TESTUDO EPHIPPIUM. 81 part of the skull. The Albemarle races, 7. microphyes as well as T. vicina, have the outer pterygoid expanded, differing therein from 7. elephantopus and T. nigrita, and agreeing in this respect with 7. ephippiwm, in which, however, the expansion is excessive and the concavity less. Our largest male skull of the North-Albemarle race (Plate XLV. figs. A—C) is 42 inches long: in this skull the foramina palatina are 29 millims. distant from each other, and the posterior roots of the pterygoid edges 25 millims.; so that the convergence of the latter is very gradual. The distance of a foramen palatinum from the anterior extremity of the vomer is 30 millims. The ridges on the base of the skull are obsolete; but the impression in front of the occipital condyle is deep. Parietal crest raised above the surface of the cranium, of moderate length. Articular surface of the mandible very narrow.
In all these particulars our skull of 7. vicina (Plate XLVII. fig. A) agrees with that of the northern form. It is also that of a male, 4 inches long, measured from the front margin of the intermaxillary to the occipital condyle, and 4 inches broad in its widest part, between the zygomatic arches. Its outer tympanic rim has a subcircular outline ; and the ridge which runs from the Eustachian notch (e@) to the stapedial foramen, and to which the columella is attached, is high and rather sharp.
Of the skull of the female (Plates XLII—XLIV. fig. B) I have given a detailed description (Phil. Trans. 1875, p. 276). It differs from that of the male in having the occipital crest and the tympanic case but little produced backwards, being in general configuration but little removed from the young skull, although it is one of a fully adult animal. In the depth and width of the mesopterygoid region, in the expansion and direction of the pterygoid edges, there is no difference whatever from the male. 9.
TESTUDO EPHIPPIUM.
I proposed (Phil. Trans. 1875, p. 271) this name for a species equally well characterized by the peculiar form of its carapace and of its skull. Porrsr’s remarks on the Tortoises of Charles Island (see anté, p. 5) apply so well to this species that I have no doubt that the specimen from which the following description is taken came from that island. If this is really the case, this species is probably extinct. The specimen is an adult male, 33 inches long, stuffed, and belongs to the Museum of Science and Arts, Edinburgh. It was lent to me by T. C. Arcuer, Esq., Director of the Museum, who most kindly allowed the skull and hmb-bones to be extracted, which could be effected without the least injury to the outward appearance of the specimen. Nothing is known of its history. .
A very young stuffed example, 7 inches long, in the British Museum is referred to this species on account of its oblong shape and large smooth areole. The carapace (Plates XX XIX. and XXXI. fig. B) is narrow, oblong, and deep; from the middle of the central dorsal plate to the front margin of the shell the upper profile M
82 THE RACES OF THE GALAPAGOS.
is nearly horizontal, the fore part of the shell being strongly compressed, concave on each side, with the anterior margin strongly reverted—this part of the shell having an appearance which has been so aptly compared by Porrser to a “Spanish saddle.” The hind part of the shell is rounded, with a steep posterior profile, but more gently declivous on the sides, the marginal plates above the hind legs being arched outwards with the edge somewhat reverted, but less so than on the anterior marginal plates. The anterior as well as the posterior margins are irregularly scalloped. ‘The plates are nearly smooth, the areolar portions passing gradually into the striated portions; but the striz themselves are inconspicuous and in many places nearly obliterated. ‘The sternum * is deeply concave, truncated in front and behind, the substance of the caudal plates and of the lateral portion of the abdominals being much thickened. I need not mention the scutellation of the head and legs, none of the Galapagos Tortoises showing any peculiarity in this respect. ‘The tail is very short, and without terminal“claw.”
On comparing the carapace of the young example with that of equally small specimens of other species, we find the areolar spaces larger, the concentric strie deeply sculptured, but less numerous and further apart. Especially the sternal plates are smooth, with the striz partly obliterated. Posteriorly the sternum terminates in a notch (and this appears to be uniformly the case in very young specimens of all the species); but this notch is much shallower than in 7. nigrita, obtuse-angular. The measurements of these two specimens are the following :— Length of carapace. Width of carapace. Depth of Sternum. Caudal plate. Instr.line.Overcurve.Instr.linc,Overcurve.carapace.Length.Width.Length.Width. Spec. in. in. in. in. in. in. in. in. in.
Adult 3475; 33 40 235 40 Airs 24 als 3s Young 7 93 5 84 34 4.3 62 1 17 Skull.—The skull (Plates XLII.—XLIV. fig. C) is comparatively smaller than that of T. elephantopus ; it is 43 inches long, measured from the front margin of the intermaxillary to the occipital condyle, and 53 inches broad in its widest part between the zygomatic arches. The sutures between the various bones can be clearly traced; and growth evidently had not ceased entirely—an observation confirmed by the examination of other bones extracted from the specimen. 1. The frontal region is flat, broad, passing into the very short snout, its greatest width (in front of the postfrontals) being about one half of the distance between the tympanic condyles. 2. The occipital crest is moderately developed, pointed behind, and rising but little above the level of the upper surface of the skull. 3. The tympanic case with the mastoid is produced backwards, the hind margin of the paroccipital forming a rather strong curve (Plate XLII. - *Alargeportioninthemiddleofthesternumhasbeencutoutbythepersonwhopreservedtheanimal, in order to extract the contents of the shell.
TESTUDO EPHIPPIUM. 83 fig. A, a). 4. There is no hollow in front of the occipital condyle, the space between the condyle and basisphenoid simply shelving downwards towards the latter (0). 5. On the front margin of the temporal fossa, in front of the foramen carotidis eaterne, there is a large not very prominent tuberosity (c) for the insertion of a portion of the temporal muscle; no groove separates this tuberosity from the zygomatic arch; or, in other words, the tympanic cavity is not constricted in front. 6. Tympanic cavity very large: the outer tympanic rim ovate, resembling the outline of the human concha, with the convex side in front and the pointed part above; the notch for the passage of the Eustachian tube is very broad, but shallow (e). 7. The ridge which runs from this notch to the stapedial foramen, and to which the columella is attached, is rather low and obtuse. 8-10. The points noticed under these figures in the description of the skull of 7. elephantopus (see page 65) are exactly the same in the present species. 11. The palatal region is very shallow and broad, in consequence of the outer pterygoidedge being flattened down and expanded in its whole length (Plate XLIV. fig. C, f). The triangular space, of which the foramina palatina and the anterior extremity of the vomer form the points, is isosceles in shape, in accordance with the generally short longitudinal axis of the skull. 12. Anterior surface of the tympanic pedicle with a shallow impression. 13. Lower jaw with a double alveolar ridge; its symphysial portion simply vertical, without a backward expansion of the lower margin of the bone. The parts of the angular and coronoid which face each other leave a rather wide cleft between them. Upper margin of the angular not excised.
Limb-bones.—The following bones have been preserved in the large stuffed example, and were extracted from it :—The humerus (Plate LI. figs. B, B’), distinguished by its great length and slenderness; its shaft is trihedral in the middle, and not much bent. The two hollows which we noticed in 7. elephantopus below the head and above the trochlea are here absent. The ulnar tuberosity (a) projects high above the head, which is nearly entirely raised above the level of the summit of the radial tuberosity. The canal (0) for the blood-vessels on the radial edge, close to the elbow-joint, is deep and partly open, cutting off, as it were, a splint from the radial extremity of the bone. T’.
elephantopus, T.
ephippium, 790 millims.
long.
840 millims.
long.
millim. millim.
Lengthofthehumerus(measuredinastraightlinefromthe summit of the head to the middle of trochlea).................. 216 235 Circumference of the narrowest part of the shaft ............... 89 91 longest diameter of the head ....2.2.c.e.ses1 40. -<hensswasssescess 40 40 Shortestidiameter of the head’. ..4.4ccccssscess cess sssaceisssseenen 37 35 Extreme breadth between the condyles........ Ae eee a eae 82 82 The bones of the forearm (Plate LIV. fig. B) are, like the humerus, comparatively slender ; they are remarkably smooth, without prominent ridges or tuberosities. The M 2
84 THE RACES OF THE GALAPAGOS.
ulna has its radial edge but slightly emarginate, and is twisted round its longitudinal axis, so that the transverse diameters of its proximal and distal dilatations would intersect each other at an angle of about 50°. The olecranon is not much developed. The articular facet of the radius for the articulation with the humerus is a rectangular triangle, with the point directed backwards, and the shortest side in front. T.
elephantopus, T.
ephippium, 790 millims.
long.
840 millims.
long.
millim. millim.
Wength ofulnay ca. sinc c00b ses iosteuen es eater eae eee eee 137 155 east: width: of ulna %, «, 70s dcsuste sodas sche eee nae eee 28 26 ueneth of radius4s..\scc 2b hee eee eee ee 12] 149 Least circumference of radius «..0...07..0)..s80ehe eee nee 50 5] Only a few of the carpal bones have been extracted from the specimen, among them the scaphoid and ‘“ intermedium,” which have remained perfectly separate. In the pelvis (Plate LIV. fig. A) we notice, in the first place, that all the sutures are present, and that growth was still proceeding in their vicinity. However, on the whole, the pelvis does not differ in a marked manner from that of 7. elephantopus, except that the symphysial bridge is broader (the obturator foramina consequently narrower) and not so deep. ‘The posterior part of the ossa ilii also is broader than in the other species. Other slight differences of form may be seen from the accompanying comparative measurements :— T’.
elephantopus, T.
ephippium, 790 millims.
long.
840 millims.
long.
millim. millim.
Longestinnerverticaldiameterofpelvis(fromsummitofilium COSYMPHYS1S))ischsessVestenteracaleeleddeletenettehteMexenetates170160 Longest inner horizontal diameter of pelvis...............c.ce0c000s 132 118 Shortestinnerhorizontaldiameterofpelvis(betweeniliopubic prominences) .. 24.4.0 02 1 ue ee eee ashes 112 97 Longest diameter of foramen obturatorium......................4. 4.2 42 Width of symphysial bridge ............ rn en atm aschda bcs 26 35 Depth of symphysial*bridze:= ves sssn eae eee eee 26 23 Least breadth of posterior portion of ossa ischii .................. 61 80 Hiengthiofios iit ob.c.c0e ee eee teencnee eee epee sah Aas aeons 140 130 Theast (breadth ofios alicy. ao .wtat sae ere eee eee 30 27 The femur is very similar to that of 7. elephantopus (p. 68), with the exception of its proximal portion (Plate LIII. fig. B): the head has an elliptical form, and does not rise above the level of the summit of the larger trochanter, as in 7. elephantopus, but is considerably smaller; on the other hand, the cavity separating the head from the trochanters
TESTUDO ABINGDONII. 85 is much larger, as broad as long, and the two trochanters (a@ and 6) are widely separated from each other by a smooth groove. The bones of the lower leg and carpus do not show any noteworthy peculiarity. Asin T. elephantopus, the astragalus and calcaneum are coalesced; but, owing to the lessadvanced aye, the line of separation is still visible. T.
elephantopus, LT.
ephippium, 790 millims. long.
$40 millims, long.
millim. millim. Hen cGhvonet Meme nntann.s ciccuen neces ci waubn act sngleh enms andeatncnen'e tte 169 186 Least circumference of the femur...................600..c0cceeeee ns 80 90 ihonsest diameter ofhead of femur 2... ..0c.-.te0-seesdewes cerceess 55 43 VACmotastesCOMICSIG!tratecltekte0gardahncacieieaticeNeasinehs6667 Wemetnwotsthentibiatasectesectiets.esfecesschedishewdarsateleeeees136150 iieast circumference: Of the: tibia. 2.2.20. c2ncscs.ssecsveccesenesecess 60 72 HemetheotatherNUaeteteMNateliercaeveneadeotioeegialnsdeavalveeldion123138 Least circumference of the fibula .............. ..e..cc ccc eee eee eee 45 45 6.
TESTUDO ABINGDONII. As mentioned above (pp. 5, 6), our knowledge of the existence of these animals in Abingdon was derived from two short statements by Captains James CoLnetr* (1798) and Basit Hari (1822). Since the time of these voyagers down to the present date no further information has been received; and no specimen seems ever to have been brought to Europe before Commander Cooxson’s visit to the island. He succeeded in bringing home the following specimens {, all of which are adult males :— 1. Carapace, 38 inches long, with skull and skin of neck, limbs, &c. Weight 201 Ib. 2. Stuffed specimen with skull; carapace 34 inches. Weight 131 lb. 3. Stuffed specimen with skull; carapace 384 inches. These three examples are entirely identical with regard to the form of the carapace, as well as to the structure of the skull and the other parts which have been preserved. They resemble externally 7. ephippium; but the carapace is of extraordinary thinness and lightness, with the surface deeply pitted’and grooved from the deficiency of osseous substance, which is quite in accordance with the slenderness of the cervical vertebre. For this reason these Tortoises must be much more liable to injury from blows, falls, and other rough treatment; and no doubt Commander Cookson (P. Z. 8S. 1876, p. 523) is quite right in supposing that it is owing to this cause that he could not keep them alive on board. The skull, however, does not participate in this general atrophied condition of the skeleton, but is as stout and compact as in the other species. Commander Cookson has already pointed out some of the most striking distinctive *‘VoyagetotheSouthAtlantic’(London,1798),p.152. +‘ExtractsfromaJournal’(Edinburgh,1824),2ndedit.ii.p.140. + The particulars of their capture are related by Commander Cooxson in Proe.
Zool.
Soc.
1876, p.
520.
86 THE RACES OF THE GALAPAGOS.
features of this species, of which the white colouring of the jaws and upper part of the head and of some of the nails is not the least singular; it is, perhaps, an instance of commencing albinism. ‘The circumstance that no young individuals or females were found must be accounted for by the greater difficulty of discovering small individuals, as well as by the well-known fact that the females are much less active, and more disinclined to leave their hiding-places, than the males.
The anterior portion of the carapace (Plates XL. & XLI.) is narrow, compressed, and deep; the posterior more depressed and dilated. The upper profile is almost straight, and ascending forwards almost from the areolar centre of the last scute, the highest point of the shell being quite in front, so that the anterior aperture assumes the shape of a triangle pointed at the apex. ‘The fore part of the shell is strongly compressed, concave on each side, with the anterior margin strongly reverted, having the appearance ’ of a “Spanish saddle” in a still more marked degree than in T. ephippium. Posteriorly the carapace is dilated, with rounded posterior margin, with a rather steep decline over the caudal plate, which is bent inwards, but more gently declivous on the sides, the marginal plates above the hind legs being strongly bulged outwards, and the posterior marginals being slightly reverted. ‘The anterior margins of the shell are irregularly scalloped and more or less broken, as it is this part of the shell chiefly with which the males butt each other in their combats. ‘The posterior margin of the carapace shows scarcely any indentation. The plates are without any trace of striation ; but the whole surface of the shell is deeply and irregularly pitted. The sternum is deeply concave, truncated in front and behind, with the front lobe rather contracted.
The jaws, some of the nails, and portions of the sternum are of a yellowish white colour. The tail is not very long, and without terminal claw.
Length of carapace. Width of carapace. Depth of Sternum. Postgular plates. Instr.line. Overcurve. Instr.line. Overcurve. carapace. Length. Width. Length. Width. in. in. in. in. in. in. in. in. in.
38 403 224 38 194 26 214 7 12 Skull (Plate XLV. figs. D-F).—The three skulls of this Tortoise are perfectly identical, showing more especially the same singular formation of the palate. ‘The palate is narrow and very concave, bordered by the expanded pterygoid-edges, which rapidly converge behind. Ina skull 4g inches long the foramina palatina are 23 millims. distant from each other, and the posterior roots of the pterygoid-edges 16 millims. The distance of a foramen palatinum from the anterior extremity of the vomer is 32 millims. The inner margin of the pterygoid-edge is continued backwards as a sharp ridge over the base of the skull. The impression in front of the occipital condyle is deep. Parietal crest almost in a line with the upper surface of the skull,and long. Articular surface of the mandible much dilated.
TESTUDO ABINGDONII. 87 Cervical Vertebre.—Of the specimen of which the carapace measures 38 inches in length, and the skull 4% inches, the cervical vertebre (with the exception of the first) have been preserved; in their natural connexion they measure 25 inches, which gives an idea of the slenderness of the neck. They exceed in slenderness those of all the other Galapagos Tortoises as far as we know them at present, and as will be seen from the list of measurements subsequently given. They are of extraordinary lightness compared with their size and length, perfectly normal, and not diseased, as Commander Cookson believed on his first inspection of the animal after its death (/. ¢. p. 521). In the second vertebra (Plate XLVIII. figs. A) the neural arch is compressed and elevated, and provided with a neural crest, the anterior surnmit of which is expanded. The third (Plate XLVIII. figs. B) has a condyle in front, and glenoid cavity behind; its neural crest is low, and split into three ridges; hemal crest well developed. ‘The fourth (Plate XLVIII. figs. C) is biconvex, much longer than the third, with low crests. The fifth (Plate XLIX. figs. D) has a glenoid cavity in front and a condyle behind, and is bulkier, but scarcely longer than the fourth; on its neural aspect it presents a pair of sharp low ridges, confluent in front and divergent towards the posterior zygapophyses, the median line being provided with another still lower ridge; hemal crest very low. The sixth (Plate XLIX. figs. E) is about as long and bulky as the fifth, and has a glenoid cavity in front and a condyle behind, both the glenoid cavity and the condyle showing indications of a bipartition. The neural arch is compressed, with a pair of very low, parallel, close ridges arising from the roots of the zygapophyses, and disappearing towards the middle of the bone. The hzmal surface has a higher crest than any of the preceding vertebre. The seventh vertebra (Plate L. figs. F) has a glenoid cavity in front and behind, and the bipartition of the posterior is nearly complete ; the neural crest is very high, and split into two branches diverging backwards in the direction of the zygapophyses, and leaving a deep triangular recess between them ; as in 7’. vicina, the summit of this crest is split into two prominences, separated from each other by a deep notch. Whilst in the four preceding vertebre the neural arch is deeply hollowed out inwards of and behind each anterior zygapophysis, this place is rather convex in the seventh vertebra of this species. The hamal crest is as developed as in the preceding vertebra, but higher. The ezghth vertebra (Plate L. figs. G), with its bipartite anterior and single posterior condyle, and with its expanded hamate posterior zygapophyses, does not essentially differ from that of the other species; but all the grooves and impressions are still more strongly marked.
The following Table will show the measurements of these vertebra compared with those of 7. elephantopus and T. vicina :—
88 Vertebre............ 2nd. GSS = 8 ‘Ss mm. mm. mm. Length of centrum .. 55 47 62 Depth of centrum in the middle ...... 34 34 34 Horizontal width of middleofcentrum..151415 Width of anterior con- GYI6 Ae siren aces 15 15 24 Width of anterior glenoid cavity ...... Widthofposterior condyle...+sm: Widthofposterior glenoid cavity.... 19 19 21 Distance of outer margins of anterior zygapophyses ...... 23 20 28 Distance of outer margins of posterior zygapophyses ...... 25 28 35 THE RACES OF THE GALAPAGOS. 3rd. SSS g f 5 ‘3 g. cs Seas as N NON mm. mm. mm. 67 65 88 28 26 34 L758 as 20 18 22 20 28 26 34 33 39 26 Soleil Ath. SN g : S 3 ree 'S is 8 mm. mm. mm 8588122 2725733 Ist kf 25} 9 R20 eo Ziliee2o mmol: 35 38 38 ZoeOleOl 5th. SN 5 2 S S 3 Ss ES) 8 mm. mm. mm. 8380114 27 26 34 20 20 22 30 36 34 32 28 31 38 37 43 30 35 50 6th. T.
elephantopus. \ | > | ) BB ea OVE 28 29 34 37 40 29 39 42 28 SS T.
vicina. eH oo oo ie) 29 37 42 53 36 5 T.
abingdonit. 7th. = rs "8 mm. mm. mm. 74 72 99 53 49 68 oT 226 40 41 44 43 50 47 30) HOO mae 46 55 56 E T.
elephantopus. \ 29 54 65 = T.
abingdonit. BB na aron Go F oo =) 39 60 76