descending process of lacrymal ; perforated
dorsally by a supraorbital fenestra of variable size.

Nasal either has a slender maxillary process, or there isa distinct maxillo-nasal bone ;
meets its fellow of the opposite side in the middle line above the ethmoid, so that
the latter does not appear on the dorsal surface; premaxillary groove on upper
surface of nasals extends backwards to or beyond naso-frontal suture.

Premaxilla strong ; body more or less elevated, and with a distinct prenarial septum ;
palatine processes broad and produced into more or less definite vomerine processes ;
width of body always more than half and sometimes one and a half times length of
basis cranil.

Maxilla short and narrow ; maxillo-palatine a short, flat plate, produced dorsad either
into an irregular shell of bone containing a large antrum, or into a thick, oblique
plate, containing no, or but little, trace of the antrum.

Vomer less than one and a half times length of basis cranii; consists of thin paired
plates meeting each other ventrad in an acute dihedral angle, and either quite
free or partially ankylosed with one another in front; firmly ankylosed behind, in
fully adult specimens, with palatines and pterygoids.

Palatine a thin twisted plate, about one and a fifth times length of basis cranii;
pedate posterior end produced into short mesial vomerine process; articulates at
anterior end with maxilla, and posteriorly with vomer and pterygoid, with which,
in fully adult specimens, it becomes ankylosed.

Mandible very strong; symphysis short, more or less flattened and ridged below;
distal end more or less deflected downwards.

7. THe CLASSIFICATION OF THE DINORNITHID A.

At an early period of his investigations—in 1846—Owen was led to the conclusion that
the differences between certain of the Moas were of more than specific value, and
instituted the genus Palapterya for the reception of species (ingens and dromioides) in
which the hallux was present. Further investigations, however, convinced him that
the retention of a purely vestigial structure was not, even if constant, of sufficient

VOL. xtt.—Part xI. No. 6.—October, 1895. 3.N

414 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

importance to distinguish a genus, and he therefore returned to his earlier practice of
placing all the species in the single genus Dinornis.

Reichenbach (26) was the first, in 1850, systematically to divide the family into
genera, proceeding upon the simple plan of erecting into a genus each of the seven
species known to him. His material must necessarily have been very imperfect, and
two of his genera (Moa and Movia) are undoubtedly synonyms of a third (Dinornis).

Von Haast (6) was the next, in 1873, to attack the problem. He divided the Moas
into two families, each containing two genera; but the characters upon which the
definition of the families was based were shown by Hutton (8) to be quite unreliable ;
many of the generic characters are incorrect and others inconstant ; and in at least three
instances, in the Canterbury Museum, the skull of one species was assigned to the
skeleton of another. Moreover, the brief account of the proposed classification, given
in a Presidential address, was never followed up by a detailed statement, and was
unsupported by measurements or figures. Under these circumstances the wide
acceptance of Von Haast’s views is rather remarkable: they are adopted without
remark by Wallace (28) in 1876, and with a qualifying note by Newton (14) in 1885;
and Lydekker (15) in 1889 discusses the question and comes to the conclusion that the
distinction between the two families is a valid one,—a decision which this author's later
enquiries (12) have led him to reverse. Fiirbringer (3) has a long discussion in his
usual judicial manner, and concludes that there is no evidence for the establishment of
more than one family.

Last year two classifications were propounded, unfortunately independently: one in
England by Lydekker (12), the other by Hutton in New Zealand (g). Both agree in
recognizing only a single family, which Lydekker divides into five genera—or four if
Megalapteryx be excluded—and nineteen species, Hutton into seven genera and twenty-
six species. In both schemes the definitions of certain of the genera are wanting in
exactness, especially as regards the skull, which Prof. Hutton rightly considers the most
important part of the skeleton for generic distinctions. Mr. Lydekker supplies valuable
corrections of many of Owen’s determinations, but he has only examined the British
Museum collection, which is evidently deficient in many important respects. Prof.
Hutton, on the other hand, besides examining the large public and private collections in
New Zealand, has himself collected Moa-bones in various parts of the colony, and was the
first to recognize the important bearing of the geographical distribution of the species.

The table on p. 415 gives a comparison of the arrangement of the Dinornithide by
the four authors referred to with that adopted in the present paper.

It will be seen that Haast, Lydekker, and Hutton are all agreed as to the limits of
Dinornis, and that Reichenbach’s Anomalopteryx corresponds with the similarly-named
genus of Hutton and with Haast’s Meionornis. Lydekker’s Anomalopterya includes
four of Hutton’s genera and part of a fifth, Hutton’s Huryapterya includes species
from two of Haast’s and from two of Lydekker’s genera, and so on.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDZA.

ReIcHENBACcH, 1850.

Von Haast, 1873.

Lyprxxer, 1891.

Moa
giganteus.

Movia

ingens.

Dinornis

struthioides.

Anomaloptery«
didiformis.

Cela
curta.

Syornis

casuarinus.

Emeus
crassus.

Dinornis
maximus.

robustus.
ingens.

gracilis.

struthioides.

Meionoi nis
didiformis.

casuarinus.

Euryapterye
rheides.

gravis.
Palapteryx

elephantopus.

My own observations have led me to the following conclusions, which, it must be
remembered, are founded mainly upon a study of the skull, and take no cognizance of

Dinornis
maximus.

novie-zealandize.

gracilis.
struthioides.

Anomalopteryx
dromioides.

didiformis.
parva.

geranoides.
curta.
owenl.

didina.
casuarina,

Emeus
crassus.

gravipes.
Pachyornis

elephantopus.
immanis.

species a.
b,

99

Hurron, 1891.

Parker, 1892.

Dinornis
altus.
maximus.
excelsus.
validus.
giganteus.
firmus.
robustus.
ingens.
potens.
gracilis.
torosus.
struthioides,

Palapteryx
plena.
dromioides.

Anomalopteryx
didiformis.

antiqua.
Cela
geranoides.

curta.

Mesopteryx
didina.

Syornis

casuarinus.

rheides.
crassus.

Euryapterya
gravis.

ponderosa.

elephantopus.

pygmeea.

Dinornis

maximus.

robustus.
potens (?).

torosus.
species a.

Anomalopteryx
didiformis.
parva (?).

Mesopter ya
species a.

species (3.
species y.

casuarina,

Emeus
crassus.
species a.

» PB

Pachyornis
species a.
elephantopus.
species .

” Y

several of Hutton’s and of one or two of Lydekker’s species :—

1. The tall, comparatively slender-limbed forms, with broad skull and long, wide,

3Nn 2

415

416 PROF, T, JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

deflected beak, constitute a very natural and highly specialized group. It may be
called the maximus-group, and, so far as I can make out, includesall the species placed
by Haast, Lydekker, and Hutton in the genus Dinornis, which name should therefore
be retained.

2. A second highly specialized or culminating group is constituted by heavy-limbed
forms with strongly-built narrow skull, short broad beak, and stout mandible. This
may be called the crassws-group: its type species is also the type of Lydekker’s Emeus.
T adopt this name.

3. The remaining species together form a comparatively generalized group, including
forms of small or moderate height and of varying bulk, having narrow skulls and
pointed beaks. This assemblage is divisible into three subdivisions :—

a. The elephantopus-group, characterized by a larger and relatively broader cranium
than is possessed by either of the following groups, large temporal fosse, and a wide
V-shaped mandible. It corresponds with Lydekker’s Pachyornis and with Haast’s
Palapteryx, and is placed under Euryapteryx by Hutton. The name Pachyornis should
be retained, Palapteryx having been wrongly applied by Haast.

6. The casuarina-group, so-called from its type species. The skull resembles that of
the elephantopus-group in general features, but is smaller and more delicate, and has
considerably smaller temporal fosse. This section includes Hutton’s Mesopteryx and
Cela, with part of Syornis ; it corresponds with Haast’s Meionornis, with Reichenbach’s
Syornis, and with Lydekker’s typical group of Anomalopterye. Caswarinus being the
type species, Reichenbach’s name has priority, but, as Lydekker has shown, it clashes
with Synornis of Hodgson. The group is certainly not congeneric with either the
preceding or the following one, and I therefore adopt Hutton’s name Mesopterya, the
type species of which (didina) is here considered to be a variety of casuarina}.

c. The didiformis-group, in which parva, if really distinct, should be included. It is
characterized by an unusually straight beak and immense temporal fossee. This section
corresponds with Anomalopteryx of Reichenbach and Hutton, doubtfully with Haast’s
Meionornis, and with Lydekker’s Celine group of Anomalopteryx. ‘There seems to be no
doubt that the name Anomalopteryx probably belongs to this group.

I see no evidence favouring the retention of the genera Palapteryx and Cela.
Hutton retains Palapterya for the reception of dromioides, Owen, and plena, Hutton,
the main ground of generic distinction being, as I understand, that the posterior view
of the skull of dromioides, as figured by Owen (17, pl. 54), shows an undoubted
approximation to that of Dinornis in its great relative breadth and short rounded
paroccipital processes. Lydekker, however, with the skull itself at his disposal, places
this species near casuarina. As to Cela, the type species is curta, of which oweni is,
according to Hutton, a synonym. Judging from Haast’s figure of a fragmentary skull,
I am disposed to agree with Lydekker in placing this species under Anomalopterya.

* Megalapteryx probably belongs to this group. See Note above, p. 378.—/June 1895.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA. AIT

. These conclusions may be summarized by stating that the Dinornithide are divisible
into three groups, two of highly specialized forms and one of more generalized forms,
which last may be again split up into three subgroups. These facts might be expressed
equally well by making three genera, and subdividing one of them into three sub-
genera, or by making three subfamilies, two of them containing a single genus apiece,
and the third including three genera. The latter appears to me the more convenient
method. Agreeing, as I do, with Hutton and Lydekker, that the differences between,
say, elephantopus and didiformis are of generic value, it appears to me quite clear that
those between robustus and elephantopus or between didiformis and crassus are
something more than generic.

Arranging the groups as nearly as possible according to their affinities as determined
by cranial characters, we get the following scheme :—

Family DINORNITHIDZE.

Subfamily a. DINORNITHIN &.

Genus 1. Drvornis.

Subfamily 6. ANOMALOPTERYGIN &.

Genus 2. PACHYORNIS.
, 3. MESOPTERYX.
4, ANOMALOPTERYX.

99

Subfamily c. EMEIN &,

Genus 5. EmeEvs.

8. SUMMARY OF CRANIAL CHARACTERS OF THE SUBFAMILIES AND GENERA
OF THE DINORNITHIDA.

Subfamily DINORNITHINA.

Width of cranium at paroccipital process nearly always more than twice length of
basis cranil.

Distance between optic foramina about two thirds length of basis cranii.

Orbit right-angled.

Greatest length of premaxilla more than two and a half times that of basis cranii.

Body of premaxilla deflected, bluntly rounded ; its length and width about one and
a half times length of basis cranii.

Maxillo-palatine contains a large antrum.

Nasal has a slender maxillary process.

418 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Mandible much deflected, with bluntly rounded symphysis, width of which is at least
three quarters of, and is usually more than, length of basis cranii.
Length of mandible more than four times that of basis cranii.

Genus DriNorNIs.

Occipital plane inclined forwards, condyle projecting beyond middle of supra-
foraminal ridge; occipital crest usually indistinct.

Paroccipital process short, and evenly rounded below; projects beyond occipital
condyle.

Width of cranium at squamosals from two and a third to two and a half times length
of basis cranil.

Mamillar tuberosities very large and prominent ; a large precondylar fossa.

Margin of tympanic cavity forms a continuous curve.

Temporal fossa large ; distance between temporal ridges about two thirds of width
of cranium at temporal fosse ; width of temporal fossa more than half length of
basis cranii: no mid-temporal ridge.

Temporal and lambdoidal ridges usually distinct, but occasionally in contact.

Posterior tympanic fossa wide; inferior temporal ridge prominent, usually stops
short of pretympanic process.

Zygomatic process very short.

Postorbital process angled, consisting of horizontal and descending portions.

Antorbital strong, deeply excavated above by a large supraorbital fenestra ; its ventral
edge forms a strong transverse postchoanal bar.

Prenarial septum very distinct.

Maxillo-jugal arch compressed, gently curved.

Quadrate has a rather long and compressed orbital process ; pneumatic foramen of
variable size, but usually very large, on mesial surface of otic process ; occasionally
two additional foramina on posterior surface.

Posterior angular process of mandible small or obsolete.

Subfamily ANOMALOPTERYGIN &.

Width of cranium at paroccipital processes less than twice length of basis cranii.
Distance between optic foramina usually less than one third and never much more
than one half length of basis cranii.
Orbit evenly curved, sinuous, or obtuse-angled (right-angled in Mesopterya,
species y).
Greatest length of premaxilla less than two and a half times that of basis cranii.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA, 419

Body of premaxilla narrow and pointed ; its length rarely and its breadth never
more than length of basis cranil.

Maxillo-palatine contains a large antrum.

Nasal has a slender maxillary process (qu. except in Anomalopteryx, where there is
a distinct maxillo-nasal bone 2).

Mandibular symphysis pointed ; its width always considerably less than three fourths
of length of basis cranii.

Length of mandible nearly always less than four times that of basis cranti.

Genus PACHYORNIS.

Occipital plane vertical or slightly inclined backwards; occipital crest prominent ;
supraoccipital fossee well marked ; anterior and posterior lambdoidal ridges enclose
a wide lozenge-shaped area.

Paroccipital process short, bluntly pointed.

Width of cranium at paroccipital processes from one and a half to nearly twice
length of basis cranii.

Width at squamosals more than double length of basis cranii.

Cranial roof strongly and evenly arched; an irregular shallow depression at base of
postorbital process.

Mamillar tuberosities large.

Margin of tympanic cavity evenly curved.

Temporal fossa large; distance between temporal ridges about one and a third
times width of cranium at temporal fossee; width of temporal fossa about three
fourths length of basis cranii; mid-temporal ridge small or absent.

Temporal and lambdoidal ridges may or may not be in contact.

Posterior temporal fossa narrower than in Dinornis and Anomalopteryx, but wider
than in Mesopteryx and Pachyornis; inferior temporal ridge strong ; pretympanic
process short.

Zygomatic process long, bluntly pointed.

Margin of orbit rather sinuous; postorbital process angled, consisting of horizontal
and descending portions.

Distance between optic foramina about one half length of basis cranii.

Antorbital very thin; supraorbital fenestra moderate; postchoanal bar strongly
curved outwards and forwards and more prominent than in any genus except
Dinornis.

Rostrum broad and flattened in about its posterior 15 mm., then moderately com-
pressed as far as the triangular processes, where it becomes broad and rounded
below (the anterior extremity is lost in all the specimens I have examined).

420 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Width of body of premaxilla nearly equal to its length, and about three fourths of
the length of the basis cranii.

Prenarial septum distinct.

Maxillo-jugal rather slender, compressed posteriorly, and strongly curved.

Orbital process of quadrate compressed and bluntly pointed; a single pneumatic
foramen of variable size on the mesial surface of the otic process.

Mandible strong, V-shaped, moderately deflected; posterior angular process of
moderate size.

Genus MESOPTERYX.

Occipital plane vertical or inclined backwards; occipital crest usually distinct ;
supraoccipital fossee present.

Occipital condyle often projects beyond paroccipital processes.

Paroccipital process pointed ; width of cranium at paroccipital processes about one
and a half times length of basis cranii.

Width at squamosal prominences double length of basis cranii.

Mamillar tuberosities small; posterior border of basitemporal platform nearly
straight.

Margin of tympanic cavity right-angled (evenly curved in species «).

Temporal fossa small; distance between temporal ridges about equal to width of
cranium at temporal fossee; width of temporal fossa about one half length of basis
cranii.

Temporal and lambdoidal ridges never confluent.

Posterior temporal fossa narrow ; inferior temporal ridge moderate.

Margin of orbit evenly curved or slightly smuous (right-angled in species y); post-
orbital process evenly curved.

Trigeminal foramen usually double (single in species /3), consisting of apertures
placed one above the other and separated by a narrow horizontal or oblique bar.

Distance between optic foramina about one third length of basis cranii.

Antorbital very thin; supraorbital fenestra rather small; postchoanal ridge very
thin or barely discernible.

Rostrum dilated but not flattened towards its posterior end, usually much compressed
and carinate in its intermediate portion, and slightly dilated between and in front
of the triangular processes.

Body of premaxilla narrow and pointed; its width nearly equal to its length and
about three fourths length of basis cranii; a distinct prenarial septum.

Maxillo-jugal arch slender and gently curved.

Orbital process of quadrate compressed and bluntly pointed ; a single pneumatic
foramen of variable size on mesial surface of otic process.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDZ. 421

Mandible slenderer than in any other genus, moderately dc flected ; posterior angular
process well developed.

Genus ANOMALOPTERYX.

Occipital plane distinctly inclined backwards, the condyle being hidden in a view
from above by the supraforaminal ridge and also in a view from the side by the
paroccipital process.

Occipital crest slight, but whole median supraoccipital region dilated to form a wide
transversely convex ridge.

Width at paroccipital process about one and three quarters length of basis cranii.

Anterior and posterior lambdoidal ridges usually widely separated.

Width at squamosals about double length of basis cranii.

Mamillar tuberosities large.

Margin of tympanic cavity sharply angled.

Temporal fossa very large; distance between temporal ridges about half width of
cranium at temporal fossee; width of temporal fossa more than three fourths
length of basis cranii.

Posterior limb of temporal ridge confluent with lambdoidal ridge; mid-temporal
ridge well marked and often prominent.

Posterior temporal fossa wide; inferior temporal ridge strong; no pretympanic
process.

Zygomatic process short and blunt.

Margin of orbit obtusely angled ; postorbital process evenly curved.

Distance between optic foramina less than one third length of basis cranii.

Presphenoid fossa unusually well defined.

Antorbital very thin; a large supraorbital fenestra; no definite postchoanal bar.

Lateral contour of premaxilla much straighter than in any other genus; body narrow
and pointed, its width about two thirds of its length and three fourths of length
of basis cranii; a distinct prenarial septum.

Maxillo-jugal arch stout and nearly straight.

Orbital process of quadrate long, compressed, and rather pointed at the tip; a large
pneumatic foramen on the mesial surface of the otic process and often a second
foramen on its posterior surface.

Mandible stout and nearly straight; symphysis narrow, with a moderately
prominent ventral ridge ; posterior angular process strong.

VOL. XIII.—PART xI. No. 7.—October, 1895. 30

422 PROF, T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Subfamily EMEIN &.

Width of cranium at paroccipital processes less than twice length of basis cranii.

Distance between optic foramina less than one third length of basis cranii.

Orbit evenly curved or slightly sinuous (right-angled in Pachyornis, species «).

Greatest length of premaxilla rarely more than twice, and never more than two and
a half times length of basis cranii.

Body of premaxille bluntly rounded in front; its length always less, its breadth
either less or slightly more than length of basis cranii.

Maxillary process of nasal represented by a distinct maxillo-nasal bone.

Maxillo-palatine devoid of an antrum.

Mandibular symphysis bluntly rounded ; its width at least three fourths of length of
basis cranil.

Length of mandible usually more than three and less than three and a half times
length of basis cranii.

Genus EMEus.

Occipital plane inclined backwards ; occipital condyle hidden in a view from above
by supraforaminal ridge, and in a view from the side by paroccipital process.

Occipital crest prominent ; supraoccipital fossee well marked.

Anterior and posterior lambdoidal ridges distinct, but close together.

Width of cranium at paroccipital processes usually about once and a half, but
occasionally (in H. crassus) once and three quarters length of basis cranii.

Width at squamosals rarely more than double length of basis cranii.

Mamillar tuberosities large.

Temporal fosse small; distance between temporal ridges but little less than width
of cranium at temporal fosse; width of temporal fossa usually about half length
of basis cranil.

Temporal and lambdoidal ridges do not touch; mid-temporal ridge usually well
marked.

Posterior temporal fossa narrow ; inferior temporal ridge strong.

Zygomatic process long, often bifid.

Postorbital process evenly curved, sometimes slightly inturned at distal end.

Antorbital thin ; supraorbital fenestra small; no definite postchoanal bar.

Nasal process of premaxilla very thick, its anterior end not flattened but rod-like;
prenarial septum often obscure.

Maxillo-jugal arch strong and nearly straight; maxillo-palatine gives off dorsally
an oblique vertical plate, with sometimes the merest vestige of an antrum at
its base.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDZ. 423

Orbital process of quadrate bluntly pointed; a large pneumatic foramen on mesial
surface of otic process.

Mandibular symphysis broader than long, with a very obscure ventral ridge;
posterior angular process better developed than in any of the other genera.

9. Tur PHYLOGENY OF THE Ratita '.

The most definite opinion I have met with as to the phylogeny of the Ratitz is that
expressed in the elaborate genealogical tree which illustrates Fiirbringer’s great work.
He ascribes a common origin to the Moas and Kiwis and to the Emus and Cassowaries,
but derives his four main groups of Ratite—the Struthioniformes, Rheiformes, Casuari-
formes, and Apterygiformes—separately from a primitive stock.

Mivart, in his memoir on the axial skeleton of the Ratits (13), gives no definite
opinion as to the phylogeny of the group, but his diagram illustrating the mutual
relationships of the various genera seems to indicate his belief in their monophyletic
origin. He shows a main stem dividing into two branches; one of these divides again
for Struthio and Rhea; the other forks a second time, one branch dividing again for
Casuarius and Dromeus, the other for Dinornis and Apterya.

The monophyletic origin of the Ratite is also supported by Newton, who, in his
luminous article “ Ornithology” (14), says “that these forms—Moa, Kiwi, Emu and
Cassowary, Rhea, and finally Ostrich—must have had a common ancestor nearer to them
than is the ancestor of any carinate form seems to need no proof.” Prof. Newton’s
classification indicates no closer affinity between any of the genera except the Emu and
Cassowary, which together constitute his order Megistanes; each of the other genera
has an order to itself.

A study of the skull certainly confirms the view that the nearest ally of the Dinor-
nithide is Apteryx, and that the four families of Australasian Ratitee are more nearly
related to one another than is either of them to the Asio-African and South-American
forms. Struthio and Rhea differ so much from the Australasian members of the sub-
class as to lend strong support to Fiirbringer’s view that they arose separately from a
primitive stock ; but whether the Cassowaries and Emus on the one hand and the Moas
and Kiwis on the other had a distinct or a common origin is a very complex question.

The main difficulty lies in deciding what characters should be considered as of
phylogenetic importance and what merely adaptive, but it appears to me that in the
following particulars the Emu and Cassowary show an undoub¢ed relationship to the
Moas :—

The general characters of the maxilla, maxillo-palatine, and antrum in both

genera.

1 As my conclusions are based upon a study of the skull, I have omitted all reference to dpyornis,

Dromornis, Megalapteryx, and Paleocasuarinus.
302

424 PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

The general relations of the vomer, palatines, and pterygoids in both genera.

The presence of a vestige of the maxillary process of the nasal in Dromeus.

The well-ossified antorbital ankylosed to the descending process of the lacrymal in

both genera.

The elevated body of the premaxilla with its distinct prenarial septum in Caswarius.

Dr. Forbes’s discovery (1) of a dinornithine bird which he calls Palwocasuarinus
will, if the detailed account of his very interesting researches bears out the opinions
expressed in his preliminary note, lend strong support to this view. The tibie upon
which the genus is founded have, as the name implies, a remarkable resemblance to
those of the Cassowary.

On the other hand I know of no character in the skull of Rhea by which it defi-
nitely approaches the Moas, and the presence of a maxillary process to the nasal, the
form of the cerebral fossx, and the position of the pneumatic foramen of the quadrate
seem the only particulars in which the Ostrich comes in any way near them. Struthio
and Rhea are, in fact, sharply separated both from one another and from the Australasian
Ratitee as well by the characters of the bony palate as by those of the pelvis. The
characters possessed by them in common with the other Ratite are of two kinds:
ancestral characters, such as the form of the vomer, the basi-pterygoid processes, and
the single-headed quadrate, which, according to the view taken in this paper, are
accounted for by the hypothesis of common descent from a group of generalized flying
birds or Proto-Carinate; and adaptive characters, such as those of the sternum,
shoulder-girdle, and wing, which they share to a greater or less degree with all flightless
birds.

The marked differences between the Moas and Kiwis are certainly for the most part
adaptive: the two families resemble one another in the increased size of the olfactory
organ and the reduced size of the eye; but both processes have gone so much further in
Apteryx that the differences between the two, in this respect alone, give their skulls
the appearance of being more widely separated than those of any other two ratite birds.
The real affinities underlying these differences are, however, shown by the striking
similarity of the bones of the palate in the two forms. The absence of a maxillary
antrum in Apteryx seems at first sight a difference of great importance, but the fact
that this cavity has disappeared or become vestigial in one of the most specialized
genera of Moas seems to indicate that its complete atrophy in the Kiwi is simply to
be looked upon as an instance of the extreme specialization of that genus.

As to the origin of the various genera of Dinornithide, I am not altogether in accord-
ance with Prof. Hutton (9, p. 428). I think there can be no doubt that Dinornis and
Emeus have diverged furthest from the ancestral stock, but in opposite directions ; and
that the narrow-beaked forms are the most generalized. Of the three narrow-beaked
genera, Mesopteryx appears to me to deviate least from the ordinary type of the
Ratitee, its comparatively lightly-built skull and slender mandible bringing it nearer

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA. 425

than any other genus to Dromeus and Casuarius'. On the other hand, the relative
size of the orbit is greatest in Anomalopteryx, and the presphenoid fossa or vestige of
the interorbital septum is most marked in that genus and in Dinornis: 1 am disposed,
therefore, to derive Mesopteryx and Anomalopteryx from a common ancestor.

Pachyornis appears undoubtedly to spring from the Mesopterya-stock: these two
genera are more neatly allied than any other two, Pachyornis being the more differen-
tiated in virtue of its greater bulk, broader skull, larger temporal fossze, more widely
separated optic foramina, and stronger beak.

Emeus is derived by Hutton from Mesopteryx, a view which I am strongly inclined
to adopt. The cranium of Hmeus undoubtedly comes nearest to that of Mesopteryz,

Dinornis
N
aie Si Emeus
Struthio ij focnyentis': rete w
at 0 eso &,
Duomaleus W v \ fe eryx yrery
I INS Ny
\ SS X VL

Fig. 1.—Phylogenetic diagram showing the mutual relations of the Ratite.

the differences between the two skulls depending mainly on the stronger and coarser
character of the whole skull, the broader beak, and the stronger mandible of Emeus. I
think, therefore, that the latter genus should be considered as springing from the
Mesopteryx-stock—not, of course, from Mesopteryx itself, but from an older member of
the line of descent which culminated in that genus.

Dinornis agrees with but goes beyond Pachyornis in its widely, separated optic

The striking resemblance of the dried head of Mesopteryx casuarina (Didornis didinus, Owen) to that of
an Emu is noticed by Owen (23).

PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

426

*uorjoafoad [eyuOZII0Y UI MMASeIp rvprMIs y—z “Sty

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA. A427

foramina and its broad skull, but in other respects—the form of the orbit, the length,
breadth, and strong deflection of the beak, &c.—is quite peculiar. I think that on
the whole it is reasonable to suppose this most specialized of the Moas to have sprung
from the common ancestor of the family independently of all the other existing genera.
It is, however, quite possible that future research may show Hutton to be right in
placing Palapteryx on or near the line connecting Dinornis with the generalized
ancestor of the group.

The accompanying diagrams (pp. 425, 426) express these views in a graphic form.
The first figure has the tree-shape adopted by Fiirbringer, which, after several trials, I
find more suitable to the present purpose than the usual straight-line diagram. ‘The
second figure shows the same thing in horizontal projection : the various genera of Dinor-
nithide are included in a tinted area indicating the limits of the family: the Dinornithide
and Apteryx on the one hand, and the Emu and Cassowary or. the other, are enclosed by
an even line indicating the limits of two groups, probably of subordinal value, including
respectively the New Zealand and the Australian forms: finally all these are enclosed
within a dotted line to show that the Australasian forms may be included in a natural
group, perhaps of ordinal value, clearly separated from the isolated Asio-African and
South-American genera.

These conclusions may be further expressed by a table of Classification as follows :—
Subclass RA TIT A, Merrem.

Order I. STRUTHIONES, Newton.
Fam. Srruraionipas. Genus Struthio.
Order IT. RHE, Newton.
Fam. Rueix. Genus Rhea.
Order III]. MEGISTANES, Newton.
Suborder 1. CasuarrFrormes, Firbringer.
Fam. 1. Casuartpa. Genus Casuarius.
2. Dromaipxz. Genus Dromeus.
Suborder 2. ApTHRYGIFORMUES, Firbringer.
Fam. 1. Dinornirurp2.
Subfam. a. Dinornithine. Genus Dinornis.
Subfam. 6. Anomalopterygine. Genera Pachyornis, Mesopteryx, Anomalopteryc’.
Subfam. c. Emeine. Genus Emeus.
Fam. 2. Apreryeipz. Genus Apteryx.

List oF WORKS REFERRED TO.
1. Forses, H. O. Preliminary Notice of Additions to the Extinct Avifauna of New Zealand.
(Abstract.) Trans. N. Z. Inst. vol. xxiv. (1891), p. 185.
2. Forses, H.O. Ona Recent Discovery of the Remains of Extinct Birds in New Zealand.
Nature, vol. xlv. (1892), p. 416.

» And probably Megalapteryx. See Note, p. 378.—June 1895.

428

co

20.
Die
Mr,
DB
24.
25s

20.

2

PROF, T, JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

. Firxsrincer, M. Morphologie und Systematik der Vogel. 1888. Amsterdam.
. Garrop, A. H. On the Value in Classification of a peculiarity in the anterior margin of

the Nasal Bone in certain Birds. Proc. Zool. Soc. 1878, p. 33.

. Haast, J. von. On the Measurements of Dinornis Bones, &e. Trans. N. Z. Inst. vol. i,

(1868), p. 21.

. Haast, J. von. Presidential Address. Philos. Inst. of Canterbury. Trans. N. Z. Inst.

vol. vi. (1873), p. 419.

. Haast, J. von. On Dinornis oweni. Trans. Zool. Soc. vol. xii. (1886), p. 171.
. Hurroy, F. W. Remarks on Dr. von Haast’s Classification of the Moas. Trans. N. Z.

Inst. vol. ix. (1876), p. 363.

. Huron, F.W. The Moas of New Zealand. ‘Trans. N. Z. Inst. vol. xxiv. (1891), p. 93.
. Huxtzy, T. H. On the Classification of Birds. Proc. Zool. Soc. 1867, p. 415.
_ Jaucur, G. Bericht iiber einen fast vollstindigen Schiidel von Palapteryx. Reise der

Novara, Palaontologie (1865), p. 305.

. Lyppxxer, R. Catalogue of the Fossil Birds in the British Museum. London, 1891.
. Mivart, St. G. On the Axial Skeleton of the Struthionide. Trans. Zool. Soc. vol. x.

(1871), p. 1.

. Newron, A. Art. Ornithology. Encyel. Brit. 9th ed. vol. xviii. (1885), p. 2.
. Nrcnoxson, H. A., and Lyprxxer, R. Manual of Paleontology. Edinburgh. 1889.
. Owen, R. On Dinornis: Pt. 2 (1846). Trans. Zool. Soe. vol. ii. p. 307 (Ext. Birds of N. Z.

p. 115).

. Owrn, R. On Dinornis: Pt.3 (1848). Trans. Zool. Soe. vol. iii. p. 345 (Ext. Birds of N. Z.

p. 179).

. Owen, R. On Dinornis: Pt. 5 (1850). Trans. Zool. Soe. vol. iv. p. 59 (Ext. Birds of N. Z.

p. 205).

. Owen, R. On Dinornis: Pt. 9 (1864). Trans. Zool. Soe. vol. v. p. 3387 (Ext. Birds of N. Z.

p- 151).

Owxn, R. On Dinornis: Pt. 14 (1869). Trans. Zool. Soc. vol. vi. p. 123 (Ext. Birds of
N. Z. p. 262).

Owen, R, On Dinornis: Pt. 21 (1875). Trans. Zool. Soc. vol. x. p. 147 (Ext. Birds of
N. Z. p. 427).

Owen, R. On Dinornis: Pt. 23 (1882). Trans. Zool. Soc. vol. xi. p. 233.

Owen, R. On Dinornis: Pt. 24 (1882). Trans. Zool. Soc. vol. xi. p. 257.

Parker, T. J. On the Anatomy and Development of Apferyx. Phil. Trans. vol. clxxxii. B
(1891), p. 25.

Parker, W. K. On the Structure and Development of the Skull in the Ostrich Tribe.
Phil. Trans. vol. elvi. (1866), p. 118.

Reicuensacu, H.G. L. Das natiirliche System der Vogel. Dresden. 1852.

. SrrznKa, E. Bronn’s Thierreich: Aves.
28.

Wattacr, A. R. Geographical Distribution of Animals, 1876.

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHID. 429

EXPLANATION OF THE PLATES.

PLATE LVI.

Figs. 1-4. Four views of a perfect skull of Hmeus, species «, in the Otago University
Museum. Natural size.

PLATE LVILI
Figs. 5-8. Outline sketches of figs. 1-4, with the various bones distinguished by colour.

PLATE LVIII.

Fig. 9. Back view of the skull shown in Plate LVI.

Fig. 10. Cranium of Emeus, species a, in horizontal section.

Fig. 11. The same in sagittal section.

Figs. 12 & 13. Two views of an immature cranium of Anomalopteryx didiformis,

Owen.
All natural size.
PLATE LIX.
Figs. 14-18. Outline sketches of figs. 9-13, with the various bones distinguished by
colour.
PLATE LX.

Fig. 19. Skull of Mesopteryax casuarina, in Dr. H. O. Forbes’s Collection.
Figs. 20 & 21. Two views of the skull of MMesopteryx, species 6, in the Colonial
Museum, Wellington.
Both natural size.
Fig. 22. Skull of Pachyornis elephantopus, in Dr. H. O. Forbes’s Collection.
Five-sixths natural size.

PLATE LXI.
Figs. 23-34. Outlines of the crania of various species of Dinornithide, from above.
Drawn to the same absolute size with the camera lucida.
Figs. 35-46. A similar series of outlines from the left side.

PLATE LXII.

Figs. 47-58. A similar series of outlines from behind.
Figs. 59-64. The maxillo-jugal arch (maxilla only in Rhea) of various Ratiti.
Natural size.
VOL, XIII.—PART x1. No. 8.—October, 1895. 3P

430 . PROF. T. JEFFERY PARKER ON THE CRANIAL OSTEOLOGY,

Figs. 65-70. The quadrate facet on the roof of the tympanic cavity, in various Ratitz.
Drawn to the same absolute size.
als., alisphenoid ; pr.o., prootic; sg., squamosal ; ex.oc., exocci pital.
Figs. 71-76. The optic and adjacent foramina in various Ratite. Drawn to the same
absolute size.
li, optic foramen; iii, oculomotor foramen; iv, pathetic foramen;
v!, orbitonasal foramen; vi, abducent foramen; a, foramen for internal
ophthalmic artery.

List of ABBREVIATIONS.

a.b.cr.fon. Position of anterior basicranial | Hard.fos. Harderian fossa.
fontanelle.
abd.for. Abducent foramen. inf.al.eth. Inferior aliethmoid.
a.lamb.r, Anterior lambdoidal ridge. | inf.orb.for. Inferior orbital foramen.
al.n. Alinasal. inf.temp.r. Inferior temporal ridge.
al.sph. Alisphenoid. int.aud.m, Internal auditory meatus.

ang. Angular.
a.orb, Antorbital.

ju. Jugal.
art. Articular.

lac. Lacrymal.
6.0c. Basioccipital.

b.pt.pr. Basipterygoid process.
6.sph. Basisphenoid.
6.temp. Basitemporal.

b.temp.pl. Basitemporal platform.

lac.for. Lacrymal foramen.

mam.tub. Mamillar tuberosity.
mesen.fos. Mesencephalic fossa.
mes.eth. Mesethmoid.

m.temp.r. Mid-temporal ridge.
car.for. Carotid foramen. Maxilla.

mz.
cer.fos. Cerebellar fossa.

maju.ar. Maxillo-jugal arch.
con.for. Condylar foramen or foramina.

max.na. Maxillo-nasal.

mz.pal. Maxillo-palatine.
d. Dentary. mex.pr. Maxillary process.

dor.sell. Dorsum selle.

na. Nasal.
eth. Ethmoid. na.pr. Nasal process.
eus.gr. Kustachian groove.

ex.oc. Exoccipital. oc.con. Occipital condyle.

oc.cr. Occipital crest.
for.a. Foramen a. oc,for. Oculomotor foramen.
foss.b. Fossa 6. , olf.ch. Olfactory chamber.
Jr. Frontal. opfor. Optic foramen. —

CLASSIFICATION, AND PHYLOGENY OF THE DINORNITHIDA,

op.pl.
orb.na.for.
orb.pr.

or sph.
ot.pr.

pa.

pal.
pal.pr.
par.oc.pr.
path for.
p.b.cr.fon.

pit.fos.
p-lamb-r.
p.M2z.
post.orb.pr.
post.temp.fos.
post.temp.r.
pre.con.fos.
pre.orb.pr.
pre.temp.w.
pr.lacer fos.
pr.op.r.
pr.ot.
pr.pit.r.
pr.sph.

pr. sph.foss.
pt.

Optic platform.
Orbitonasal foramen.
Orbital process.
Orbitosphenoid.

Otic process.

Parietal.

Palatine.

Palatine process.

Paroccipital process.

Pathetic foramen.

Posterior basicranial fonta-
nelle.

Pituitary fossa.

Posterior lambdoidal ridge.

Premaxilla.

Postorbital process.

Post-temporal fossa.

Post-temporal ridge.

Precondylar fossa.

Preorbital process.

Pretemporal wing.

Prelacerate fossa.

Preoptic ridge.

Prootie.

Prepituitary ridge.

Presphenoid.

Presphenoid fossa.

Pterygoid.

q-
qJ-

rost.

S.ang.
.0C.
s.0€.fos.
sq.
sq.-prm.
sup.al.eth.

sup.for.r.

sup.orb.fen.

temp.fos.
temp.r.

tent.r.
trig.for.

tri.pr.
turb.

vag.for.

vo.
vo.pr.

2yg-pr

Quadrate.
Quadrato-jugal.

Rostrum.

Supra-angular.
Supra-occipital.
Supra-occipital fossa.
Squamosal.

Squamosal prominence.

Superior aliethmoid.
Supra-foraminal ridge.
Supra-orbital fenestra.

Temporal fossa.
Temporal ridge.
Tentorial ridge.
Trigeminal foramen.
Triangular process.
Turbinal.

Vagus foramen.
Vomer.
Vomerine process.

Zygomatic process.

431

5g, . Prits-

Par.oc. pre"

Lane. Eh.

J.Thomson. to.
M.P.Parker, we

HMEUS

post .orb .pr.
_--post.orb. pr. 5

\

\ > “N
mer. Ab Pay. 0c. pr.

i 0e.con
pricom. Fos

inf. team r.

1

OSE. E. . Fos
PR oe Fos

West, Newman, inp,

Sp. a.

Perey

Prams Loot. Soe. VE au PL

-- ---post.orb [apa eS

a.b.cr fon.--

}-1-/.
LE

con. For.

Hy

op. for. fora. Fos.b. 4
e772, For. : “| Hard, fos

sup orb. fer

T.d-P. delt . 5
MPParker Chromo, West Newman imp.

WES Bye: Ce

Irans. Toot. Soc. Vol MIL F¢, LVUL

post orb PP

par. oc pr

f
f
'
H

0c. com. mam. titb

ie temp ia

| ttrig. for:
‘post. Lemp. fos.

mor. Lub,

JThomson. .photo. West, Newman imp.
MP.Parker. lith.

Pig sa 2 i avery Ss) SONGe,
Figs .la&13, ANOMALOPTERYX DIDIFORMIS.

Frans Locl toc Vel ta LE IIX.

Big. 14.

_---post. arb. pr.

fib ==

Come

orb. na, for.

brig. for. \

He Aneesislere \ _ felay sible
! forP' \ i
mau tubs" inf orb. for.

Weat, Newmam imp

Fifs. 14-16 EMEUS 8p. mW.
Figs.17.218, ANOMOLOPTERYX DIDIFORMIS

; Sop EI MGI.

Sh Riga tes = West ,Nevwman.imp
4 Eve. 29), MES OP Wh Fux CASUARINA Owen. Pigs, 20-21, ME SOPTERYZE< SV. Pp. ; , ;

Big, 22, PACHYORNIS BLEPHANTOPUS, Owen.

nf
)

eR

ets

Trans. Loot. Soc Wel AIUMF¢V.LXT.

D .torosus, Hutt. A. didiformis,Ow. P. elephantopus. Pachyornis, sp. a.

Mesopterye, sp. B

33.

D, torosus, Hutt. A. didiformis, Ow. M.casuarine , Ow.

Pach .elephantopus, Ow. ; Mesopteryx sp. p . FE. crassus,Ow.

TdIP., deit 3 ¢ ! ’
MP Parker lith. West, Newman imp.

OUTLINES OF CRANIA OF DINORNITHID-A

et

Trans. Loot. Soc Wel AM PE LXM.

D. torosws. Hitt.

M. casuarinca, Ow.

55.

4 Hamilton & TIP. del
MPParker. lith

A. cidiformis, Ow.

52.

Pach . elephantopus, Ow.
58.

li Riv iii? ©

Casuarius. Apterye

ch »v1

Dromaeus Dinornus

West, Newrian imp

Fiégs.47-58, OoUTLINES OF CRANIA OF DINORNITHIDE.

”

HO = 16,12Ar

Rue Ss)

OF

SKULL

Om VARIOUS) FAI AS.

433

LIST OF THE PAPERS CONTAINED IN VOL. XIII.

Bepparp, Frank E., M.A., F.R.S., F.R.S.E.,
F.Z.8., Prosector to the Society.
Contributions to the Anatomy of the
ABM VOC! AGNI 555006000000000000
Bett, F. Jerrrey, M.A., Sec. R.M.S., F.Z.S.,
Professor of Comparative Anatomy and
Zoology in King’s College, London.
Contributions to our Knowledge of the
Antipatharian Corals .............-
Description of a remarkable new Sea-
urchin of the Genus Cidaris from Mauri-
tius

Bovutenser, G. A., F.R.S., F.Z.S8.

Catalogue of the Reptiles and Batrachians
of Barbary (Morocco, Algeria, Tunisia),
based chiefly upon the Notes and Collec-
tions made in 1880-1884 by M. Fernand
atbastes ape iets merkeasetcieud:cystelsuclepate

On Remains of an Extinct Gigantic Tortoise
from Madagascar (Testudo
dieri, Vaillant)

Gavow, Hans, Ph.D., M.A., F.R.S., F.Z.8.
On the Remains of some Gigantic Land-

grandi-

BaCnCeCEONC SCION nC nC ECnC nt aC nCnr CnC)

Tortoises, and of an Extinct Lizard,

recently discovered in Mauritius

Gapow, Hans, Ph.D., M.A., F.R.S., F.Z.8., and

Newton, Sir Epwarp, K.C.M.G., F.L.S.,
C.M.Z.S.

On additional Bones of the Dodo and other

Extinct Birds of Mauritius obtained by

Mr. Théodore Sauzier .... .........
Greeory, J. W., B.Sc., F.Z.S., British Museum
(Nat. Hist).

On the British Paleeogene Bryozoa......

VOL. XII.—PaRT xt. No. 9.—October, 1895.

87

303

93

305

313

281

; Page
Hickson, Sypnuy J., M.A. Cantab., D.Sc. Lond.,

F.Z.8., Fellow of Downing Coll. Camb.
A Revision of the Genera of the Alcyonaria
Stolonifera, with a Description of one
new Genus and several new Species .. 325
Newron, Sir Epwarp, K.C.M.G.,F.L.S.,C.M.Z.S.,
and Gapow, Hans, Ph.D., M.A., F.R.S.,
E.Z8.
On additional Bones of the Dodo and other
Extinct Birds of Mauritius obtained by
Mr. Théodore Sauzier
Newton, E. T., F.G.S., F.Z.S., Geol. Survey.
On a Skull of Trogontherium cuviert from
the Forest Bed of East Runton, near
Cromer
Parker, T. Jurrury, D.Sc., F.R.S., C.M.ZS.,
Professor of Biology in the University
of Otago, Dunedin, New Zealand.
On the Cranial Osteology, Classification,
and Phylogeny of the Dinornithide .. 373
Parker, W. Kircunn, F.R.S., F.Z.8.
On the Morphology of a Reptilian Bird,
Opisthocomus cristatus .........4.44-

281

165

43
Rosertson, Davin, F.L.8., F.G.8., and Sressine,
Rey. THomas R. R., M.A.
On four new British Amphipoda
Srreppine, Rev. THomas R. R., M.A.
On the Genus Urothoe and a new Genus
Grothovdesmermeereree ee
Descriptions of nine new Species of Amphi-

31

podous Crustaceans from the Tropical

Atlantic

Srepsine, Rev. THomas R. R., M.A., and
Rosertson, Davip, F.L.S., F.G.S.

On four new British Amphipoda ..

3Q

a :
ae : heat be id 4 “a Rein (

3 Wait Wely

a TT

Laan

(oxatly AR NY |

hyp aeteai

ee Ah eee

bias

bt

, ae a) a

i i ed » ‘
ONY ENT ea

INDEX OF SPECIES, ETC., IN VOL. XIII.

Acanthodactylus bedriagai, 103.

boskianus, 94, 96, 100, 129, 130.

, var. asper, 129.
lineomaculatus, 94, 100, 131.
pardalis, 94, 96, 101, 103, 129, 131.
, var. bedriage, 131.

, var. lineomaculatus, 132.

savignyi, 131.

scutellatus, 96, 100, 129, 130, 131.
, var. exiguus, 130.
vulgaris, 96, 100, 129, 131.
Adeona grisea, 241.

(Dictyopora) cellulosa, 242.
Adeonella atlantica, 244.

dispar, 244.

distoma, 241, 245.

—— ——, var. imperforata, 245.
fuegensis, 244.

intricaria, 244.

japonica, 245.

pallasi, 244.

pectinata, 223, 244,

—— platalea, 244.

polymorpha, 241, 242, 244,
polystomella, 244.

regularis, 244.

sparassis, 245,

sulcata, 244.

tuberculata, 245.

Adeoneilidi, systematic position of the, 241.
Adeonellopsis coscinophora, 245, 246.
distoma, 246.

—— heckeli, 245.

amperforata, 245.

incisa, 247, 263, 277.

perforata, 247.

Agama agilis, 100 117.

wetherelli, 245, 246, 247, 263, 277, 278.

Agama bibronii, 96, 100, 117, 118, 163.
colonorum, 100, 118.

—— wmermis, 94, 96, 100, 117, 118.
—— mutabilis, 117.

—— ruderata, 117.

savigny2, 94.

—— tournevillii, 96, 103, 117, 118, 163.

Alcyonaria, classification of the, 328.

Alcyonaria Stolonifera, a Revision of the Genera
of the, with a description of one new Genus
and several new Species, by Sydney J. Hickson.
325-347.

, list of the literature upon the, 344-346.

Aleyonium coralloides, 336.

Algira barbarica, 128.

microdactyla, 103.

Amphipoda, on four new British, by T. R. R. Stebbing
and D. Robertson, 31-42,

Amphipodous Crustaceans, descriptions of nine new
Species of, from the Tropical Atlantic, by
T. R. R. Stebbing, 349-371.

Amphisbena cinerea, 99.
Y} ’

elegans, 99, 122.

Anas bernieri, 291.

melleri, 291, 292.

—— punetata, 291.

theodori, 282, 291, 292, 300.
Anguis fragilis, 98, 120.
punctatissimus, 98,

Anomalopteryx antiqua, 415.

casuarina, 415.

curta, 415.

—— didiformis, 378, 383, 385, 387, 389, 390, 395,
397, 415, 428.

—- didina, 415.

——- dromioides, 415.

gerenoides, 415.
owent, 415.

319) 2

436

Anomalopterya parva, 379, 415.

Anser cinereus, 290.

Anthelia capensis, 333, 336.

desjardiniana, 333, 336.

—— domuncula, 336.

—— filippii, 333, 336.

— glauca, 333, 336.

olivi, 336.

— purpurascens, 333, 336.

rubra, 336.

strumosa, 333, 336.

Anthropoid Apes, contributions to the anatomy of,
by F. E. Beddard, 177-218.

, introductory remarks upon, 177-201.

Anthropopithecus calvus, 177.

Antipatharian Corals, contributions to our knowledge
of the, by F. Jeffrey Bell, 87-92.

Antipathes robillardi, 91, 92.

Antipathid from the neighbourhood of Mauritius,
91.

Aphanapteryx broecki, 282, 293, 294, 301.

Apteryx australis, 398, 404, 405.

oweni, 404.

Aptornis defossor, 76.

Ardea garzetta, 282.

—— (Butorides) nigricollis, 289.

—— (Nycticorax) megacephala, 289.

Asio capensis, 287.

Astacus crassicornis, 351.

Astur alphonsi, 282, 285, 299.

—- melanoleucus, 285, 286.

sp., 282.

Athene murivora, 287.

Aulopora tenuis, 333, 334.

Barbary, catalogue of the Reptiles and Batrachians
of, by G. A. Boulenger, 93-164.

——., distribution of the Reptiles and Batrachians

of, 96.

, natural divisions of, 94.

Beddard, F. E. Contributions to the anatomy of the
Anthropoid Apes, 177-218.

Bell, F. Jeffrey. Contributions to our Knowledge of

the Antipatharian Corals, 87-92.

Description of a remarkable Sea-urchin of
the genus Cidaris from Mauritius, 303.
Bernicla brenta, 290, 291.

Biflustra ecocena, 228.

INDEX.

Biflustra macrostona, 232.

(Membranipora) eocena, 228.

Biselenaria offa, 235, 236, 263, 277.

placentula, 234, 236.

Black Coral of the Mediterranean, observations on a

particularly fine example of, 87.

Blanus cinereus, 96, 121.

Boulenger, G. A. Catalogue of the Reptiles and
Batrachians of Barbary (Moroceo, Algeria,
Tunisia), based chiefly upon the notes and
collections made in 1880-84 by M. Fernand
Lataste, 93-164.

——. On Remains of an extinct Gigantic Tortoise

from Madagascar (T'estudo grandidieri, Vaillant),

305-311.

Bryozoa, on the British Paleogene, by J. W. Gregory,

219-279.

, affinities of the fauna of the British Palzogene,

264, ”

, bibliography of the British Paleogene, 266.

, classification of the British Paleogene, 221.

——,, miscellaneous records of the British Paleogene,

221.

, Stratigraphical distribution of the British

Paleogene, 263.

, systematic synopsis of the British Paleogene,
225.

, terminology of the British Paleogene, 220.
Bubo madagascariensis, 287.

virguianus, 287.

Bufo arabicus, 99, 158.

boulenger, 158.

—— mauritanicus, 97, 98, 99, 102, 158, 159.
pantherinus, 102, 158.

variabilis, 94, 100, 158.

viridis, 95, 97, 100, 103, 158, 159.
vulgaris, 97, 102, 103, 158, 159.

sp., 98.

Butorides mauritianus, 282, 289, 290, 300.
Cacatua galerita, 283, 284.
Calyptorhynchus funereus, 283, 284.

Carine murivora, 286.
Carpophaga pacijica, 298.
Casuarius galeatus, 398, 403, 404.
Catenicella amphora, 222.

pulchella, 222.

utriculus, 222.

INDEX. 437

Cela curta, 415.
geranoides, 378, 415.
Cellaria beyrichi, 241.
macrostoma, 232.

michelini, 256.

Cellepora, sp., 202.
globularis, 252.
gontostoma, 240.

— gracilis, 237.
heckeli, 245.
imbricata, 245.

—— multiradiata, 252.

petiolus, 253.

pumicosa, 252.
— quadrata, 234.

schizogaster, 243.
scrobiculata, 252.

(Discopora) koninckiana, 239.
Cerastes cornutus, 97, 99, 155.

vipera, 97, 101, 155, 164.
Chetura caudacuta, 71.

Chalcides lineatus, 96, 104, 138, 139, 140, 164.
mauritanicus, 97, 99, 138, 141.
— mionecton, 96, 138, 140.

— ocellatus, 96, 98, 138, 141.

, var. polylepis, 96, 139, 164.
, var. tiligugu, 96, 139.

, var. vittatus, 96, 139, 164.
sepoides, 97, 100, 188, 141, 142.
tridactylus, 96, 98, 138, 140.

Chameleon africanus, 142,

cinereus, 142.

saharicus, 142.

vulgaris, 94, 97, 98, 142.
Chauna chavaria, 62, 73, 76.
Chelone midas, 321.

viridis, 53.

Chrysisina coronopus, 259.
Cidaris curvatispinis, 304.
Cistudo europea, 100, 105.

lutaria, 105.

Clavularia alba, 336.

arctica, 335.

australiensis, 333, 335, 338, 339, 341, 346,
347.

bathybius, 336.

borealis, 335.

Clavularia capensis, 333, 336.
catenata, 332, 336.
— celebensis, 333, 335, 342, 346.

—— colinabwuin, 336.

—— concreta, 335.

—— crassa, 335.

—— cylindrica, 335.

— dane, 332, 336.

—— desjardiniana, 333, 336.
—— elongata, 335.

—— filiformis, 332, 336.
—— filippii, 333, 336.

| —— flava, 335, 341, 347.

frigida, 335.

garcie, 333, 335, 341, 342, 346, 347.
glauca, 333, 336.

magelhenica, 335.

margaritaceum, 336.

marioni, 335.
modesta, 333, 336.
ochracea, 335,

petricola, 335.

—— primula, 332, 336.

purpurascens, 333, 336,

ramosa, 335, 340, 347.

reptans, 835, 342, 346.

TOSCH, B3D.

ruse, 335.

stormi, 335.

strumosa, 333, 336.

tenuis, 334,

—— thalassantha, 332, 336.

tubaria, 335.

violacea, 335.

—— viridis, 326, 328, 329, 333, 335, 342, 343,
344, 346, 347.

Clemmys leprosa, 96, 97, 105, 106, 314.

sigris, 314.

Clydonia gracilis, 350, 351.

longipes, 350, 351, 356.

Ceelopeltis lacertina, 97, 98, 151.

monspessulanus, 151.

producta, 97, 100, 151.

Coluber esculapii, 99.

agassiztt, 98.
austriacus, 99.

—— hippocrepis, 98.

438 INDEX.

Coluber monspessulanus, 98.

natria#, 98, 100.

Columba livia, 298.

Conescharellina clithridiata, 252, 254, 263, 278.
rosula, 252.

stoliczkai, 252.

Conodontes boisvillettz, 166, 171, 172, 174, 175.

Corallium rubrum, 332.

Cornularia aurantiaca, 334.

cornucopia, 334,

crassa, 335.

multipinnata, 334.

—— subviridis, 334.

Cornulariella modesta, 333.

Coronella amalice, 97, 101, 144, 145, 150, 164.
austriaca, 101, 144.

brevis, 149, 150.

girondica, 97, 102, 103, 144, 145, 150.
— levis, 99.

—— (Macroprotodon) brevis, 94.

—— ( ) cucullata, 94.

Corvus frugilegus, 48.

Corythaix buffoni, 56, 62, 73, 76.

Crax globicera, 56, 57, 62, 63, 73, 74, 76.

Cribrilina menzonii, 237.

—— philomela, 237.
radiata, 262.

viner, 236, 263, 277.
Cupularia umbellata, 223.
Cyclodus brandti, 137.
Cyclostoma carinatwm, 299.
Cygnus olor, 63.

Cyprea caput-serpentis, 299.

Cypselus affinis, 71.

Dendrocygna arcuata, 291.

Diabroticus schmerlingi, 174.

Drachoris intermedia, 262.

Dicholophus cristatus, 73.

Didosaurus mauritianus, 281, 323, 324.
Didunculus strigirostris, 298.

Didus ineptus, 281, 282, 283, 296, 298, 301, 302.
Dinornis altus, 415.

crassus, 379.

—- didiformis, 379, 425.
elephantopus, 376, 377, 379.
eacelsus, 415.

—— firmus, 415.

Dinornis giganteus, 415.

gracilis, 415,

gravus, 379, 380.

ingens, 375, 377, 415.
maximus, 374, 375, 415,

—— novee-zealandic, 415.

parvus, 379.

potens, 375, 383, 415.

rheides, 376, 377, 378.

robustus, 375, 383, 396, 415.

struthioides, 375, 376, 415.

torosus, 374, 375, 383, 390, 395, 397, 415.
validus, 415.

sp., 379, 377, 387, 415.

Dinornithide, on the Cranial Osteology, Classification,
and Phylogeny of the, by T. J. Parker, 373-429.
, Classification of the, 413.

, comparative account of the skull in the, 380.
, comparison of the skull of the, with those of
the other Ratitz, 398.

, cranium of the, 380.

——., hyoid of the, 398.

——, list of specimens examined of the, 374.

——, list of works referred to on the, 427.

——, mandible of the, 396.

——-, maxillo-jugal arch of the, 393.

——.,, palatine of the, 395.

——.,, premaxilla of the, 392.

——.,, pterygoid of the, 395.
, quadrate of the, 395.
, summary of cranial characters of the sub-

families and genera of the, 417.
, vomer of the, 394.
Diomedea exulans, 67.

Discoglossus auritus, 160.

pictus, 97, 98, 102, 160.

scovazzit, 102, 160.

Ditaxia variabilis, 262.

Dittosaria wetherelli, 226.

Dodo, on additional Bones of the, and other Extinct
Birds of Mauritius obtained by Mr. Théodore
Sauzier. By Sir E. Newton and Hans Gadow,
281-302.

Echidna atricauda, 155.

mauritanica, 100, 154.

Echis carinata, 97, 155.

| Egidia pulchella, 4, 9, 11.

INDEX. 439

Emeus crassus, 376, 377, 379, 383, 386, 389, 394, | Huryapterya pygmea, 415.

395, 397, 415, 422, 425.
—— gravipes, 415.
sp., 379, 380, 382, 383, 387, 390, 394, 415.
Emys europea, 314.

gigantea, 305.

leprosa, 99.

— lutaria, 105.

orbicularts, 96, 100, 105.
vulgaris, 106.

Entalophora clavula, 261.

palmata, 261.

tergemina, 260, 263, 279.
wanganwensis, 261.

Eremias guttulata, 96, 103, 1382, 133.
pardalis, 133.

simoni, 133.

Eryx jaculus, 97, 101, 143.
Eschara brongniarti, 247, 248, 251.
ciliata, 245,

— coscinophora, 245, 246.

— crenatula, 250.
——— cyclostoma, 239.
—— diplostoma, 247.
fenestrata, 251.
—— lichenordes, 245.

—— membranacea, 240.

—— mucronata, 245.
—— ornatissima, 245,
perforata, 247.
prominens, 248,

pulchra, 245.

—— santonensis, 239.

— semitubulosa, 250.

— simplex, 240.

—— stipitata, 250.

—— sucimargo, 250.

sulcata, 244.

syringopora, 245, 250.

Eumeces algeriensis, 96, 99, 136, 164.
pavimentatus, 136.

— schneideri, 96, 136.
Euprepes vittatus, 101.
Euproctus rusconit, 100, 162.
Euryapteryx elephantopus, 415.

gravus, 415.

ponderosa, 415,

rheides, 415.

Fascicularia edwardsi, 326.

Flustra crassa, 228,

dentata, 231.

sulcata, 245,

sp., 245.

Fortunata, 350.

Fulica atra, 292.

newtoni, 282, 292, 293, 294, 301.

Funingus madagascariensis, 295.

sp. ine., 295.

—— (Alectorenas) nitidissimus, 282.

Gadow, Hans.
Land-Tortoises, and of an extinct Lizard,

On the Remains of some Gigantic

recently discovered in Mauritius, 313-324.
Gallinula chloropus, 294.
Gammarus elegans, 5, 13.
Gecko turcicus, 98.
(Hemidactylus) verruculatus, 98.

Gemellaria prima, 226.
Gerardia lamarcki, 89, 91.
savalia, 91, 92.
Gibbulina sulcata, 299.
Gigantopora lyncoides, 244.

Glossoliga hagenmuelleri, 103.

Gongylus ocellatus, 94,

Gorgonia savoglia, 90, 91.

Gorilla gina, 182.

Goura coronata, 298.

Gregory, J. W. On the British
Bryozoa, 219-279.

Gymnodactylus mauritanicus, 98.

steudneri, 108.

trachyblepharus, 96, 102, 110.

Gymnosarca bathybius, 336.

Heliodilus soumagnit, 286, 287.

Hemeschara variabilis, 241.

Paleogene

Hemidactylus cyanodactylus, 115.
turcicus, 94, 96, 115, 116.
verruculatus, 94, 115.
Hemipodius varius, 76.

Heterodon diadema, 100.
Heteropora conifera, 261.

—— glandiformis, 261, 263, 279.
stellulata, 261.

stipitata, 262.

440)

Hickson, Sydney J.
the Alcyonaria Stolonifera, with a Description

of one new Genus and several new Species,

325-347.
Hippopotamus lemerlir, 305.
Hornera concatenata, 260.
farehanensis, 260, 268, 279.
Jflabelliformis, 259, 262.
—— frondiculata, 259.

—— minuta, 262.
ramosa, 260,
Hyla arborea, 99, 102, 159.

, var. meridionalis, 97, 159.

barytonus, 159.

perezi, 159.

Hyperia cornigera, 365.

Idmonea bialternata, 257, 263, 278.
carinata, 258.

—— compressa, 258.

coronopus, 257, 259, 262, 268, 279.
—— giebeli, 259, 263, 279.

grebeliana, 256.

gracillima, 257, 260, 261.

laticosta, 258.

milneana, 257.

—— reticulata, 258.
seriatopora, 258, 263, 279.

triquetra, 256.

—— (Vervia) seriatopora, 259.
— (Tubigera) giebeli, 256.
Lacerta agilis, 98, 99, 125.
algira, 98.

chalcides, 98.

chameleon, 98.

—— dumerilii, 99.

guttulata, 99.

muralis, 94, 96, 99, 103, 123, 125, 126.
, var. tiliquerta, 126.

neapolitana, 126.

ocellata, 97, 98, 99, 123, 124, 164.

, var. pater, 96, 98, 99, 103, 128, 124, 164.
, var. tangitana, 96, 104, 124, 164.
, typica, 124.

palustris, 98, 99.

— pardalis, 99, 129.

perspicillata, 96, 100, 128, 126.
salamandra, 98.

savignyt, 129,

A Revision of the Genera of |

INDEX.

Lacerta vaillanti, 129.
viridis, 99, 123, 124,
—— ——.,, var. schreiberi, 124.

viridissima, 98, 123.

vulgaris, 98.

Laura gerardie, 89.
Leiopathes lamarcki, 90, 91.
Lepralia angiostoma, 248.
bisculca, 251.

lonsdalei, 247, 263, 278.

manzonii, 237.

—— pauper, 241.
schizogaster, 243.
serrulata, 255.

tenera, 255.

tetragona, 234.

unicornis, 241.
violacea, 243, 245.
Lachenopora, sp., 263, 279.
mediterranea, 262,

Licmetis tenuirostris, 283.

Lnophis regine, 146.

Lophopsittacus mauritianus, 281, 282, 283, 284,
285, 299.

Loricaria americana, 226.

Lunulites goldfussi, 234.

quadrata, 234.

radiata, 233, 234.

subplana, 234.

—— transiens, 233, 263, 276, 277.

urceolata, 233, 234.

Lycognathus teniatus, 149.

textilis, 149.

Lysianassa magellanica, 1.

Lytorhynchus diadema, 97.

Mabuia vittata, 96, 135.

Macrocephalus, 367.

Macrocercus macao, 288.

Macroprotodon cucullatus, 97, 99, 100, 145, 149.
maroccanus, 149, 150.

mauritanicus, 100, 149.

Macropus major, 53.

| Mauritius, de cription of a remarkable Sea-urchin of

the genus Cidaris from, by F. Jeffrey Bell,
303.

, on Remains of an Extinct Gigantic Tortoise
from, by G. A. Boulenger, 305-311.

| ——-, on the Remains of some Gigantic Land-

INDEX,

Tortoises, and of an extinct Lizard, recently
discovered in, by Hans Gadow, 313-324.
Megalapteryx tenuipes, 378.
Meionornis casuarinus, 415.
didiformis, 415.
Meleagris gallopavo, 65.
Melita palmata, 37, 38.

Membranipora appendiculata, 229.
bushi, 229, 263, 276.
catenularia, 233.

crassomuralis, 229, 230, 231, 263, 276.
disjuncta, 232, 263, 276.
elliptica, 230.

eocent, 228, 229, 230, 263, 276.
—— lacroiai, 229, 230, 231, 282.
lawxa, 231, 232.

lowopora, 230.

macrostoma, 229, 232.

— membranacea, 231.

monostachys, 232.

reticulum, 229.

savarti, 228.

—— sigillata, 232.

subtilimargo, 231.

temporaria, 230.

tenuimuralis, 231, 263, 276.

— tuberculata, 231.

virguliformis, 232, 263, 276.

Membraniporella nitida, 236, 277.

, var. eocena, 236.

Meniscopora bigibbera, 251, 263, 278.

Mesopteryx casuarina, 377, 378, 382, 385, 387, 395,
397, 415, 429.

—— didina, 377, 378, 415.

huttonii, 377.

sp., 378, 382, 383, 395, 415.

Micropora coriacea, 238.

cribriformis, 237, 263, 277.

, var. perforata, 237.

—— gracilis, 237, 238.

minsteri, 238.

Microporella fissa, 245.

grisea, 242.

membranacea, 240.

violacea, var. fissa, 245, 247.
Mou giganteus, 415.

Molge hagenmuelleri, 97, 98, 162, 164.
poireti, 97, 98, 100, 162.

VoL. XI.—Part x1. No. 10.—October, 1895.

44]

Molge vulgaris, 98.

waltliz, 95, 97, 98, 162, 163.
Mollia schizogaster, 243.

Movia ingens, 415.

Mucronella angustoecium, 254, 263, 278.

chilopora, 255.

—— hérnest, 254.

—— serrulata, 255.

—— tenera, 255.

ventricosa, 254,

Naia haie, 94, 97, 101, 102, 103, 152.
, var. annulifera, 152.

Natriv torquata, 149.

Necropsittacus rodericanus, 283, 284, 285.

Nerita polita, 299.

Nettapus auritus, 291.

Newton, Sir E., and Gadow, Hans. On additional
Bones of the Dodo and other Extinct Birds of
Mauritius obtained by Mr. Théodore Sauzier,
281-302.

Newton, E. T. OnaSkull of Trogontherium cuviert
from the Forest Bed of East Runton, near
Cromer, 165-175.

Notamia americana, 226.

loricata, 226, 227.

prima, 226.

— wetherelli, 221, 226, 263, 276.

(Gemellaria) loricata, 226.

Nycticorax megacephala, 290.

Ocydromus australis, 294.
Onychocella allica, 239.

angulosa, 238.

— archosia, 239.

— charonia, 239.

— chito, 239.

— cressida, 239.

—— cyclostoma, 239.

—— drya, 239.

—— koninckiana, 239.

—— magnoaperta, 238, 239, 263, 277.
santonensis, 239.

Ophiomorus punctatissimus, 120.

Ophiops elegans, 134.

occidentalis, 96, 134.

Ophisaurus koellikeri, 96, 120.

(Pseudopus) apus, 120.

Opisthocomus cristatus, on the Morphology of, by W.
K. Parker, 43-85.

3R

442

Opisthocomus cristatus, early stages of, 48.
, embryos and adult of, 49.

, hind limb of, 77.

——.,, hip-girdle of, 74.

——.,, light cast upon the Ontogeny of Birds by the
Morphology of, 81.

, ornithological position of, 80.

, skull of, 49.

——, sternum and shoulder-girdle of, 64.

——., vertebral chain of, 59.
, Wings of, 69.

Orang reputed to be Stmia morio, 201-217.

Orbitulipora haidingeri, 253.

lenticwlaris, 254.

petiolus, 253, 263, 278.

Pachyornis elephantopus, 375, 376, 377, 387, 415,
429.

—— immanis, 376, 415.

sp., 376, 377, 415.

Pachystyla inversicolor, 299.

Palxogene Bryozoa, on the British, by J. W. Gregory,
219-279.

Paleornis alexandri, 284.

Palapteryx dromoides, 413, 415.

elephantopus, 415. °

ingens, 375, 413.

plena, 415.

Parker, T. J. On the Cranial Osteology, Classifi-
cation, and Phylogeny of the Dinornithide,
373-429,

Parker, W. K. On the Morphology of a Reptilian
Bird, Opisthocomus cristatus, 43-85.

Patagona gigas, 71.

Pezophaps solitaria, 298.

Phaps chalcoptera, 298.

Phoxus holbélli, 7.

Phyllodactylus europaus, 96, 110.
Platydactylus facetanus, 115.
fascicularis, 98, 115.

—— muralis, 115.

Plestiodon aldrovandii, 136.
Pleurodeles waltlii, 192.

Plotus anhinga, 289.

—— melanogaster, 289,

—— nanus, 282, 288, 289, 300.
nove-hollandie, 280
Podarces simoni, 108, 132.

INDEX.

Podicepes auritus, 289.

cornutus, 289.

cristatus, 289.
—— minor, 289.

pelzelni, 289.

philippensis, 289.
ruficollis, 289.
sp. ine., 289.

Podoceropsis palmatus, 36, 42.
Podocerus cumbrensis, 38, 42.
falcatus, 39.
minutus, 41.

Porella coneinna, 255.
, var. eocena, 254.

Porina subsulcata, 245,

Porphyrio melanonotus, 292, 294.

Porzana carolina, 76.

Psammodromus algirus, 96, 97, 98, 127, 128.
, var. dorie, 128.

, var. nollii, 128.
blanci, 96, 97, 127, 163.
hispanicus, 127.

microdactylus, 96, 127.
Psammophis punctatus, 100, 150.
sibilans, 150.

, var. punctatus, 97, 150.

Pseudopus apus, var. ornata, 103.

serpentinus, 98.

Pterygocera arenaria, 20.

Ptyodactylus lobatus, 96, 108, 111, 112, 113.
oudrit, 103, 111, 112, 113, 114, 163.
Pustulipora clavula, 261.

palmata, 261.

Rana esculenta, 95, 97, 98, 102, 157, 159.
, var. latastii, 157.

, var. ridibunda, 157.

picta, 99.

temporaria, 98.

vwridis, 157.
Ratite, measurements of the skulls of the, 405.
, phylogeny of the, 423.

, summary of the cranial characters of the, 408.
Retepora trigona, 259.

Rhabdonectes, 367.

Rhabdosoma armatum, 368.

brachyteles, 368, 369, 371.

—— brevicaudatum, 366, 368.

Rhabdosoma lilljeborgi, 368.
piratum, 368, 371,
white, 368.
Rhagasostoma hewagonum, 238.

Rhamphastos toco, 73.
Rhinechis amaliv, 103, 144.
Rhizoxenia alba, 336.
Jiliformis, 331, 332, 336.
primula, 332, 336.
rosea, 331, 332, 336.
thalassantha, 332, 336.

Robertson, D., and Stebbing, T. R. R. On four new

British Amphipoda, 31-42,
Salamandra corsica, 161.
maculosa, 98, 104, 161.
, var. algira, 97, 161, 164.
Sareidiornis africanus, 290.
mauritianus, 282, 290, 291, 300.
melanotus, 292.
Sarcodictyon cutenata, 332, 336.
colinabum, 336.

Saurodactylus mauritanicus, 96, 109, 163.

Saurothera vieilloti, 76.
Savaglia lamarckt, 91.

savaglia, 90.

Scaptira inornata, 130.

maculata, 131.

Seeloglaux nove-zealandice, 286, 287.

Schismopora costata, 224.
Schismoporella schizogaster, 243.
Schizoporella beyrichi, 241.
biturrita, 244.

dunkert, 240,

goniostoma, 240.

gonversi, 241.
insignis, 241.
magnoaperta, 239, 263, 277.
magnoincisa, 240, 263, 277.
— — pauper, 241.

requlosa, 240.

—— simplea, 240.

—— tuberosa, 244.

—— unicornis, 240, 241.
variabilis, 241.

Schizoretepora tessellata, 224.
Scina acanthodes, 352, 360, 366, 370.
atlantica, 352, 357, 360, 365.

INDEX.

Seina borealis, 350, 351, 352.

bovallii, 350, 351, 352.

clausi, 350, 351, 352, 360.
concors, 352, 360, 362, 363, 371.
cornigera, 351, 352, 365,

—— crassicornis, 352.

—— ensicorne, 351, 352.

—— gracilis, 352.

lepisma, 351, 352, 364.
longipes, 352.

— mnarginata, 351, 352, 360, 364.
-— edicarpus, 352, 356, 371.

rattrayt, 352, 358, 371.
—— sarsi, 352, 360.

similis, 352, 362, 363, 364, 371.
stenopus, 352, 354, 366, 371.
— tullbergi, 352, 360, 362, 363.
uncipes, 352, 363, 366, 371.
Scincus cyprius, 99.

fasciatus, 96, 137.

officinalis, 96, 137.
(Plestiodon) cyprius, 136.
Scops rutilus, 287.

pacifica, 352, 360, 362, 363, 365.

443

Sea-urchin, description of a remarkable, from Mau-

ritius, by F. Jeffrey Bell, 303.
Seps chalcides, 140.
mionecton, 102.

Simia morio, on the Orang reputed to be, 201-217.

, foot of, 204.
—— ——., hand of, 203.

—— ——, muscular anatomy of fore limb, 205.
—— ——,, muscular anatomy of hind limb, 211.

, muscular system of, 204,
satyrus, 201, 217.
Smittia tubularis, 255, 263, 278.

Sophrosyne murrayi, 33.
robertsoni, 31, 33, 41.
Sphenops capistratus, 100, 141.

Spizella pusilla, 79.

Stebbing, T. R. R. Descriptions of nine new Species
of Amphipodous Crustaceans from the Tropical

Atlantic, 349-371.

——. On the Genus Urothoe and a new Genus

Urothoides, 1-30.

Amphipoda, 31-42.

, and Robertson, D. On four new British

444

Stenodactylus guitatus, 96, 99, 107, 108.
mauritanicus, 107, 108.
petersti, 108.

wilkinsonti, 108.
Stereosoma, gen. noy., 337.
celebense, 337, 338, 346.
Stolonifera, genera of the, 331.
Striv flammea, 286, 287, 288.
sauzieri, 282, 286, 287, 299.
Struthio camelus, 58.

Sympodiun coralloides, 336.

margaritaceum, 336.

Syornis casuarina, 377, 415.

crassus, 415,

rheides, 415.

Syrrhoé fimbriatus, 34, 41.

Talegalla lathani, 73, 75.

Tarentola angusticeps, 116.

—— mauritanica, 96, 97, 115, 116.

, var. deserti, 115, 116, 163.
neglecta, 96, 116.

Teichopora clavata, 249, 250, 263, 278.
Tervia solidula, 259.

Testudo abrupta, 305.

campanulata, 105.

coriacea, 98.

—— daudini, 307, 317, 318, 319, 322.
elephantina, 306, 307, 308, 310, 317, 318,
319, 320.

elephantopus, 309, 319, 320.

—.. gigantea, 305, 306, 307.

greca, 98, 104, 105.

—— grandidierr, caudal vertebrae of, 308.
, cervical vertebra of, 308.

, limb-bones of, 310.

. pectoral arch of, 310.

—— ——, pelvis of, 310.

, sacral vertebree of, 308.

, shell of, 308.

—— ——,, skull of, 307.

—— guentheri, 320.

hololissa, 306, 307, 317, 318.

abera, 96, 97, 98, 99, 104.

indica, 313, 314, 316, 320, 323.

inepta, 313, 314, 315, 316, 319, 320, 321,
322, 324.

leptocnemis, 318, 314, 321.

INDEX.

Testudo marginata, 98, 105.
mauritaniea, 104,

—— microtympanum, 307.

—— nigra, 319.

—— nigrita, 319.

—— perraulti, 313, 316,

ponderosa, 317, 318, 319.

—— pusilla, 104.

—— sauzeri, 314, 315, 316, 318, 323.

sulcata, 309.

—— sumeirei, 314, 317, 318, 319, 320, 322, 323,
324,

—— triserrata, 313, 314, 315, 316, 317, 318, 319,

320, 321, 322, 324.

vicina, 309.

vosmaeri, 314.

Tetrao urogallus, 50, 53, 57.

Tiliqua ocellata, 98.

Topaza pella, 71.

Tortoise, on Remains of an Extinct Gigantic, from
Mauritius (Zestudo grandidieri, Vaillant), by
G. A. Boulenger, 305-311.

Tortoises, on the Remains of some Gigantic Land-,
recently discovered in Mauritius, by Hans
Gadow, 313-324,

Treron olax, 298.

Triton nebulosus, 100, 162.

porreti, 99.

Trocaza meyert, 295.

Troglodytes aubryi, 179, 180, 181, 182, 184, 185.

—— bicolor, 203.

—— calvus, brain of, 198.

—— ——, comparison of musculature of, with that

of the common Chimpanzee, 196.

——,, ear of, 178.

, external characters and anatomy of,

178-201.

, foot of, 179.

—— ——, hand of, 179.

—— ——, muscular anatomy of fore limb, 185.

—— ——, muscular anatomy of hind limb, 190.

—— ——, muscular system of, 185.

—— ——, palate of, 197.

_, skull of, 181.

—— niger, 180, 181, 182, 183, 184, 196, 198, 199,
200, 201, 218.

—— ischego, 182, 183, 184, 201.

INDEX.

Troglodytes vellerosus, 180.
T'rogonophis wiegmanni, 96, 99, 122. |

Trogonotherium of the Forest Bed, East Runton, com-
parison with that of Conodontes boisvilletti, 172.

» comparison with that of Fischer’s Z'rogono-
thervum cuvieri, 171.

——, dental characters of, 170.

——.,, description of the skull of, 166.

——,, distinctive characters of, 169.

, introductory remarks upon, 165.
——.,, measurements of, 173.

cuvieri, on a skull of, from the Forest Bed of
East Runton, near Cromer, by E. T. Newton,
165-175.

Tropidonotus aurolineata, 149.

murorum, 149,

natriv, 97, 148.

ocellata, 149.

-— persa, 149.

viperinus, 95, 97, 99, 149.

, var. aurolineatus, 99.

Tropiocolotes steudneri, 108.
tripolitanus, 96, 108.
Tubucellaria opuntioides, 256.

Tubularia cornucopic, 334.
Turnix rostrata, 73, 76.
Turtur picturatus, 295.
Tyro atlantica, 365.

cornigera, 351, 365.

pacifica, 365.

sarsi, 351, 365.

tullbergii, 365.

Umbonula bartonense, 248, 249, 263, 278.
calcariformis, 249, 263, 278.
Uria troile, 75.

Uromasti« acanthinurus, 96, 97, 99, 101, 119.
spinipes, 101, 119.

temporalis, 119.
Urothoe, on the Genus, by T. R. R. Stebbing, 1-30. |
abbreviata, 3, 9, 27.
-—— bairdii, 7, 27. |
brevicornis, 7, 10, 12, 13, 15, 17, 28, 25, 27,
28, 29, 30.

445

Urothoe elegans, 4, 5, 6,9, 10, 18, 14, 17, 22, 25, 27,
28.

wrrostratus, 4, 10, 27, 28.

lachneéssa, 4, 26, 27.

marinus, 5, 6, 7, 8, 12, 13, 14, 15, 16, 21, 23,
24, 27, 28, 29.

Se VATA DECLUNCLUSS

8, 27, 28.

norvegica, 6, 7, 9. 21, 25, 27, 28, 30.
pinguis, 4, 27.

poucheti, 9, 25, 27, 28.

—— pulchella, 4, 11, 14, 15, 17, 21, 27, 28, 30.
rostrata, 3, 4, 10, 27.

Urothoides, gen. nov., 26.

lachneéssa, 26, 27.

Varanus arenarius, 121.

griseus, 96, 97, 121.

scincus, 121.

Vinago australis, 295.

Vincularia fragilis, 256.

Vipera ammodytes, 153.

arietans, 97, 102, 153, 154.

aspis, 101, 153.

avicennce, 101, 155.
brachyura, 99, 154.

cerastes, 99, 155.

daboia, 98.

echis, 99.

euphratica, var. mauritanica, 103.
latastvi, 97, 101, 103, 153.
lebetina, 97, 98, 99, 100, 108, 153, 154.
Xiphocephalus, 367.

Zamenis algirus, 97, 100, 145, 146, 147.
—— ater, 146.

atrovirens, 146.

cliffordii, 101, 148.

diadema, 97, 101, 147, 148.

—— florulentus, 94, 100, 146, 147.
gemonensis, 108.

hippocrepis, 97, 98, 146, 147.

nummifer, 148.
viridiflavus, 103.
Zerzoumia blanci, 103.
Zootoca deserti, 101, 13).

END OF VOLUME XIII.

PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLEET STREET.

i
Me

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XIII.—Parr 1.

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS. LONGMANS, GREEN, AND C0., PATERNOSTER-ROW.

January 1891.
Price 21s.

Taylor and Francis, Printers. |

[Red Lion Court, Fleet Strect.

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.

To Fellows. To the Publie.

£5. ees eile
VOLUME I. (1833-1835, containing 59 Plates). . Price 313 6 418 0O*
VOLUME II. (1835-1841, containing 71 Plates). . Price 4 0 O 5 6 6%
VOLUME III. (1842-1849, containing 63 Plates). . Price 3 8 6 . 411 0*
VOLUME IV. (1851-1862, containing 78 Plates). . Price 6 2 O . 8 2 6*

VOLUME _V. (1862-1866, containing 67 Plates). . Price 5 4 3 619 0

VOLUME VI. (1866-1869, containing 91 Plates). . Price 11 5 O . 15 0 0

VOLUME VII. (1869-1872, contaming 73 Plates) . . Price 10 4 0 . 13 12 0

VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 3 . Fa © 0)

Parr ol (W872) a5 Veil yc mereka ee meds tue aio OL Ome 014 0

» 2 (1872). Pa ale Goi} 2 2 0

iy Bo. (ei) Einwliinya Ope 116 0

» 4 (1878) . Oils 2) 0)

» ». (1878) . » O18 O 1 4 0

» 6. (1878) . » O19 6 1 6 0

» 7. (1874) » O16 6 1 2 0

» 9. (1874) 5 O18) 20 14 0

SRO MB VA ci ca Bal Ga Miuectatecmee baal Ue am Berea Ll lea 2 2 0

VOLUME IX. (1875-1877, containing 99 Plates) . . Price 12 0 O 16 0 0

Part 1. (1875, contaiming 22 Plates) . . . . . re sab ee @

» 2. (1875, containmg 12 Plates) . ... . oe Ae 2 2 0

» 98. (1875, containmg 5 Plates) . . .. . yy Okie O 1 4 0

, 4. (1875, containmg 14 Plates) . . .. . preiond bea ae pe 2 2 0

» 5. (1875, containing 9 Plates) . . .. . tlie see Grats 110 0

57) 6. (L876, contamine @4q Places) e-em en eis) 0 eLOn a6 014 0

» 7. (1876, containmg 7 Plates) . . .. . » O15 9 a 0

» 8. (1876, contamimg 10 Plates) . . .. . mi FO elOms0 ae 14 0

5 9. (1876, contaimmg 6 Plates) . . .. . a A Oe 112 0

3 10. (1877,,containing y/oPlaes)meg a i) wee he earn elec) sO 110 0

» ll. (1877, containing 5 Plates and Title and Index),, O15 9 . TQ 40)

* Only odd parts of these yolumes can be supplied.
Continued on page 3 of Wrapper.

THE

Taylor and Francis, Printers. |

TRANSACTIONS

OF

ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XII.—Parr 2.

\BRA,
«AS
4 i's
: Of « &
erates oN
ie fa
=< a\
2 ‘3 ’ \
% ry
WH) a
}
A

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;

AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.

April 1891.

Price 21s.

{Red Lion Court, Fleet Street.

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.
To Fellows. To the Public.
£ os. ad. BGs ids
VOLUME I. (1833-1835, containing 59 Plates). . Price 313 6 . . . 418 OF
VOLUME II. (1835-1841, containing 71 Plates). . Price 4 0 0... 5 6 6%
VOLUME III. (1842-1849, containing 63 Plates). . Price 8 8 3 . . . 411 O*
VOLUME IV. (1851-1862, containing 77 Plates). . Price 6 2 0... 8 2 6%
VOLUME _V. (1862-1866, containing 67 Plates). . Price 5 4 3 . . . 619 O
VOLUME VI. (1866-1869, containing 92 Plates). . Pricell 5 O 15 0 0
VOLUME VII. (1869-1872, containing 73 Plates) . . Price 10 4 0 . . .13812 O
VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 3 . . .1211 O
VOLUME IX. (1875-1877, containing 99 Plates). . Price 12 1 6... .16 2 0
VOLUME X. (1877-1879, containing 95 Plates). . Price 10 0 38 13 7 O
Parr 1. (1877, with numerous woodcuts) . . . . pe) 0 012 0
» 2 (1877, contaming 27 Plates) OU ee Sp en eG 110 O
» 93. (1877, contaming 6 Plates) » 9018 O 1 4 0
5, 4. (1878, contaming 9 Plates) . ... . Ba pans (6) 110 0
» 5. (1878, contaming 3 Plates) . ... .- Os 9 20 012 0
5). 6. (1878, contamine 99) Plates), 24S <2) sel eG 110 0
» 7. (1878, containmg 7 Plates) . . .. . » 2018 O 140
», 8. (1878, containing 8 Plates) » O15 O 100
» 9. (1878, containing 4 Plates) 40) (9:80 012 0
» LO. (1879, contaming 6 Plates) » O12 O 016 O
> 11. (1879, containing 5 Plates) . ... . Oso O 012 O
», 12. (1879, containmg 4 Plates) . ... . » O15 9 110
» 18. (1879, containing 7 Plates and Title and Index) ,, 018 O 140
GeEneErRAL Inpex, Vols. I. to X. (1833-1879) MO ei 010 O

* Only odd parts of these volumes can be supplied.
Continued on page 3 of Wrapper.

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XIUI.—Parr 3.

LONDON: $C’

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.

October 1891.

Price 24s.

Taylor and Francis, Printers. | [Red Lion Court, Fleet Stree

ot

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.

To Fellows. To the Public.

BB. Gh ah £ s. d.
VOLUME I. (1833-1835, contaming 59 Plates). . Price 313 6 . . . 418 O*
VOLUME II. (1835-1841, contaimg 71 Plates). . Price 4 0 0 ... 5 6 6%
VOLUME III. (1842-1849, containing 63 Plates). . Price 3 8 3 . .. 411 O*
VOLUME IV. (1851-1862, containing 77 Plates). . Price 6 2 0 . .. 8 2 6%
VOLUME _ V. (1862-1866, containing 67 Plates). . Price 5 4 3 . . . 619 O

VOLUME VI. (1866-1869, containing 92 Plates). . Price 11

Or
i=)
ax
=}
So

VOLUME VII. (1869-1872, containing 73 Plates) . . Pricel10 4 0 . . .1812 O
VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 38 . . .1211 O
VOLUME IX. (1875-1877, containing 99 Plates). . Price 12 1 6 16 2 0
VOLUME X. (1877-1879, containing 95 Plates). . Price 10 0 8 13 7 0
Part 1. (1877, with numerous woodcuts) . » 0 9 0 012 0

> 2. (1877, containing 27 Plates) Bed eG 110 O

5) Os (187% contamine .60Plates)\ 2 a) ee 0 18 00 I 410

», 4. (1878, contaming 9 Plates) . . .. . lee! 1G 110 0

» 9. (1878, containmg 3 Plates) fe OO) 012 0

» 6. (1878, containing 9 Plates) eal 62006 11080

» 7. (1878, contaming 7 Plates) » 918 O 1 0

», 8. (1878, containing 8 Plates) . . .. . >» O15 O 100

» 9. (1878, contaming 4 Plates) . ... . 5 O829'0 012 0

5, 10. (1879, containing 6 Plates) » 012 0 016 0

» 11. (1879, containing 5 Plates) > O08 9 0 012 O

», 12. (1879, containing 4 Plates) » O15 9 110

» 13. (1879, containing 7 Plates and Title and Index) ,, 018 O 1 4 0
Genera Inpex, Vols. I. to X. (1833-1879) . . . Oat 6 010 O

* Only odd parts of these volumes can be supplied.
Continued on page 3 of Wrapper.

i ?

Nw

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XIII.—Parr 4.

MAY 16 1892

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.

April 1892.
Price 6s.

Taylor and Francis, Printers.] [Red Lion Court, Fleet Street.

>

PRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.

To Fellows. To the Public.

£s. ad. £ os. d.
VOLUME I. (1833-1835, containing 59 Plates). . Price 313 6 . . . 418 O*
VOLUME II. (1835-1841, containing 71 Plates). . Price 4 0 0... 5 6 6%
VOLUME III. (1842-1849, containing 63 Plates). . Price 3 8 3... 411 O*
VOLUME IV. (1851-1862, containing 77 Plates). . Price 6 2 0... 8 2 6%
VOLUME _V. (1862-1866, containing 67 Plates). . Price 5 4 3 spleen gOmLo) 0
VOLUME VI. (1866-1869, containing 92 Plates). . Price 11 5 O ce ae OO
VOLUME VII. (1869-1872, containing 73 Plates) . . Pricel10 4 0 . . .1312 O
VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 38 . . .1211 0

VOLUME IX. (1875-1877, contaiming 99 Plates). . Price 12 1

fer)
a
er)
cS)
i)

VOLUME X. (1877-1879, containing 95 Plates). . Price 10 0 3 13 7 0
Part 1. (1877, with numerous woodcuts) . 0. 0 . O12 0
» 2. (1877, containing 27 Plates) . ... . ba can) - 110 0

» 9. (1877, containing 6 Plates) . ... . ea OM Gia On. 1 4 0
» 4. (1878, containmg 9 Plates) .... . 5 lb Be - 110 0

» 0. (1878, contaming 3 Plates) .... . yp O OO ~ 01420

» 6. (1878, contaimmg 9 Plates) . ... . ae eth G . 110 0

» 7@. (1878, containg 7 Plates) . ... . » 018 O . 140
», 8. (1878, contamimg 8 Plates) .... . OM DOna. - 100

» 9. (1878, contaming 4 Plates) . ... . ap OO) . O12 0

», 10. (1879, contaming 6 Plates) » O12 0 - 016 O

», 11. (1879, containing 5 Plates) OO - O12 0
», 12. (1879, containing 4 Plates) . eine » O15 9 ronal ie 0)

, 13. (1879, containing 7 Plates and Title and Indes) » 918 O “1 4 0
Generat Inpexx, Vols. I. to X. (1833-1879) . . . Py he O . 010 0

* Only odd parts of these volumes can be supplied.
Continued on page 3 of Wrapper.

dae
~

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XII.—Parr 5.

bd

MIA

,
ey) OH
‘S ty e

Zracnian BW

AP 2 OL 7
h

LONDON:
PRINTED FOR THE SOCIRATY,

SOLD AT THEIR HOUSE IN HANOVER-SQUARE;

AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.
February 1893.
Price 15s.

Taylor and Francis, Printers, |

[Red Lion Court, Fleet Street.

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.
To Fellows. To the Public.
£9 ER hs £ s. d.
VOLUME I. (1833-1835, containing 59 Plates). . Price 313 6 . . . 418 O*
VOLUME _ II. (1835-1841, containing 71 Plates). . Price 4 0 0... 5 6 6%
VOLUME III. (1842-1849, containing 63 Plates). . Price 3 8 3 . . . 411 O*
VOLUME IV. (1851-1862, containing 77 Plates). . Price 6 2 0... 8 2 6%
VOLUME _ V. (1862-1866, containing 67 Plates). . Price 5 4 3 . . . 619 O
VOLUME VI. (1866-1869, containing 92 Plates). . Pricell 5 O 15 0 0
VOLUME VII. (1869-1872, containing 73 Plates) . . Price10 4 0 . . .13812 O
VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 3 . . . 1211 O
VOLUME IX. (1875-1877, containing 99 Plates). . Price 12 1 6. . .16 2 O
VOLUME X. (1877-1879, containing 95 Plates). . Price 10 0 3 .138 7 0
Parr 1. (1877, with numerous woodcuts) . . . . 5s Oar Oi ON: 012 0
» 2. (1877, containing 27 Plates) . . .. . seule ae GLope, 1 10.0
>, 9. (1877, containing 6 Plates) pO Sy Os. 1 4 0
» 4. (1878, contaming 9 Plates) . ... . Pie ERG 110 0
» D. (1878, containing 3 Plates) . ... . mW OO 5 012 0
», 6. (1878, containing 9 Plates) . ... . iy eee apo 110 0
» 7. (1878, contaming 7 Plates) » O18 O . 1 0
>» 8. (1878, contaming 8 Plates) .... . » O15 0. 1 0 0
» 9. (1878, contammg 4 Plates) . ... . Si OnE Oy Olive 012 O
», 10. (1879, contaming 6 Plates) .... . » 012 0 016 0
5 11. (1879, contaiing 5 Plates) . ... . OO sOwe 012 0
», 12. (1879, containing 4 Plates) . . .. . 5 OD 1s | @) 1 0)
» 13. (1879, containing 7 Plates and Title and Index) ,, 018 O . 14 0
Grnerat Inpex, Vols. I. to X. (1883-1879) . . . » O 7 6 010 O

* Only odd parts of these volumes can be supplied.
Continued on page 3 of Wrapper.

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XIII.—Parr 6.

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE:
AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.

June 1893.

Price 12s.

Taylor and Francis, Printers,]

[Red Lion Court, Fleet Strect.

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY
OF LONDON.

To Fellows. To the Public.

Lisi ds Soe Ses
VOLUME I. (1833-1835, contaming 59 Plates). . Price 813 6... 418 O*
VOLUME II. (1835-1841, containing 71 Plates). . Price 4 0 0... 5 6 6%

VOLUME III. (1842-1849, containing 63 Plates). . Price 5 8 38 . . . 411 O*
VOLUME IV. (1851-1862, containing 77 Plates). . Price 6 2 O .. . 8 2 6
VOLUME _ V. (1862-1866, containing 67 Plates). . Price 5 4 3 . . . 619 O
VOLUME VI. (1866-1869, containing 92 Plates). . Pricel1l 5 0 .. .15 O O
VOLUME VII. (1869-1872, containing 73 Plates) . . Pricel10 4 0 . . .1312 0
VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 38 . . . 1211 0 |

VOLUME IX. (1875-1877, containing 99 Plates). . Price 12 16 .. .16 2 O

VOLUME X. (1877-1879, contaiming 95 Plates). . Price 10 0 3 . . .13 7
GENERAL INDEX, Vols. I. to X. (1883-1879) . Price 0 Gg he 610 0
VOLUME XI. (1880-1885, contaiming 97 Plates). . Price 9 12 O 12 16 O
Part 1. (1880, containing 4 Plates) . . pee Oe 0) 016 0
» 2. (1880, containing 7 Plates) » 0 18 0 1 4 0
» 3. (1881, containing 8 Plates) eel 2G 110 0
» 4. (1881, contaming 38 Plates) . . She 6 010 0
» >». (1881, containing 13 Plates) . . . . . Ole) 1 4 0
» 6. (1882, containing 6 Plates) > 0. 12> 0 016 0
7. (1882, containing 9 Plates) sy 10) Uy 1020
» 8. (18838, containing 11 Plates) OM 016 O
» 9. (1883, containing 10 Plates) Ose YO 016 0
» 10. (1885, contaming 12 Plates) . . .. . pees tan Ll bain (6) Qe au)
» 11. (1885, containing 14 Plates and Title and Index),, 111 6 2) 92

* Only odd parts of these volumes can be supplied.

Continued on page 3 of Wrapper.

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XIII.—Parr 7.

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS. LONGMANS, GREEN AND CO., PATERNOSTER-ROW.

August 1893.

Price 12s.

Taylor and Francis, Printers, ] [Red Lion Court, Fleet Street.

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.
To Fellows. To the Public.
£5 By th LS.
VOLUME I. (1833-1835, contaming 59 Plates). . Price 8138 6 . .. 418 O*
VOLUME II. (1835-1841, containing 71 Plates). . Price 4 0 O Bie) dh =
VOLUME III. (1842-1849, containing 63 Plates). . Price 3 8 3 . . . 411 O*
VOLUME _ IV. (1851-1862, containing 77 Plates). . Price 6 2 0... 8 2 6

VOLUME _ V. (1862-1866, containing 67 Plates). . Price 5 4 8 . . . 619 O

VOLUME VI. (1866-1869, containing 92 Plates). . Price 11

Or
j=}
x
i=}
Oo

VOLUME VII. (1869-1872, containing 73 Plates) . . Pricel0 4 0 .. . 1812 0
VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 3 . . .1211 O
VOLUME IX. (1875-1877, containing 99 Plates). . Price 12 16. . . 16 2 0
VOLUME X. (1877-1879, containing 95 Plates). . Price 10 0 8 . . .18 7 O

GENERAL INDEX, Vols. I. to X. (1833-1879) Bp ee Ose.

(or)
o
=
i=)
S

VOLUME XI. (1880-1885, contaming 97 Plates). . Price 9 12
Part 1. (1880, contaming 4: Plates)

» 2. (1880, contaiming 7 Plates)

» 0. (1881, containing 8 Plates)

» 4. (1881, containing 3 Plates)
5. (1881, containing 13 Plates

—
CY)
=
(oy)

)
6. (1882, containing 6 Plates)
7. (1882, containing 9 Plates)
» 8. (1883, contaming 11 Plates)
9. (1883, containing 10 Plates)
», 10. (1885, containing 12 Plates) . . . . .
. (1885, containing 14 Plates and Title and Index)

7
iret el SS) OO 1S) LO) le OQ) 2)
—
~ I w
(ey ery is) eS) ee) Se) 2) fe) en Se)
Wh ne) Ce eS) Ca ime ey a (i)
=
(or)
SS) ere) eS) see) er dey) Sy iS}

9

* Only odd parts of these volumes can be supplied.

Continued on page 3 of Wrapper.

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

Vou. XIII.—Parr 8.

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS, LONGMANS, GREEN AND CO., PATERNOSTER-ROW.

April 1894.

Price 12s.

aylor and Francis, Printers. | {Red Lion Court, Fleet Street.

TRANSACTIONS OF THE ZOOLOGICAL SOCTETY

OF LONDON.
To Fellows. To the Public.
£s. d. £8. ds
VOLUME I. (1833-1835, containing 59 Plates). . Price 313 6 . . . 418 OF
VOLUME II. (1835-1841, containing 71 Plates). . Price 4 0 0 . .. 5 6 6*
VOLUME III. (1842-1849, containing 63 Plates). . Price 3 8 3 . . . 411 O*
VOLUME IV. (1851-1862, containing 77 Plates). . Price 6 2 0... ..8 2 6G
VOLUME _ V. (1862-1866, containing 67 Plates). . Price 5 4 3 . . . 619 0
VOLUME VI. (1866-1869, containing 92 Plates). . Pricell 5 O . . .15 O O
VOLUME VII. (1869-1872, containing 73 Plates) . . Pricel0 4 0 . . .1312 O
VOLUME VIII. (1872-1874, containing 82 Plates). . Price 9 8 3 . . . 1211 O
VOLUME IX. (1875-1877, containing 99 Plates). . Pmce 12 16. . .16 2 0
VOLUME X. (1877-1879, containing 95 Plates). . Price 10 0 3 . . .18 7 O
GENERAL INDEX, Vols. I. to X. (1833-1879) . pe Olver 010 0
VOLUME XI. (1880-1885, containing 97 Plates). . Price 9 12 O 1216 O
Part 1. (1880, containing 4 Plates) » O12 0 016 O
» 2. (1880, containing 7 Plates) » 918 O 1 4 0
» &. (1881, contaming 8 Plates) ek 8 110 0
» 4. (1881, contaming 3 Plates) 50 6 010 0
» ». (1881, contaming 13 Plates) oy Alte}, 0) 1 4 0
» 6. (1882, contaiming 6 Plates) A Oe O 016 0
» “@. (1882, containing 9 Plates) » 915 O Tes Gl, 0)
» 8. (1883, containing 1] Plates) » O12 O 016 O
» 9. (1888, containing 10 Plates) OF 220) 016 0
» 10. (1885, containmg 12 Plates) . . .. . i ed EG 2 2270
», 11. (1885, containing 14 Plates and Title and Index),, 111 6 2.20

* Only oid parts of these volumes can be supplied.

Continued on page 3 of Wrapper.

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

~ 8004988
‘You. XIM.—Parr 9.

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS. LONGMANS, GREEN AND CO., PATERNOSTER-ROW.

October 1894.

Price 15s.

Taylor and Francis, Printers,] ; [Red Lion Court, Fleet Street .

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY
OF LONDON.

To Fellows. To the Public.

£s. d. £s d.
VOLUME I. (1833-1835, containing 59 Plates). . Price 813 6 .. . 418 OF
VOLUME II. (1835-1841, containing 71 Plates). . Price 4 0 0 ... 5 6 6*
VOLUME III. (1842-1849, containing 63 Plates). . Price 3 8 3 . 411 0*
VOLUME IV. (1851-1862, containing 77 Plates). . Price 6 2 0 ...8 2 6
VOLUME _V. (1862-1866, containing 67 Plates). . Price 5 4 3 619 0
VOLUME VI. (1866-1869, containing 92 Plates). . Pricell 5 O .15 0 0
VOLUME VII. (1869-1872, containing 73 Plates) . . Pricel0 4 0 . 1312 0
VOLUME VIII. (1872-1874, containing 82 Plates) . . Price 9 8 38 .1211 0
VOLUME IX, (1875-1877, containing 99 Plates). . Price 12 1 6 .16 2 0

VOLUME X. (1877-1879, containing 95 Plates). . Price 10 0
GENERAL INDEX, Vols. I. to X. (1833-1879) . OSes

a ow
—
oO w
eS
(SOA
(=>) X=)

VOLUME XI. (1880-1885, containing 97 Plates). . Price 912 0 ~ 1216 0
Part 1. (1880; containing 4 Plates) . . » O12 O 016 0
» 2 (1880, containing 7 Plates) » O18 O ‘ 1 4 0
» 9 (1881, containing 8 Plates) . yoda 26 110 0
» 4. (1881, containing 3 Plates) pe Orfeo) 010 0
» 0. (1881, containing 13 Plates) » 018 O. 14 0
» 6. (1882, containing 6 Plates) » 012 O 016 O
» @ (1882, containing 9 Plates) 2 » 915 0. 100
, 8. (1883, containing 11 Plates) . . .. . » O12 0. 016 0
» 9. (1883, containing 10 Plates) . ... . » 012 0 016 O
», 10. (1885, containing 12 Plates) . . Ee a AAG ait ee
» 11. (1885, containing 14 Plates and TitleandIndex),, 111 6 . 2 2 0

* Only odd parts of these volumes can be supplied.
Continued on page 3 of Wrapper.

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY

OF LONDON.

310606
Vou. XII.—Parr 10.

LONDON: |

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.

February 1896.

Price 12s.

Taylor and Francis, Printers, | {Red Lion Court, Fleet Street.

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.

To Fellows. To the Public.

. £s d. £8. d.
VOLUME I. (1833-1835, containmg 59 Plates). . Price 313 6 . . . 418 OF*
VOLUME | II. (1835-1841, containing 71 Plates). . 5 400... 5 6 6*
VOLUME III. (1842-1849, containing 63 Plates) . EO NO aae ian yay aaa a Nol Oe

VOLUME IV. (1851-1862, contaming 77 Plates). . ,, 6 290 ...8 2 6

VOLUME _ V. (1862-1866, containing 67 Plates) . DAS eer Mens OLsLON 0)

VOLUME VI. (1866-1869, containing 92 Plates) . sere On MO wae OOO

VOLUME VII. (1869-1872, containing 73 Plates) . LOS AM Osa od alieenO

VOLUME VIII. (1872-1874, contaming 82 Plates). . ,, SO iitcwmnco aN prone eal bene by la (0)

VOLUME IX. (1875-1877, contaming 99 Plates) . Bere OS Ol aG)c soe meee hOwe nO

VOLUME X. (1877-1879, containmg 95 Plates). . ,, 10 0 8 ...18 7 O

GENERAL INDEX, Vols. I. to X. (1833-1879) . ,, 0 010 0

VOLUME XI. (1880-1885, containing 97 Plates). . Price 9 12 O 1216 O

Parr 1. (1880, containing 4 Plates) i OnleerO 016 0

» 2. (1880, containmg 7 Plates) pea Oma) Le 420

» 98. (1881, contaming 8 Plates) eee EN DG 110 0

» 4. (1881, contaming 3 Plates) » O 6 010 0

i oe (LSS contamimeeS Plates) tease vp Walks) 1 4 0

» 6. (1882, contaming 6 Plates) 5 OO We) 016 0

» 7. (1882, contamimg 9 Plates) pp Ob Ue 0) 12020

» 8. (18838, contaming 11 Plates) . . . 5 O25 0 016 0

» 9. (1883, containing 10 Plates) . . . SS OMAP AE) 016 O

», 10. (1885, containing 12 Plates) . . . . . pee ella Pe AO)

» 11. (1885, containing 14 Plates and Title and Index),, 111 6 2 2 0

* Only odd parts of these volumes can be supplied.

Continued on page 3 of Wrapper.

he

TRANSACTIONS

OF

THE ZOOLOGICAL SOCIETY:

OF LONDON.

Vou. XIT.—Parrt 11.

LONDON:

PRINTED FOR THE SOCIETY,
SOLD AT THEIR HOUSE IN HANOVER-SQUARE;
AND BY MESSRS. LONGMANS, GREEN, AND CO., PATERNOSTER-ROW.

October 1895.

Price 21s.

Taylor and Francis, Printers, | [Red Lion Court, Fleet Street

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY

OF LONDON.
To Fellows. To the Public.
£s. ad. £-s, d.
VOLUME I. (1833-1835, containing 59 Plates). . Price 313 6 . . . 418 O*
VOLUME II. (1835-1841, containing 71 Plates). . ,, 400... 5 6 6*
VOLUME III. (1842-1849, containing 63 Plates) . SG AO sama ae eee Os
VOLUME IV. (1851-1862, containing 77 Plates). . ,, 6 20...8 2 6
VOLUME _ V. (1862-1866, containing 67 Plates) . 35 BRA iS atievesre Or Lou
VOLUME VI. (1866-1869, containing 92 Plates). . ,, 11 5 0... 15 0 0
VOLUME VII. (1869-1872, containing 73 Plates). . ,, 10 4 0 . . .1812 O
VOLUME VIII. (1872-1874, containing 82 Plates) . 5 9 8 38 . 1211 0
VOLUME IX. (1875-1877, containing 99 Plates) . bees biden te (5) 16 2 0
VOLUME X. (1877-1879, containing 95 Plates). . ,, 10 08 .. .18 7 0
GENERAL INDEX, Vols. I. to X. (1833-1879) ” 0 7 6 010 0
VOLUME XI. (1880-1885, containing 97 Plates). . Price 912 0 . . 1216 O
Part 1. (1880, containing 4 Plates) » O12 0 016 0
» 2. (1880, containing 7 Plates) » 018 O 1 4 0
» 9. (1881, containmg 8 Plates) 5 aad 6 110 0
» 4. (1881, containing 3 Plates) asnOcend) 1G 010 0
», ». (1881, contaiming 13 Plates) » 918 O 1 4 0
» 6. (1882, contaming 6 Plates) . ... . » O12 0 016 0
» %. (1882, containing 9 Plates) » O15 0. 10 0
» 8. (1883, containmg 11 Plates) » O12 0 016 O
» 9. (18838, containing 10 Plates) » 012 0 016 0
» 10. (1885, containing 12 Plates) . . .. . thoes = @ B®)
» 11. (1885, containing 14 Plates and Title and Index),, 111 6 2 2 0

* Only odd parts of these yolumes can be supplied.

Continued on page 3 of Wrapper.

TRANSACTIONS OF THE ZOOLOGICAL SOCIETY OF LONDON
(continued).

VOLUME XII. (1886-1890, containing 65 Plates) .

Partl. (1886, containing 6 Plates)

22

29

ey)

VOLUME XIII. (1891-1895, containing 62 Plates) .

2.
3. (1886, containing 2 Plates)
4, (1886, containing 12 Plates)
5. (1886, contaming 5 Plates)
6.
7
8
9

(1886, containing 7 Plates)

(1887, containing 7 Plates)

. (1888, containing 8 Plates)

. (1889, containing 3 Plates)
. (1889, containing 7 Plates)

10. (1890, containing 8 Plates and

Partl. (1891, containing 6 Plates)

10
11

. (1891, containing 6 Plates)
. (1891, containing 6 Plates)
. (1892, containing 1 Plate).
. (1893, containing 9 Plates)

. (1893, containing 4 Plates)

. (1893, containing 6 Plates)

. (1894, containing 6 Plates)
. (1894, containing 6 Plates)
. (1895, containing 5 Plates)

. (1895, containing 7 Plates and

Title and Index}

Title and Index)

To Fellows.

coor Se

Price 5 8

a WY)
One)

pp (Oa
Sec O. 15
oe 09
+ Qn 12
Ook
006
Oks
» O15

Price 6 8

Onis
C015
e018
CORA
OntE
0 9
e020
OR
yal
Oe
O15

ooooeoeceonanoc os

a oc O& © w

SS)

coowoc9qce~e

To the Public.

ess
4,

12
16

16
12

th Or One iad Sit Ose
=
rw)

11

Ke oOocoocaeaeoo KH KY HY ow
ed
cw)

eo te ee) Oe) fea Se as

Sle Se OS aN Sy Se HENS) eel

CONTENTS.

XV. On the Cranial Osteoloyy, Classification, and Phylogeny of the Dinornithide. By
T, Jnrrery Parkur, D.Se., F.R.S., Professor of Biology in the University of

Otago, Dunedin, New Zealand. (Plates LVI.-LXII.) . . . . page 373
Listsot the Papers:contamediin® Viol Mla’ sar ee Nee eins
Endéxiob Species, Wc. smV ok: hse eae oe eR 2 ee ie Pee

Titlepage and Contents to Vol. XIII.

THE PUBLICATIONS OF THE ZOOLOGICAL SOCIETY OF LONDON.

Tur scientific publications of the Zoological Society of London are of two kinds—* Proceedings,”
published in an octavo form, and ‘“‘ Transactions,” in quarto.

According to the present arrangements, the ‘“‘ Proceedings”’ contain not only notices of all busi-
ness transacted at the scientific meetings, but also all the papers read at such meetings and recom-
mended to be published in the “ Proceedings” by the Committee of Publication. A large number
of colonred plates and engravings are attached to each annual volume of the “ Proceedings,” to
illustrate the new or otherwise remarkable species of animals described in them. Amongst such
illustrations, figures of the new or rare species acquired in a living state for the Society’s Gardens
are often given.

The “ Proceedings ” for each year are issued in four parts, on the first of the months of June,
Angust, October, and April, the part published in April completing the volume for the preceding
year.

The “‘ Transactions’ contain such of the more important communications made to the scientific
meetings of the Society as, on account of the nature of the plates required to illustrate them, are
better adapted for publication in the quarto form. They are issued at irregular intervals.

Hellows and Corresponding Members, upon payment of a Subscription of £1 1s. before the day of
the Anniversary Meeting in each year, are entitled to receive all the Society’s Publications for the
year. They are likewise entitled to purchase the Publications of the Society at 25 per cent. less
than the price charged for them to-the Public. A further reduction of 25 per cent. is made upon
purchases of Publications issued prior to 1871, if they exceed the value of five pounds.

Fellows also have the privilege of subscribing to the Annual Volume of the ‘ Zoological Record ”
for a sum of £1 (which includes delivery in the United Kingdom only), but this privilege only
holds good if the subscription is paid before the First of December in each year.

Such of those publications as are in stock may be obtained at the Society’s Office (3 Hanover

Square, W.), at Messrs. Longmans’, the Society's publishers (Paternoster Row, H.C.), or through
any bookseller.

October 1895, P. L. SCLATER,
Secretary.

913

Matt

* *
basi
7

b
ina
Vf

ey PAS MRT Sa
; cant HA ‘ i SRL Ay hed
it ne f feats , i HEY ; 4 f olde Vai
; ig hy \ Ay A l " J ny
Y ie) 4)

ager =

E Se