C<D 
Reprinted from The Emu, Vol. 75, Part 2, pp 49-54, April 1975 
THE EMU 
VOL. 75 APRIL 1975 PART 2 
THE FOSSIL RAILS OF C. W. DE VIS, BEING MAINLY AN EXTINCT 
FORM OF TRIBONYX MORTIERU FROM QUEENSLAND 
STORRS L. OLSON 
Received 18 March 1974. 
SUMMARY 
OLSON, S. L. 1975. The fossil rails of C. W. De Vis, being mainly an extinct form of Tribonyx mortierii 
from Queensland. Emu 75: 49-54. 
The fossil rails described by C. W. De Vis from late Pliocene or early Pleistocene deposits (Chinchilla 
Sand) in south-eastern Queensland were re-examined. Fulica prior was found to be synonymous with 
F. atra. All the other rallid fossils represent an extinct form of Tribonyx mortierii, recognizable by its 
slightly smaller size as a distinct subspecies for which the name Porphyrio reperta De Vis has priority. The 
rest of De Vis's names (Gallinultt peralala, G. strenuipes, Tribonyx effluxus, and Porphyrio mackintoshi) 
are synonymized with Tribonyx mortierii reperta. Tribonyx mortierii is believed to have evolved on the 
mainland and spread to Tasmania when the two were connected. At some time after the final opening of 
Bass Strait, the mainland population became extinct, leaving a relic on Tasmania that evolved a slightly 
larger size. 
INTRODUCTION 
Six rails named by C. W. De Vis (1888, 1892) from 
south-eastern Queensland provide the only fossil 
evidence of the Rallidae so far recorded from 
Australia. De Vis's work is highly unreliable; his 
'descriptions are often inadequate and difficult to 
interpret, his figures generally poorly executed' 
(Bartholomai 1966: 115). Consequently, the identity 
of De Vis's fossil rails has been far from certain. 
Through the courtesy of the Queensland Museum I 
have been able to examine all but one of the types, 
as well as some additional referred material of these 
forms. The synonymies resulting from these studies 
not only greatly simplify our notion of Australian 
fossil rails, but shed interesting new light on the 
evolution of the flightless Tasmanian Native-hen 
Tribonyx mortierii. 
THE SPECIMENS: THEIR PROVENANCE 
AND AGE 
The first rallid specimens described by De Vis (1888: 
1277) were 'yielded by the Darling Downs in the 
immediate neighbourhood of Chinchilla, a township 
200 miles [320 km] by rail west of Brisbane. The 
Chinchilla deposits are beds of sand . . . about three 
miles from the township on the north bank of the 
River Condamine . . .' The names Porphyrio reperta, 
Gallinula strenuipes and Fulica prior were introduced 
in this publication. Part of the type material of 
F. prior later (De Vis 1892) became the type of 
Tribonyx effluxus; so it may be assumed that all four 
of these types came from the locality described 
above. Likewise, the referred specimens of P. reperta 
(see below) are labelled 'Chinchilla, Darling Downs.' 
The species Porphyrio mackintoshi, Gallinula pera- 
lata and Tribonyx effluxus were described by De Vis 
in 1892. This paper was evidently intended as a 
continuation of the 1888 account because it is with- 
out an introduction and begins immediately with 
accounts of species. Unfortunately, no definite men- 
tion of localities is made and one wonders therefore 
if the Chinchilla locality mentioned in the introduc- 
tion to the 1888 paper was intended also to apply to 
the specimens described in the 1892 paper. 
A perhaps deceptive allusion to a locality is made 
in the account of Porphyrio mackintoshi, which 
species De Vis (1892: 440) 'dedicated to a gentle- 
man who rendered most kindly aid to the collector, 
Mr. Hurst, during his search for fossils of this kind 
near Warwick.* Note, however, that there is no 
indication that the type itself came from Warwick, 
only that Mr Mackintosh was helpful in looking for 
fossils there. The whereabouts of the type of P. 
mackintoshi is not known. That of Gallinula peralata 
is labelled simply 'Darling Downs.' 
49 
50 S. L. OLSON:  FOSSIL RAILS OF C. W. DE VIS EMU  75 
There is a problem in that there are fossil deposits 
of different ages in Darling Downs, those at Chin- 
chilla being the oldest and those in eastern Darling 
Downs, including the vicinity of Warwick, being 
younger. At least one of De Vis's fossil species 
(Lithophaps ulnaris) is stated definitely to have come 
from Warwick (De Vis 1891a), as was a femur 
assigned to the extinct species Uroaetus brachialis 
(De Vis 1891b). The only certain locality for any 
of De Vis's fossil rails is that at Chinchilla and the 
types of Gallinula peralata and Porphyrio mack- 
intosh! could as well have come from there as 
elsewhere. 
The deposits at Chinchilla were termed the Chin- 
chilla Sand by Woods (1960). East of Chinchilla, 
also along the River Condamine, are several fossil 
localities in fluviatile deposits believed on the basis 
of their marsupial fauna  (particularly Diprotodon 
optatus)  to be Pleistocene in age (Woods 1960). 
The diprotodontid and macropodid faunas of the 
Chinchilla Sand are different from those of eastern 
Darling Downs. These differences are believed to 
signify a difference in age; the geographic proximity 
of the two localities and the widespread nature of 
the latter fauna make it unlikely that they are merely 
of ecological significance' (Woods 1960: 396). For 
these reasons Woods tentatively referred the Chin- 
chilla Sand to the Pliocene. It seems quite possible, 
however, that such faunal changes could occur in 
the Pleistocene. Dr Mary Wade, Curator of Geology 
at the Queensland Museum, informs me {in Utt.) 
that 'the youngest date that can be fastened on to 
Chinchilla Sand is pre-glacial Pleistocene, and we 
don't know vertebrate ranges well enough to tell how 
much older it is.' The fossil birds from these deposits 
that have been restudied have all proved to be 
referable to modern species (Miller 1966a and b; 
Olson this paper) and are thus not inconsistent with 
a Pleistocene age. No doubt most of De Vis's other 
fossil avian taxa can likewise be synonymized with 
modern   forms.  Thus,   for  the  present,   the  most 
accurate age that can be affixed to the Chinchilla 
rails would  appear to  be late Pliocene or early 
Pleistocene. 
Some particulars of De Vis's type-specimens are 
summarized in Table I. In addition to these I have 
also examined the following topotypes identified in 
De Vis's hand as Porphyrio reperta: two left femora 
(QM F7007, F7008), a right tibiotarsus lacking the 
distal end (QM F7009), a very worn proximal right 
tibiotarsus (QM F1127). A distal third of a right 
tibiotarsus with no data but bearing De Vis's annota- 
tion 'C. fulicd was found to fit perfectly onto a 
proximal portion of a tibiotarsus (also with no data) 
identified as P. reperta by G. F. van Tets. The com- 
plete bone is now entered as QM F7030. None of 
these specimens was mentioned in De Vis's publica- 
tions but all are rallid and all pertain to the form 
that ultimately comes under the name reperta. 
THE DISPOSITION OF FUUCA PRIOR 
Fulica prior was originally based on a proximal 
right humerus and a distal right humerus of another 
individual (De Vis 1888). Later, De Vis (1892) 
recognized that the distal humerus was not that of 
a coot and made it the type of Tribonyx effluxus, 
leaving the proximal humerus as the type of F. prior. 
Most of the distinguishing characters of F. prior 
were actually based on the distal humerus, but De Vis 
did not trouble to redefine the species. 
The type of F. prior is indeed from a coot but on 
comparison was found to be inseparable from F. atra. 
De Vis mentioned its smaller size, but the type- 
specimen falls within the lower ranges of the nom- 
inate race F. a. atra (Table II), which is consistent 
with the reputedly smaller size of the Australian race 
F, a. australis. Other alleged differences in F. prior, 
such as the narrower head with more convex articular 
surface, either are not apparent or will not distinguish 
the type from an adequate series of F. atra. As Miller 
(1966a: 185) has noted, De Vis'evidently proceeded 
on the general belief that all fossils should be 
designated as separate species, whether or not they 
differed significantly from their modern relatives.* 
Fulica prior De Vis is here synonymized with Fulica 
atra Linnaeus. 
A MAINLAND FORM OF TRIBONYX MORTDZRD 
Because I believe all of De Vis's remaining types 
and referred material to represent a single species of 
large flightless rail only subspecifically distinct from 
Tribonyx mortierii, all of these specimens are here 
considered together. 
TABLE I 
The type-specimens of fossil rails from Queensland. 
Species De Vis Type element QM Cat. No. 
Porphyrio reperta 1888 Distal right tarsometatarsus F1126 
Porphyrio mackintosh! 1892 Distal right tarsometatarsus Not at QM 
Gallinula strenuipes 1888 Left tarsometatarsus F1128 
Gallinula peralata 1892 Right humerus F1144 
Tribonyx effluxus 1892 Distal right humerus F1138 
Fulica prior 1888 Proximal right humerus F1129 
1975 S. L. OLSON: FOSSIL RAILS OF C. W. DE VIS 51 
The elements of the hindlimb indicate a rail much 
larger and heavier than Gallinula tenebrosa, Tribonyx 
ventralis, Amaurornis olivaceus, Fulica atra or any 
of the smaller Australian rails. Although I was un- 
able to examine a skeleton of Eulabeornis castaneo- 
ventris, the fossil tarsometatarsi are considerably 
larger and heavier than those in skins of that species. 
Furthermore, the fossil form was flightless (see 
below), but E. castaneoventris is not. 
Compared to Porphyrio, all the fossil hindlimb 
elements are much more robust. The femora are 
longer and have much heavier and more curved 
shafts. The tibiotarsi, though shorter than in 
Porphyrio, are nearly twice as thick. The tarso- 
meta tarsus of Porphyrio is highly specialized for a 
rail (Olson 1973a), with the shaft greatly thinned 
proximally, the trochleae more nearly in the same 
proximo-distal plane, the scar for the hallux deeply 
excavated and the distal foramen large and rounded. 
These specializations are also evident even in the 
large flightless form of Porphyrio, 'Notornis' mantelli. 
The Queensland tarsometatarsi show none of these 
characters and clearly are not referable to Porphyrio, 
In all of their more important features these fossil 
hindlimb elements agree perfectly with Tribonyx 
mortierii (Plates 4 and 5). 
Even from De Vis's illustration, the type humerus 
of Gallinula peralata may be recognized as that of a 
flightless rail. This is shown by the low elongate 
flattened head, on nearly the same level as the 
internal tuberosity, the greatly curved shaft, the 
thickened and reduced deltoid crest and the reduced 
bicipital crest, all of which are characters that the 
humeri of most flightless rails assume. The only 
flightless rail in Australia is Tribonyx mortierii, to 
which species the humerus of peralata is similar in 
form and length (Plate 4, Table II). 
The type of Tribonyx effluxus is one of the more 
fragmentary, abraded and perhaps least diagnostic 
of De Vis's rallid types. Ridpath (1972: 118) studied 
this specimen and suggested that it may have come 
from a flightless species but he did not think that it 
was a 'sufficiently characteristic part of the skeleton 
upon which to decide the fossil's relationships.' The 
brachial depression is too small for Porphyrio and the 
shaft is too heavy for Tribonyx ventralis, Gallinula 
tenebrosa or Fulica atra. Except for having a more 
slender shaft, the specimen is very similar to the type 
of G. peralata and it agrees well with Tribonyx 
mortierii. As noted by Ridpath {1972: 118) the 
'radial trochlea* (= external condyle) is somewhat 
more oblique to the long axis of the shaft than in 
T. mortierii but the difference is slight and 'the shape 
of the radial trochlea varies appreciably between 
different individuals of T. mortierii.' There is no 
reason to suppose that the type of T. effluxus came 
from a species different from the rest of the Chin- 
chilla fossil rails. 
TABLE II 
Measurements (mm) of Tribonyx m. mortierii compared with T. m. reperta and of 
Fulica atra compared with F. prior. * = estimated. 
Proximal. Width Least Width Shaft Distal Width Length 
range mean s.a. range mean s,d. range ma* s.d. range            mean s.d. 
Femur mortierii  (n-17) It.9-20.0 18.3 .70 5.9-7.3 6.4 .45 15.2-18.4 17.1 .72 78.0- 87.2      S3.9 2.57 
Femur reperta OM F7008 1(1.1 6.1 15,1 80.0 
Famur reperta OK F7O07 16.9 ft. 3 15.2 72.3 
Tibia aortierii  (n-16) 13.3-16.1 14.8 .70 5.7-6.7 6.2 .40 11.6-12.9 12.4 .54 123.0-140.9    133.6 5.12 
Tibia renerca OM F7009 15.1 6.5   128* 
Tibia mitu OM F7029 13.3 5.9 12.0 121 
Tataua mortl?rii  (n-11) 12.7-14.4 13.3 .76 5.3-6.0 5.7 .33 12,7-14.3 13.5 .63 75.0- 90.7      83.1 4.86 
Tarsus riKru QM F112B 12.6 5.6 12.1 74.6 
Tarsus reperts QH F1126 
  
5,5 12.2 70* 
Humerus mortierii (n-14) 13.9-15.5 14.5 .6ft 4.2-4.7 4.5 .20 9.3-10.3 9.9 .26 60.7- 68.4      65.0 2.25 
Humerus reperta QH F1144 14.2 5.0. 10.4 62.8 
Kumerua reperta QM F1138   4.5* 9.9   
Humerua Fulica atra (n-13) 13.5-15.8 14.8 .90 4.2-4.9 4.5 .28     
Humerus Fulica prior 13.ft 4.7     
52 S. L. OLSON:  FOSSIL RAILS OF C. W. DE VIS EMU 75 
The distal end of a tarsometatarsus forming the 
type of Porphyrio mackintoshi was apparently never 
placed in the collections of the Queensland Museum, 
and I have been unable to locate it. Although De Vis 
(1892: 440) compared it only with Porphyrio he 
indicated some uncertainty as to its proper place- 
ment by saying that 'it is probable that this [species] 
and P. reperta, m? will eventually be placed in a 
new genus, as in both the hind toe is less elevated 
than in the recent genus, and the inner trochlea 
(imperfect in the cotype [sic] of P. reperta) is found 
in the present fossil to be distinctly shorter, or rather 
not to extend so far distad . . .' These characters are 
found in Tribonyx mortierii, and along with De Vis's 
illustration (Plate 4, inset), show definitely that 
mackintoshi is not referable to Porphyrio. The only 
difference De Vis noted between mackintoshi and 
reperta was the larger size of the former. In view of 
the great variation in size in modern Tribonyx 
mortierii and in the Queensland fossil form (Table II, 
Plates 4 and 5), and also in view of the abundance 
of the latter in the Chinchilla deposits, it seems safe 
to assume that mackintoshi, too, belongs to the 
extinct form of Tribonyx mortierii. 
Though all the Queensland fossil rails, except the 
type of Fulica prior, appear to be referable to the 
species Tribonyx mortierii, not all agree exactly with 
the modern Tasmanian form, even though there is a 
great deal of individual variation in the size and 
proportions of the elements of the limbs of modern 
mortierii (Plates 4 and 5). The two fossil femora are 
smaller than modern specimens with similar propor- 
tions (Plate 5). One of these is absolutely smaller 
than any modern specimen (Table II). One of the 
fossil tibiotarsi is inseparable from modern mortierii, 
but the other is just below the minimum for that 
form. The fossil tarsometatarsi are likewise slightly 
smaller than in the modern form and the proximo- 
anterior part of the shaft in the type of strenuipes is 
somewhat more excavated. The fossil humeri are 
within the range of size of mortierii. The shaft of 
peralata is considerably heavier than in any modern 
specimen (Plate 4) but appears as if it may be from 
an abnormal individual. The shaft of effluxus is not 
any heavier than that of modern mortierii. 
The differences shown by the fossils (mainly 
smaller size) I do not consider to be of more than 
subspecific value (see discussion). Of the five names 
applicable to the fossil form, Porphyrio reperta and 
Gattinula strenuipes are the two oldest and were 
proposed simultaneously. As first reviser, I choose 
reperta because of page priority and in view of the 
more widespread use of this name, at least in labelling 
specimens, Gallinula strenuipes, Gallinula peralata, 
Tribonyx effluxus, and Porphyrio mackintoshi thus 
become synonyms of Tribonyx mortierii reperta. 
Elsewhere  (Olson  1973b), I have reduced Tri- 
bonyx to a sub genus of Gallinula, but I have used 
Tribonyx in the present account for the sake of 
clarity. The extant Tasmanian Native-hen ought then 
to be known as Gallinula mortierii mortierii (Du 
Bus) and the extinct mainland form as Gallinula 
mortierii reperta (De Vis). For this reason I have 
retained the spelling reperta throughout, although if 
used with either Porphyrio or Tribonyx it should be 
rendered repertus, as already noticed by Brodkorb 
(1967). 
DISCUSSION 
Because the volant Black-tailed Native-hen Tribonyx 
ventralis occurs on the mainland of Australia but not 
on Tasmania, and because T. mortierii is found on 
Tasmania but not on the mainland, it has usually 
been thought that mortierii is an insular derivative 
of ventralis (e.g. Mathews and Iredale 1921). How- 
ever, Ridpath and Moreau (1966) thought that the 
differences between them suggested that the two 
species had been separated for some time. The 
late Pliocene or early Pleistocene occurrence of 
T, mortierii in Queensland, documented here, not 
only confirms the antiquity of the separation of 
mortierii and ventralis, but also shows that mortierii 
is evidently not a Tasmanian autochthon. On New 
Zealand there was another flightless species, Tribonyx 
hodgeni, now extinct, that was nearer in size to 
ventralis but more similar in cranial morphology to 
mortierii (Olson in press). Tribonyx hodgeni prob- 
ably arose from an ancestor closer to the common 
ancestor of both mortierii and ventralis than to 
ventralis itself. This is further evidence that ventralis 
is not a primitive or ancestral stock. 
Because Bass Strait was in existence in the Mio- 
cene (Ridpath and Moreau 1966) and because 
T. mortierii was in Queensland before Pleistocene 
glaciations caused lower sea-levels, T. mortierii prob- 
ably would not have evolved on Tasmania and spread 
from there to the continent. Successful colonization 
of mainland areas by insular species is rare in any 
case. 
It is not certain just how the ancestral populations 
of Tribonyx in Australia became geographically 
separated so as to evolve into two species. X. ven- 
tralis, in any event, became more aquatically adapted 
and now inhabits temporary waters in semi-arid 
country characterized by erratic rainfall (Ridpath 
1972; Ridpath and Moreau 1966). As a consequence, 
it is highly nomadic and retains a need for its power 
of flight. T. mortierii, on the other hand, became 
more terrestrially adapted and evidently evolved in 
a more equable and constant environment such as 
that of the present coastal areas of eastern Australia. 
It must have reached Tasmania when that island 
was connected to the mainland and was being colon- 
ized by such animals as emus Dromaius and various 
54 S. L. OLSON:  FOSSIL RAILS OF C. W. DE VIS EMU 75 
 . In press. A review of the extinct rails of the       ? -, and R. E. MOREAU. 1966. The birds of Tas- 
New Zealand region  (Aves: Rallidae). Rec. natn. mania: ecology and evolution. Ibis 108: 348-393. 
Mus. NZ. WOODS, J. T. 1960. Fossiliferous and cave deposits. In 
RmPATH,   M,   G.   1972.  The  Tasmanian  Native  Hen, Hill, D. and A. K. Denmead (Eds.) The Geology 
Tribonyx mortierii. III. Ecology. CSIRO Wildl. Res. of Queensland. Melbourne Univ. Press. (Originally 
17: 91-118. published as Vol. 7 of J. geol. Soc. Aust.) 
DR STORES L. OLSON, National Museum of Natural History, Smithsonian Institution,  Washington, DC 
20560, USA. 
1975 S. L. OLSON: FOSSIL RAILS OF C. W. DE VIS 53 
large marsupials. After Tasmania was finally isolated 
from the mainland about 12,000 years ago (Ridpath 
and Moreau 1966), T. mortierii survived there but 
became extinct in Australia. This pattern of distribu- 
tion (Fig. 1) is closely paralleled by that of such 
mammals as the Thylacine Wolf Thylacinus cyno- 
cephalus and the Tasmanian Devil Sarcophilus 
harrisi, which became extinct in pre-European times 
on the mainland but persisted later in Tasmania. 
The larger size of T. mortierii mortierii has prob- 
ably arisen since the final isolation of Tasmania. 
Insular forms of many species commonly become 
larger than their mainland counterparts, and the 
differences in size between mortierii and reperta are 
very likely a result of this effect. 
ADDENDUM 
After the present manuscript had been submitted 
and accepted, P. V. Rich and G. F. van Tets called 
to my attention the fact that certain other bones 
among the De Vis material might be referable to 
Tribonyx mortierii reperta. I was able to investigate 
this personally and discovered a few more such bones 
on a recent visit to the Queensland Museum. A distal 
end of a right femur (QM F7022), which De Vis 
had labelled 'Megapodius', is from T. m. reperta 
(distal width 15.1 mm). A distal half of a left 
humerus (QM F7058) from Chinchilla identified in 
De Vis's hand as the fossil species 'Anas elapsd is 
also from T. m. reperta. Fortunately De Vis never 
published on either of these two specimens. The 
humerus, like that of the type of 'Gallinula peralata', 
is heavier than in modern T. m. mortierii (distal 
width 11.0, least width shaft 5.2) so that it appears 
as if this may be a valid character for T. m. reperta. 
Two distal portions of right tibiotarsi (QM F5554 
and F5555) that De Vis (1892: 444) referred to 
his extinct species 'Ptatalea subtenuis' are from 
T. m. reperta. These are labelled simply 'Darling 
Downs'; the abovementioned femur has no locality 
data. De Vis based most of his description of 
Platalea subtenuis upon the proximal two-thirds of 
a right femur (QM F1140), which may now be 
regarded as the lectotype. This femur is definitely 
not rallid. The total number of fossils of T. m. reperta 
thus far identified from the De Vis material is four- 
teen, and these represent at least four and probably 
five individuals; so the species may be assumed to 
have been rather abundant. 
ACKNOWLEDGEMENTS 
This study was made possible through the generosity 
of Drs Mary Wade and Alan Bartholomai of the 
Queensland Museum (QM) who made the fossils 
available and answered many questions, I am most 
grateful to M. G. Ridpath (MR), CSIRO, A. R. 
McEvey, National Museum of Victoria (NMV), 
and N. K. Johnson, Museum of Vertebrate Zoology 
(MVZ) for lending comparative material. G. F. van 
Tets drew my attention to some fossil specimens and 
provided valuable comments. The manuscript has 
benefited considerably from critical reading by 
M. G. Ridpath, J. Farrand, Jr, and R. J. Emry. 
The photographs are by V. E. Krantz. 
Figure 1. Distribution of the modern and fossil forms 
of Tribonyx mortierii. T. m. mortierii (solid) 
and T. m. reperta (x). 
REFERENCES 
BARTHOLOMAI, A. 1966. The type specimens of some of 
of De Vis' species of fossil Macropodidae. Mem. Qd Mus. 14:  115-126. 
BRODKORB, P. 1967. Catalogue of fossil birds:  Part 3 
(Ralliformes, Ichthyornithiforrnes, Charadriiformes). 
Bull. Fla St. Mus. biol. Sci. 11: 99-220. 
DE VIS, C. W.  1888, A glimpse of the post-Tertiary 
avifauna of Queensland. Proc. Linn. Soc. NSW (2) 
3: 1277-1292. 
 1891a. On the trail of an extinct bird. Proc. Linn. 
Soc. NSW (2) 6:  117-122. 
  1891b. Note on an extinct eagle. Proc. Linn. Soc. 
NSW (2) 6: 123-125. 
 1892. Residue of the extinct birds of Queensland 
as yet detected. Proc. Linn. Soc. NSW (2) 6: 437- 
456. 
MATHEWS, G. M., and T. IREDALE. 1921. A Manual of 
the Birds of Australia. London: Witherby. 
MILLER, A. H. 1966a. The fossil pelicans of Australia, 
Mem. Qd Mus. 14: 181-190. 
  1966b. An evaluation of the fossil anhingas of 
Australia. Condor 68: 315-320. 
OLSON, S. L. 1973 a. Evolution of the rails of the South 
Atlantic Islands (Aves: Rallidae). Smithson. Con- 
tribn Zool. 152:  1-53. 
  1973b. A classification of the Rallidae. Wilson 
Bull. 85: 381-416. 
PLATE 4 
I 1 
71,ml,in mil,i lllllllllll 
Plate 4.   Limb bones of modern and fossil rails (a-d, tarsometatarsi; e-j, humeri), a Tribonyx 
m. mortierii (MVZ 140780); b T. in. reperta (type of Porphyrio reperta); e T. in. 
rvnortr,     Ctypp     nf     Pnrnhyri,-,     ,-op,>rt,,\-     C     f.     111.     mOrtlCm     (NMV     B5226)     nOte    the 
difference from a; d T. in. reperta (type of Gallinula strenuipes); e T. in. mortierii 
(MR 1012; f T. in. reperta (type of Gallinula peralata); g T. m. mortierii (MVZ 
140780) note difference from e; h T. m. reperta (type of Tribonyx effluxus); i type 
of Fulica prior; j Fulica atra atra (USNM 318530). Inset: type tarsometatarsus of 
Porphyrio mackintoshi (from De Vis  1892, pi. xxiv, figs 2a and 2b), scale =  1 cm. 
PLATE 5 
B 
H 
*A 
Plate 5. Limb bones of modern and fossil Tribonyx mortierii (ad, tibiotarsi; e-h, femora). 
a T. m. reperta (QM F7030); b T. m. mortierii (NMV B5226); c T. m. reperta 
(QM F7009); d T. m. mortierii (USNM 18408); e T. m. reperta (QM F7008); 
f T. m. mortierii (MVZ 140780); g T. m. reperta (QM F7007); h T. m. mortierii 
(NMV B5226).