TESTUDO DAUDINII. aR rise to the level of the summits of the trochanters, from which it is separated by a broad and deep cavity. The larger trochanter (a) rises to a higher level than the lesser (2), from which it is separated by a shallow incision. A deeper but narrow depression in the outer margins marks the boundary between the trochanters and the head. ‘The lateral portions of the body of the bone below the trochanters show scarcely any impression. The length of the femur of our largest example (¢) is 250 millims., with a least circumference of 128 millims.; the width of the condyles is 120 millims. Of the lower leg and foot no part deserves to be mentioned particularly. The astragalus and calcaneum appear to coalesce at an early stage of growth. And it would appear that in almost all the Round-headed Tortoises the fibula is nearly equal to or even exceeds the ¢id7a in length (see table of measurements, p. 39), whilst in the Galapagos Tortoises the tibia is considerably longer than the fibula.
2.
TESTUDO DAUDINII.
This name was given by Duméril and Bibron (Erpét. Gén. ii. p. 125) to a skeleton (with carapace) in the Paris Museum which seems to agree closely with the example in the British Museum ; it surpasses the latter in size, being 99 centims. (or 394 in.) long, and is stated to have come from the “ Kast Indies ;” probably it came on board a vessel returning from the Indian Ocean.
This species is distinguished by the very peculiar form of its carapace, and in this respect as well characterized as Testudo ephippium from the Galapagos Islands. I have before me two examples, of which one is an adult male (Plate V.), stuffed, in the British Museum, the carapace being 344 in. long. The second is a half-grown stuffed male, with a carapace 23 in. long, and belongs to the Free Public Museum, Liverpool. Fortunately the skulls of both specimens could be extracted; but the remainder of the skeleton is lost. Nothing whatever is known of the history of these specimens; and as no example has reached Europe for a long period, this species may be regarded as probablyextinct.
The carapace (Plate V.) is narrow, oblong, and rather deep, not much broader behind than in front; its upper profile from the nuchal plate to the fourth vertebral is nearly horizontalal, though slightly undulating ; but there is scarcely any downward inclination of the first vertebral plate towards the nuchal. <A second very characteristic point is the strong reversion of the two anterior and the two posterior marginal plates. The caudal plate does not participate in the outward reversion of the marginals, but is rather bent inwards, so as to present a convex surface. The anterior and posterior margins are irregularly scalloped. ‘The vertebral plates, as well as the costals, are perfectly smooth and polished, no trace of an areolar surface being visible ; but those of their margins which are in contact with the marginal plates are coarsely striated, as are _ also the marginal plates themselves. ‘The nuchal plate is of considerable size, broader than long. Caudal plate divided into two unequal portions by an irregular eccentric F
34 THE RACES OF ALDABRA.
suture, which is evidently accidental in this individual. However, it is singular that also Duméril and Bibron describe this plate as divided. The sternum (Plate IV. fig. B) is deeply concave, broad, with the anterior portion somewhat contracted, truncated in front and behind, the substance of the caudal plates and of the lateral portion of the abdominals being much thickened and very callous. The scutellation of the head and legs does not differ from that of the other Aldabra species, the upper part of the snout being protected by the same large plates, and a longitudinal series of large scutes running along the outer edge of the forearm. The tail is long (about 9 inches), terminating in a long, strong, and flat claw. Duméril and Bibron describe the tail of the skeleton in the Paris Museum as “ inonguiculée.” The colour is a uniform deep black. The younger specimen differs in some respects, particularly as regards the form of the carapace ; but the sculpture of its scutes is exactly such as we should expect in an immature individual of this species, and the reversion of the anterior and posterior marginals is already distinctly indicated. The upper profile, commencing from a protuberance near the hind margin of the first vertebral plate and continued to the extremity of the caudal, is convex; that between that tuberosity and the nuchal plate gently slopes downwards towards the latter. The whole carapace is of a regularly elliptical form, not broader behind than in front, and rather deep. ‘The anterior as well as the posterior margins are indistinctly scalloped ; all the upper plates consist of a perfectly smooth areolar portion, which is not much smaller than the scute itself, the striated marginal portion being Very narrow, with the striz shallow and broad. Nuchal plate, as in the adult specimen, large, broader than long; caudal without any division. Sternum slightly concave, with a very shallow notch in front and behind. ‘Tail with a well-developed terminal claw. Colour less deep black than in the adult specimen. The measurements of these two specimens are the following :— Length of carapace. Width of carapace. Depth of Sternum. Caudal plate. In str. line. Over curve. Instr. line. Over curve. carapace. Length. Width. Length. Width. Specimen. in. in. in. in. in. in. in. in. in.
Ad.¢. 344 40 1) 38 16 23 174 4, 7 Vie. 23 29 144 28 103 18 124 12 5 The skull * of the larger of these examples measures in length from the occipital condyle to the extremity of the snout 4} inches, or from the extremity of the occipital crest to that of the snout 5 inches, its greatest width being 3? inches. It is very similar to that of 7. elephantina, but differs from it in the following points:—3. The tympanic case, with the mastoid, is much produced backwards, the hind margin of the pavoccipital forming a strong curve. 5. The tuberosity on the front margin of the temporal fossa is distinctly indicated. 6. The tympanic cavity is large and deep, the outer tympanic rim having a subsemicircular outline, the convexity being anterior. *ProfessorA.Milne-EdwardsinformsmethattheskullattachedtotheskeletonintheParisMuseum evidently belongs to another animal,
TESTUDO PONDEROSA. 39 Very few parts of the remainder of the skeleton have been preserved and extracted from the stuffed example. The wlna has a length of 150 millims. and a width of 29 millims. in its narrowest part; the radius a length of 145 millims. and a circumference of 67 millims., also measured in its narrowest part. Olecranon very prominent, the articular surface forming a steep decline towards that of the radius. Shaft of the radius subcylindrical in its middle. The arrangement of the carpal bones does not differ from that of 7. elephantina. The tibia is 187, the fibula 139 millims. long. 3.
TESTUDO PONDEROSA.
Of this form I have seen one specimen only, the complete skeleton of a female in the British Museum (Plate VI.). It was bought some fifteen years ago by myself for the Zoological Society, in whose gardens it lived for a short time; nothing could be ascertained at the time of its earlier history.
A pelvis of a larger example in the British Museum is referred to this species on account of its complete resemblance to that of the first specimen. The carapace (Plate VI.) is very regularly shaped, a little wider behind than in front, with neither the anterior nor posterior margins reverted, and with shallow notches between the marginal plates. It is rather high, with the sides nearly perpendicular, and with the centre of each of the vertebral scutes raised. Nuchal excision very narrow. Fore part of the shell declivous from the hind part of the first vertebral scute. Caudal scute not reverted, but rather convex. Vertebral and sternal scutes almost entirely smooth, costals and marginals with broad flat concentric striation. Nuchal platenarrow,linear.
‘Sternum truncated in front, with a rectangular excision behind; the lateral edges of its front lobe are conspicuously convex along the whole extent of the postgular scutes. Carapace of a uniform dark horn-colour, nearly black.
The bony substance of the carapace is of considerable thickness in every part, three fourths of an inch in a longitudinal section of the lateral portion of the sternum. The measurements of this specimen are as follows :— Length of carapace. Width of carapace. Depth of Sternum. Caudal plate. Instr. line. Overcurve. Instr.line. Over curve. carapace. Length. Width. Length. Width. in. in. in. in. in. in. in. in, in.
274 363 19 37 144 224 17i 23 64 Osteology.—The skull of this species (Plates VIII. & IX.) differs more from those of the preceding two species than these do from each other. It is particularly distinguished by its relatively inconsiderable breadth and the shortness of the occipital crest. 1. The frontal region is very convex and broad, its greatest width (in front of the postfrontals) being more than one half of the distance of the tympanic condyles (outside measure). 2. The upper surface of the parietals is a flat triangle, the outer edges of F 2
36 THE RACES OF ALDABRA.
which are dilated, converge towards, and finally are merged into, the short tetrahedral occipital crest. 3. The tympanic case (with the mastoid *) is not produced backwards; so that the posterior margin of the paroccipital is but little excised. 4. The groove on the lower surface of the occipital, in front of the condyle, is very shallow. A styloid process (Plate IX. fig. B, a) projects obliquely outwards on each side of the occipital condyle, and appears to be quite peculiar to this species. 5 to 11. In these points the skull of the present species agrees completely with that of 7. elephantina. 12. The palate is deeper than in any of the other species, with a very low median crest along its bottom. The distance between the foramina palatina is conspicuously less than that between a foramen and the anterior extremity of the vomer. 13. Anterior surface of the tympanic pedicle very slightly concave. 14. Lower jaw as in 7. elephantina and LT.
daudinit.
Cervical vertebre.—On comparing the neck-vertebre of 7. ponderosa with those of T. elephantina we find the greatest similarity between them. All the distinctive characters by which the latter differs from the Galapagos species are repeated here; and the first and seventh are the only vertebree which exhibit peculiarities indicative of specific distinctness. In the atlas (Plate XIII. fig. C) the lateral portion of the neural arch is very broad, much broader than the zygapophysis, which is nearly as long as that part of the bone which forms the roof of the arch. The centrum (odontoid process) is longer than broad, a tectiform body with the visceral surface compressed and with a broad articular surface in front and behind. The fourth is the first vertebra of the series which shows a deep hollow in the neural arch behind the anterior zygapophysis to receive the end of the zygapophysis of the preceding vertebra. ‘The dorsal surface of the szath has a very low median ridge, but no crest ; and in the seventh the “ summit” is raised quite to the level of the top of the anterior zygapophyses. Measurementsofcervicalvertebrae:— 2nd. 3rd. Ath. 5th. 6th. 7th.
millim.
millim.
millim.
millim.
millim.
millim.
Weneth offcentram. 4.0 tan tetastess tice een ener 36 53 51 63 66 57 Depthrof centrumin ‘the smiddle sys... eee 27 26 24 26 33 43 Horizontal width of middle of centrum..................... 12 13 13 15 22 24 Wadthof anterior:condylé | y.s...042-1. ee eee 12 16 ae ah ae bes Width of anterior glenoid Cavity.c......0..25..eee eer ae oe 22 22 28 . 30 Wadthvof posterior icondylesee....n-waneree nee vee ee 19 20 24 28 we Width of posterior glenoid cavity .................. ee 15 ae ss we ae 35 Distance of outer margins of anterior zygapophyses...... 24 31 31 33 34. 32 Distance of outer margins of posterior zygapophyses .... 27 29 30 30 30 42 Dorsal vertebre.—Also this part of the vertebral column differs but slightly from that of 7. elephantina, except in the singular shortening of the eighth vertebra, which is coalesced with the two following, the sutures between them having nearly entirely * This bone has been lost by the preparator.
TESTUDO PONDEROSA. oe disappeared. For comparison with that species I give the length of the centra of the severaldorsalvertebrae:— Dorsal vertebre...... Ist. 2nd. 3rd. 4th. Sth. 6th. 7th. 8th. 9th. 10th. 11th. 12th. Length of carapace. mm. mm. mm. mm. mm. mm. mm. mm. mm. mm. mm mm. T. elephantina... 31 in. 45 70 82 86 80 65 48 39 24 19 16 16 T. elephantina... 294in. 48 70 75 76 74 57 40 29 19 17 [lost | T. ponderosa ... 274in. 35 63 70 70 67 55 42 17 #+%18 %&J17 14 16 The number of caudal vertebre is twenty-five, as in T. elephantina, a close and singular approximation to the numbers observed in the Galapagos Tortoises. Throughout this portion of the vertebral column the neural arch is much less thick and high than in 7’. elephantina. The first four vertebree are much more depressed than in that species and have much longer diapophyses. The transition from the transversely elongate condyle of the fourth vertebra to the narrow and globular one of the fifth is abrupt. The sixth and seventh have their diapophyses obliquely directed forwards, whilst these processes are vertical to the axis of the vertebra from the eighth backwards, and gradually diminish in length. ‘There is scarcely a trace of a hemal ridge on the first nine vertebre, the neural surface being longitudinally hollowed out. The last seven are ankylosed as in 7’. elephantina. ‘The length of the entire caudal portion of the vertebral column is 114 inches (that of the carapace being 273); but the shortness of the tail is evidently due to sex. Limb-bones.—The scapulary (Plate XXIV. fig. A) is stout and massive. The angle at which the scapula proper and the acromion meet is very obtuse, about 130°. The body of the scapula is compressed, trihedral in form, a transverse section through its middle resembling exactly that of 7. elephantopus (see p. 67); not only is the anterior surface convex, as in that species, but also one of the lateral surfaces is deeply scooped out. Acromion compressed, twisted round its longitudinal axis. The coracoid is ankylosed to the scapula, with the neck constricted and slightly twisted. The measurements of this bone, compared with those given of 7. elephantina, are the following :— 1.
elephantina. T.
ponderosa. Spec.
c.
Spec.
U.
millim. wmillim. millim. Length of scapula (measured from the suture with the coracoid)... 265 172 160 Cincumiferenhce-im. tts MIMMe:
sas cccc ceed evess cores ce ywecdtbes cocetaneseeneees 145 73 71 Longitudinal diameter-of glenoid cavity ..............cccceceseneseeeeees 99 53 42 Menge throws COrACOIGs «fia stha i eiscc lee caisbecieegis pedecsedadersestawavane ds 150 100 84 Greatest: width:
Of COLACOIG c11i2:iicidadevascacecs dens eeccmtornerenvceteuens 133 80 80 MenoCMsOfACCOMM(schchecc.arstitetiecescctecseceesatdiovedaetaceaaecdes1107884. The humerus resembles so much that of 7. elephantina that it will suffice to give the measurements :-—
38 THE RACES OF ALDABRA.
millim Lengthiof -humerus,s422.<550) sateen te eee eae 180 Circumference of the narrowest part of the shaft .................. 83 Iongest diameter of themeada saute ieee eee tne ee 36 Shortest diameter ofthe head i... eee eee renner 30 Hixtreme breadth between the condyles )).7e.m-.2...0 eee eee 63 The bones of the forearm differ in several respects from those of 7. elephantina and T. daudini. The shaft of the radius is laterally strongly compressed and much narrower than deep. The ulna has its radial margin much curved, so that a wide oval vacuity exists between this bone and the radius. The olecranon is as prominent as ‘in T. daudini, the articular surface forming a steep decline towards that of the radius. The arrangement of the carpal bones is the same as in the other two species. The comparative measurements of the bones of the forearm in the three species are :— T. elephantina. T. daudinu. T. ponderosa.
Spec.c.Spec.7.Spec.&.
millim. millim. millim. millim. millim.
Eengthyof ulna {jks eee ee 195 110 142 150 112 éeastawidthsor winas.)) 1.4... ee: 31 18 22 29 18 Mengthvorsradws se se ee 175 97 130 145 112 Least circumference of radius ...... 85 40 63 67 46 Of the pelvis (Plate XVIII.) two specimens are in the British Museum. All the sutures have disappeared; the iliac bones are broad and stout, and the vertical and horizontal diameters of the pelvis are nearly equal. The lower part of the pubic bones is: gently inclined downwards, prolonged into a tapering lamelliform extremity, and provided below with a sharp median crest. It emits laterally a long, strong process, which is obliquely directed outwards. The posterior part of the ossa ischii is of moderate width, concave above, terminating in a short hamate posterior process on each side, and provided with an exceedingly high symphysial crest below, which, expanding towards behind, forms a large triangular tuberosity. Lateral margin of the ossa ischii excised in the shape of aC. Obturator foramina very wide. Ist spec, 2nd spec.
millim.
millim.
Longest inner vertical diameter of pelvis (from summit of ilium to symphysis) ... 158 116 Longest mner horontal-diameter of pelvis 4 ..01-seeee ee eee eee 164 129 Shortest inner horizontal diameter (between ilio-pubic prominences) ............... 97 82 Longest diameter of foramentobturatomum) 2) s..c eee eee 54 4] Width.of symphysial bridge F225; gmcycerc re en eee eee eee cee 19 10 Depth of symphysial bridges ....do3.ccencie a Sooner ee Le eee 45 30 Teast ‘breadth’ of posterior portion of ossa ischii (iqy-7.ee- eee eee 63 55 Length: of os Tio. 26. she. Ee eo ee 116 100 Jueast breadth. of og. iil; . 3.52 ie ee REE enn 39 27
TESTUDO HOLOLISSA. 39 The femur differs in some measure from that of T. elephantina. Its shaft is much flattened in the direction from the front backwards; a narrow and deep cavity separates the head from the trochanters; the two trochanters are entirely separated from each other; and a deep incision forms the marginal boundary between the head and each of the trochanters. The lateral portions of the body of the bone below the trochanters areconspicuouslyimpressed.
The bones of the lower leg and carpus do not show any noteworthy peculiarity, except that the proximal portion of the tibia is deeply grooved on its hinder surface. T. ponderosa. T. elephantina. T. daudini. Spec.c.Spec.1.Spec.n.Spec.v.
millim. millim. millim. millim. millim. millim. Htemethvof the femur 6... iee. Perse ees 148 250 145 185 Least circumference of the femur ...... 67 128 76 100 MWidthror the condyles... :.:...6..+.6 60 120 66 85 ee Or Meme thvol the tibia. 2. ...us cede ssewtovese 113 170 110 rae 136 137 Least circumference of the tibia......... 48 95 52 ar 70 75 Wench of the fibula. ....c40.$si0x, c0-e ov 108 170 110 one 131 139 Least circumference of the fibula ...... 34 70 37 ve 51 53 4, TESTUDO HOLOLISSA.
This Tortoise may be considered to be the form which is analogous to the 7. microphyes of the Galapagos group. Like that species it is perfectly smooth, even from a comparatively early stage of growth. I have seen two carapaces of males, one being fully adult, the other about half-grown: both are preserved in the Museum of the Royal College of Surgeons (Nos. 1020 and 1021) *. With regard to their history, we learn from the Cat. Osteol. Ser. Coll. Surg. i. p. 198, that the larger (No. 1021) “was a native of the Seychelle Islands, and was being sent to General De Caen, Governor of the Isle of France, in the French corvette ‘ Gobemouche,’ which was captured by Captain Corbet, of H.M.S. ‘ Nereida,’ and the animal brought to the Cape of Good Hope. It was sent to England by Admiral Bertie, who commanded at the Cape, and remained in a living state at Petworth, the seat of the Earl of Egremont, from August 1809 until April 1810. Its weight was 207 pounds.” We have no evidence that any of these Tortoises were indigenous in the Seychelle Islands; on the contrary, it is the belief of the persons best acquainted with this group, that all the individuals kept there in a state of domestication were originally imported from the Aldabra group; and therefore this specimen had probably the same origin as the living female in the Zoological Gardens (see ‘ Nature,’ vol. xii. 1875, p. 260), which _ evidently belongs to the same species. * I am indebted to Professor Flower, who gave me every facility for a careful and repeated examination of these specimens.
40 THE RACES OF ALDABRA.
Nothing is known of the history of the half-grown male, beyond that it was a donation from Sir Joseph Banks in 1810.
The carapace (Plate VII.) is very regularly shaped, scarcely wider behind than in front, with the anterior and posterior margins slightly reverted, and with very shallow notches between the marginal plates. It is rather high, with the sides nearly perpendicular, and with the centre of each of the vertebral plates slightly raised. Nuchal excision very shallow. Fore part of the shell declivous from the centre of the second dorsal plate. Caudal scute not reverted, but rather convex. The plates of the back, as well as the sternum, are perfectly smooth, without a trace of concentric strie. Nuchal plate narrow in its upper, and much dilated in its lower portion. The sternum has no excision, either in front or behind. ‘The lateral edges of its front lobe are somewhat emarginate, not convex. As in ali adult males, the sternum is deeply hollowed out and provided with thick callosities on the sides and behind. This is already conspicuous in the young male.
Both carapaces are entirely of a uniform deep black colour. The bony substance of the shell is not particularly thick. I subjoin the measurements of these three specimens :— Length of carapace. Width of carapace. Depth of Sternum. Caudal plate. Instr.line.Overcurve.Instr.line.Overcurve.carapace.Length.Width.Length.Width. Spec. in. in. in. in. in. in. in. in. in.
Ad. g (1021). 36 474 22 48 16 254 21 3 9 H.-gr. d (1020). 23 284 14 28 MW vs 13 12 5 Ad. 9 (living). 32 394 21 423 163 25 204 34 7