eee HARVARD UNIVERSITY ail LIBRARY OF THE Museum of Comparative Zoology HARVAGD MSR FT | UAIVERSIT| AP it embed YALE PEABODY MUSEUM oF Natural History Number 61 May 10, 1962 New Haven, Conn. a THE AFFINITIES OF THE PINK-HEADED DUCK (RHODONESSA CARYOPHY LLACEA ) Puinie 8S. HuMPHREY AND S. Ditton RIPLEY The rare or extinct Pink-headed Duck of India has had a checkered taxonomic history, having been placed at one time or another with the perching ducks (Cairinini), the dabbling ducks (Anatini), and the pochards (Aythyini). Delacour and Mayr (1945:23-24) in their brilliant revision of the family Anatidae considered Rhodonessa as belonging to the tribe Anatini because of similarities in display and posture. Later, Delacour (1956:197) stated that it “is probably related to Anas more nearly than to any others, but it may also have some connection with the pochards, as it somewhat approxi- mates in proportions the species of Netta, and it has a similar trachea. It certainly shows no close relationship to the Woced Ducks (Cairinini).” Verheyen (1955:22) places Rhodonessa with the pochards, an alliance which had been suggested earlier by Garrod (1875 :153-154) on the basis of the trachea. Peters (1931:170) put Rhodonessa between Malacorhynchus (Pink- eared Duck) and Aix (Wood Duck) in his subfamily Anatinae, which also included the dabbling ducks and a variety of other forms. Phillips (1922:90-93) and Salvadori (1895 :61-63) placed the genus in a subfamily Plectropterinae among genera which are now considered to be perching ducks (Cairinini). 2 Postilla Yale Peabody Museum No. 61 Most recently Johnsgard (1961:78,80) has recognized the aythyine affinities of Rhodonessa; he considers Rhodonessa a connecting link (along with Marmaronetta) between the Ay- thyini and the Anatini. Woolfenden (1961 :114), using several osteological features as evidence, has placed Rhodonessa in the tribe Aythyini. Rhodonessa has had an uncertain status because it combines some of the characters of two very different groups of water- fowl: the pochards on the one hand, and on the other, ducks which are better. adapted for a more terrestrial existence, namely the dabbling ducks and perching ducks. This combina- tion of characters has led some workers to suggest that the Pink-headed Duck might be a “link”? relating in a phylogenetic sense the pochards and the dabbling ducks. Humphrey’s interest in this problem was aroused when it was noted that the trachea of the male Pink-headed Duck is very similar to tracheae of males of species in the tribe Ay- thyini, differing from these only in small details. This striking morphological similarity and Ripley’s interest in the curicus distribution of Rhodonessa and other Indian birds have prompted us to investigate further the affinities of this puzzling genus. ACKNOWLEDGMENTS Humphrey’s part of this study was supported by the Na- tional Science Foundation as part of a project on the anatomy of the trachea of ducks and the classification of that group. Anatomical specimens examined during the course of this study were obtained through the generosity of M. C. Downes (Department of Fisheries and Game, Australia), H. J. Frith (Wildlife Survey Section, C.S.I.R.O, Australia), R. P. Gros- senheider, G. S. Hunt, G. V. T. Matthews (The Severn Wild- fowl Trust), K. C. Parkes, W. H. Partridge, Peter Stetten- heim, the Departments of Conservation of the states of Michi- gan and Washington, the Department of Animal Pathology (Cambridge, England), and the University of Washington Marine Biological Station at Friday Harbor; this study would not have been possible without the invaluable help of these people and institutions. We are indebted to the British Museum of Natural History, the American Museum of Natural History, “O" COMP. 2001 LisRARY MAY 31 1962 HARVARD May 10, 1962 The Pink-Headed Duck the United States National Museum and the Universit Michigan Museum of Zoology for the use of specimens in their care. Many of the specimens examined during the course of this study were first obtained while Humphrey was at the University of Michigan Museum of Zoology; we are especially grateful to R. W. Storer and the late Josselyn Van Tyne for allowing this material to be incorporated in the collections of the Yale Peabody Museum of Natural History. Our thanks must also go to Mrs. Shirley Hartman for pre- paring the illustrations and to Mr. George EK. Watson for many helpful suggestions. STRUCTURAL FEATURES Little is known of the anatomy of the Pink-headed Duck. Garrod (1875 :153-154) described and figured the trachea and syrinx of both sexes; Verheyen (1955:22) and more recently Woolfenden (1961 :14, 41, 52, 54, 114) have commented on the osteology of the species. Plumage pattern and general appearance. he Pink-headed Duck was a long-necked, rather awkward looking bird and at first sight, little hike a pochard in bodily proportions. The pos- ture of the species and its display habits (discussed below) have led some authors, notably Delacour and Mayr (1945), to ally it with the dabbling ducks. A general comparison of the plumage patterns of the Pink- headed Duck and all other waterfowl leaves us with the impres- sion that the species has in this character more in common with the pochards than with any other group. The coloration of the Pink-headed Duck (apart from the pink head and _ neck, which are in color unique among waterfowl) is very much like that of the pochards. The similarity is especially noteworthy in the pattern of coloration of the wing (Ripley, 1950 :903-904). Rhodonessa lacks an iridescent speculum, and in fact has sec- ondaries which are practically identical in markings with those of many pochards. Trachea. The trachea of the male Pink-headed Duck is in its general features indistinguishable from the tracheae of males of the tribe Aythyini. However, it differs in the ag- 4 Postilla Yale Peabody Museum No. 61 gregate of several details of structure from the tracheae of males of any of the pochards. We have examined specimens or figures of tracheae of males of all species in the tribe Aythyini except Netta erythroph- thalma, Aythya nyroca, and Aythya novae-seelandiae. Through the courtesy of Dr. James D. McDonald of the British Museum of Natural History we have been able to study a specimen of the caudal part of the trachea of a male Pink- headed Duck. Males of Rhodonessa and the various species of Aythyini are alike in general conformation of the syringeal region. The similarity is most striking in- 1) the form of the partly bony, partly membranous swelling or dilatation to the left, and 2) the conformation of the laterally expanded, partly fused rings antericr to the tracheo-bronchial junction (Figure 1). In form of the dilatation to the left, male Rhodonessa differs from males of the tribe Aythyini as follows: 1) the membrane-covered fenestrae are poorly developed. 2) the dilatation is not as strongly laterally compressed as in the Aythyini. 3) the dilatation does not extend as far anteriorly (cepha- lad) as in the Aythyini. 4) the lateral plane of orientation of the dilatation is dorsal to ventral not dorso-medial to ventro-lateral as in the Aythyini. 5) the dilatation is more expanded or swollen caudally than in the Aythyini. In every respect save the last, the dilatation of the caudal end of the trachea of Rhodonessa is less well developed than in the Aythyini. In Rhodonessa this dilatation is clearly a some- what less elaborate version of the same structure in the Ay- thyini. Rhodonessa resembles no other group of ducks in this respect. According to Garrod (1875:154), the trachea of the male Pink-headed Duck has ‘ta shght fusiform dilatation” anterior to the syringeal region. Mid-tracheal swellings of one kind or another occur commonly in only two of the major groups of waterfowl, the Mergini (Bucephala, Mergus, Histrionicus, Melanitta) and the Aythyini (several species of Aythya, May 10, 1962 The Pink-Headed Duck 5 Figure 1. Caudal ends of tracheae of males of A) Rhodonessa caryophyl- lacea, B) Metopiana peposaca, and C) Aythya affinis; each specimen drawn in the following views: 1) dorsal, 2) left lateral, 3) caudal, and 4) ventral. Magnification x 1.18. 6 Postilla Yale Peabody Museum No. 61 Netta). One, and possibly more, species of Anas (Anas versi- color, but not all species are known anatomically) have a mid- tracheal swelling. Garrod (1875:154) describes the caudal end of the trachea just anterior to the syringeal region of the male Pink-headed Duck as follows: “the lower end of the trachea is hardly con- tracted at all. There is, however, a slight thinning of the anterior portions of some of the inferior tracheal rings. ... a small, transverse, anterior fenestra being the result.” Garrod’s figure of the trachea of the male of this species illustrates 13 such fenestrae. A similar modification of the ventral parts of some of the more caudal tracheal rings occurs in male Clangula (tribe Mergini), but there are in that species only seven fen- estrae. Males and females of Sarkidiornis (Cairinini) have fen- estrae of this kind in the caudal part of the trachea; these fenestrae are fewer and less well developed in the females. The structure of the syringeal region of the tracheae of males of the genera Aythya, Netta, and Metopiana is peculiar to the group. From the standpoint of the structure of the male syrinx, Rhodonessa clearly belongs in the tribe Aythyini. The syringes of males of the tribes Anatini and Cairinini have much in common structurally and differ significantly from those of Rhodonessa, Aythya, Netta, and Metopiana. Humerus. Woolfenden (1959:184) has described a method of distinguishing “the humeri of the Anatinae from those of the Aythyinae [classification of Peters, 1931], based on certain characters of the pneumatic fossa .... In the Anatinae the fossa is deeper and partially excavates the medial bar. The construction is such that the palmar surface of the bar is not completely visible. Furthermore, the fossa usually possesses many bony struts. In the Aythyinae the pneumatic fossa is shallower, and the medial bar is essentially continuous with the shaft, exposing its palmar surface. Struts within the fossa are rare; in most cases the wall is solid.” Woolfenden found that the pneumatic fossa of Metopiana peposaca agrees “in all respects with those of the Anatinae.” He says further that “this deviation from what seems a reliable method of distinguishing the two subfamilies may be of phylo- May 10, 1962 The Pink-Headed Duck 7 gentic significance, for Delacour and Mayr (1945 :25-26) consider Metopiana, along with Netta rufina and Aythya erythrophthalma, to ‘constitute a bridge between the river ducks and the more specialized pochards of the genus Aythya....” We have examined the pneumatic fossa of the humerus of Anas platyrhynchos, A. fulvigula, A. falcata, A. poecilorhyncha, Aythya marila, A. fuligula, A. ferina, Metopiana peposaca, d Figure 2. Head of left humerus, anconal view; a) Rhodonessa caryophyl- lacea, b) Aythya marila, c) Metopiana peposaca, da) Mergus serrator, and e) Anas platyrhynchos. Magnification x 1.75. Rhodonessa caryophyllacea, Histrionicus histrionicus, Mela- nitta fusca, M. nigra, Somateria mollissima, Bucephala albeola, Mergus serrator, M. merganser, and Aix sponsa. Using Wool- fenden’s criterion for classifying pneumatic fossae into “aythy- ine” or “anatine” types, we find the following: *“Anatine” pneumatic fossa: Anas, Metopiana, Rhodonessa, Mergus, Aix. (Figure 2.) “Aythyine” pneumatic fossa: Aythya, Bucephala, Histrion- icus, Melanitta, Somateria. 8 Postilla Yale Peabody Museum No. 61 The fact that Mergus has an “anatine” pneumatic fossa suggests to us that this character has undoubtedly arisen in- dependently in three or more different groups (Mergini, Aythy- ini, Anatini). Delacour and Mayr (1945) have pointed out the close relationship between Bucephala and Mergus; Humphrey (1955) has shown that Mergus probably evolved from a Bucephala-like ancestor. We feel that the “aythyine” pneuma- tic fossa of Bucephala is evidence that the ‘ ‘anatine”’ pneumatic fossa of Mergus is a derived condition which in no way indicates relationships with the tribe Anatini. Therefore, the Anas-like condition of the pneumatic fossa of Metopiana does not neces- sarily indicate that this genus has any close affinity to the Anatini. In view of the foregoing, the “anatine” condition of the pneumatic fossa of Rhodonessa cannot be used as evidence to clarify the relationships of this genus. Feet. The feet of the Pink-headed Duck have a number of characters in common with those of Anas, e.g., lack of a lobe on the hallux, digits III and IV approximately equal in length or digit III slightly longer, digits relatively short compared to length of humerus. However, there are some features of the foot of Rhodonessa which suggest that its resemblance to the feet of dabbling ducks is secondarily derived. The fact that the Pink-headed Duck, by its tracheal anatomy ebviously a pochard, has feet like a dabbling duck prompted us to compare the feet of dabbling ducks and pochards. To this end we measured skeletal elements of the feet of the following species : Anas discors 6 Aythya collaris ¢ ” acuta ¢ i americana é * querquedula & 2 ferina * gibberifrons é a valisineria ¢ ” cyanoptera ¢ eg affints 8 4 ” rubripes 10 4 Metopiana peposaca 4 Aythya marila é Netta rufina if fuligula é Rhodonessa caryophyllacea & 399 nyroca & May 10, 1962 The Pink-Headed Duck 9 Measurements of these specimens are presented in Table 1 which also includes means, standard deviations. minima and maxima of the measurements of samples of Anas rubripes and Aythya affinis. As is apparent from examination of specimens of dabbling ducks (Anas) and pochards (Aythya), these two groups of waterfowl differ in size of foot, the pochards having relatively - much larger feet than the dabbling ducks. Using greatest length of humerus as an index of general body size, the greatest length of the tarsometatarsus and of each digit (minus ungual pha- lanx) was expressed as a per cent of humerus length. These ratios are presented in Table 2 where it can be seen that in every case except length of tarsometatarsus the ratios of the elements of the foot of Rhodonessa are much smaller than those of Aythya and fall among those of the Anas group. The tarsometatarsus of Rhodonessa is relatively somewhat shorter than that of any dabbling duck but rather long for a pochard. although these of some pochards (Aythya nyroca, A. valisi- neria, Metopiana) are about the same relative length or slightly longer. In common with the dabbling ducks and Rhodonessa, Metopiana has relatively short digits: except for the hallux, the digits of Netta rufina are also very short. The relative length of the tarsometatarsus of Netta is short like that of the more typical pochards. Possibly there is a difference in the relative lengths of the humeri of dabbling ducks and pochards correlated with dif- ferences in the flying abilities of the two groups. Although we know of no way of testing for this possibility, we doubt that there is enough of an adaptive difference in relative length of the humerus in the two groups to invalidate using length of this element as a measure of body size and as a means of com- paring the relative lengths of the elements of the foot. There is a clear cut difference between the species of Anas and those of Aythya in the relative lengths of digits III and IV. In Aythya (and Netta rufina) digit IV is longer than digit III (see Table 3). In Anas digit III is usually longer but may be equal to or slightly shorter than digit IV. In Anas rubripes (sample of ten males) digit IV is usually slightly shorter (up to 3.6 per cent shorter) than digit III: in five out of ten speci- No. 61 Postilla Yale Peabody Museum 10 OES eral GSI 621 LST ez wale Sie sho neh tris ker iol ehisivelnetce é voor Aydodaeo Bssouopoy yy 6'EI G61 OFT CLT 0'FZ OSI (kr fak MOOS 2 Pp aclan sone cae 2 vavsodad vuridoyof FT 6°61 0°91 GLI it ic Z'61 Cage Se Se ie os SRO VENA VIN Sil SOT Lit SST 0°06 OF COG -"° a aes meee ge SG rem 006 OST T&T 8°SG 9ST OUR esr en te ae se eee mee COLL COLE LeL GOS OL Cele Ole Gol = Gl twice = Wibarouh a Lil ec Gi te ote eo bho eee S LULL 691 6'€% SLT 80% 9°82 raat OGG) 2) ae i ae ey ELOISE oP 0'0% O'ST Isl (GZ 9'8T ZGZ wae os pnoooOFoOoOKDOnoGoOob MOD 6 BULLY GST POG LST [81 0'GS O61 COG ee et eee ey BUbO OnE SPL L'6I GPL SOL nd eT Z&s SOs Oh ceo . @ vie) ese) 8) 6) . . P SLIB][OoO coal | Gary CEL GG] Tez Q'Gl L0G o- . efvvike] (eifelie)is/ice) ele lvitelsise . $ ROOLTAU g'El CsI ov Vol rater OLT id cae nm mame SS CLA ATT een L0G Nest OSI IGz, red | Z 9G Cec Oecd uono Or tit Oscied . . o000 P ByLLeUL BAYJAYV Ae yy GLI GCI OFT OIG 6ST CAG Seana pt A oS eas "* gSodraqna CET 061 Vel ZOOL OFZ VLE [IG sy Ae ROO pet nd RIOD TOLD Kcr zsodraqna COE TCh GOSS! VOLE VOSS LOSce OF LOL COA Tco ee vor qsedriqna 6 erent ZO Al COT LZT ZOL . ajlevrelelie aiie}ia)ia/ie) isle) (sieKelielia P eioydour Ao L'6 FL FIL ZI GLI PEL BU Cote eet een P suoaplaoqqis Qh OI 9'8 ZOr LPI TT (ij je eC O BO OR On Ao on j vnponbaonb O'ZI 691 Let O'ST Liz LOL L'0@ CODCOD o e] ye) pire! /ee/ialisleMeme is! si (elie)\s P BNP Z'8 wal reare) SOT SFT 901 To ee EGov oeIdGGan abd 51 oi (0) $.1OdSIp seuy CAI a AT PHT Cll aT TT PIL ayy AT ROT oe — jInbhiq —~ ‘sopeul g jo sopduies JO UOlZRIAVp plepuR|s pue UPo Pp “UU NUATUL A aes ‘MO[aq Poyworpul sev ydvoxoa poansevour sofoods yora Jo usutfoads ouQ—) ‘sopeul GT Jo [dures “UMMULIXB TAL SG JO UOTZVIAOp plepurys pue ues ‘T ‘phyjip pue spup Jo sorods snoltea pue ‘woontiydosva nssauapoyy ‘vovsodad numdojayy ‘nuns 01}9 47 JO YOoF JO SPUOULIJo [RJo[IYS FO (SAo}OUITT[[UL) UL S}USUIANSBI I] }U GARIN] [, 11 606 L986 60+ TSE 0'SP LLé S'6E Lvs GEE tem ct of OSE SPP __ 0'6F GI + 69P OPE G9 OTS GT GIEé The Pink-Headed Duck Oy = 9°GG 6°29 GSP 8g L°0¢ 88 LO LCS GOP PSs ¢ a a lo lo) = a Dp © O26 19) 199 Sk Sk Sod eo) 19 > eb 8'6P 64 G09 D g G 9°6¥ 6'ES 1+ eg 06 i | v Les 8°60 ye G9§ 86 SP TL& OTP Date sOs 6°86 SOE 9GG 8°96 OSG 9'V1 rail OST OD == OFZ ram GSI SZ Lab ZOL OTL Pri L6 OT G0 + $01 GL g'8 69 GOT V9 4 vooepAydoArno BSSOUOPOY Y P vovsodod vuridoy 6 vuyni BIIIN eSTUye ZSTUyye FSTULyR sss D BLLOUISITBA 6 BULLOF 7!) (OP UboLtoule P Sliv[[Oo & RooIAU P BynapNy P Bleu eAYAW eSodtiqna csodriqna psodriqna sss OP vtazdour so “+ PsuOdpLIOgqIs *+ § epnponbaonb — Bynoe P SLOOSIP seuy SNSLD] -D]IUWOSMD T, AT ned TIT OU TT ROUT T nei OTL L'6 oral erie SI OSL COL 96 SSI SIT OL+FEcl LOFT o'ST a SI LG area | SSL 6'SL CCL LZl WIT LoL Pl VIL GET OST 66 0'6 GIT ToL FOFOOL FOF 96 G8 GL 16 GL OL 6S SIL 0'6 Lg ro PAI GAT Sane Al n6ugq—— May 10, 1962 ‘MOTO poeyBorpul se ydooxe poansvout satoeds Yova Jo uouttoads aug ‘why ip pue spup jo ‘paonyiydo.na vssourpoyy ‘vavsodad wumdojapy “nuns v}}9N JO Jos JO SpUoulsfo [RJ IYS FO (S.toyoULT] [Ud ‘SoyRUl g Jo Sofduies JO UOTPRIAOp plepueys pue UR, “fp UNUIUL, *"& (ponuryuo)) [ ATavy, “UINULIXB IL *% ‘sopeul QT Jo soptduies Jo uolpeIAop plvpuRys pur Ue 'T sgjoods snoliea pue ) UL SPUSMIOINS BIT WH Postilla Yale Peabody Museum No. 61 mens, digit IV was about the same length as or slightly longer (up to 1.4 per cent longer) than digit III. Rhodonessa is Anas-like in this character; digit IV is 0.9 to 2.1 per cent (three specimens) shorter than digit III. We have compared the relative lengths of the phalanges of the feet of Rhodonessa, Netta, Metopiana, and the various species of Anas and Aythya using the proximal phalanx of digit III as the basis for the intramembral proportions. The relative lengths of the phalanges of Rhodonessa as compared with those of the pochards and the dabbling ducks lead us to beheve that the dabbling-duck-like foot of Rhodonessa has evolved from a typical pochard foot. Unfortunately, lack of material makes it impossible to analyze the variability of the relative lengths of the phalanges of Rhodonessa. We suspect, however, that the variability is of the same order of magnitude as found in a sample of eight Aythya affinis. If this is true, we see no other explanation for the peculiar phalangeal propor- tions of the foot of Rhodonessa than that they are the result of modification of an ancestral pochard-like foot. The relative length of the proximal phalanx of digit II of Rhodonessa is greater than the maximum found for Anas and well within the minimum range for Aythya. Allowing for variability, one could safely say that this element is on the large side for Anas or on the small side for Aythya. The distal phalanx of digit II of Rhodonessa is relatively rather short for either Anas or Aythya. In digit III of Rhodonessa the second phalanx is relatively shorter than in any of the eight pochards studied and is slightly below the average of the six dabbling ducks. The distal phalanx is relatively smaller than the smallest of Anas and much smaller than in Aythya. The hallux (digit I) of Rhodonessa is relatively longer than in Anas and among the shorter of Aythya. (See Table 4.) In summary, the foot of Rhodonessa is Anas-like in 1) the relative shortness of digits II, II] and IV, 2) the absence of a lobe on the hallux and 3) the relative lengths of digits III and IV; it is more pochard-like (but perhaps intermediate) in 1) length of tarsometatarsus and 2) length of hallux. The intramembral proportions of the foot of Rhodonessa suggest that the phalanges have undergone in evolution a differential 13 The Pink-Headed Duck 4 May 10, 196 60 + Lor Is Nn == Ming mar =u Ley le) u nal == = = NC = XD = ae OST GOS SOP § Log OSS GLY o6P f6P g19 9GL sé+ FOL gL OOL £89 OGL GGL GILL GTh Ts g’9g eT == 07eS T6¢ 0°8Sg¢ 860 £6P GOV O'8S 8°9¢ 0°69 619 989 GGG 9°9¢ Sole Strong C'6¢ 6°8¢ Sg OPS Gre SOV LV VSP GSE OSS 9S + 6'0G 0°Sg “ig 0'0¢ Vé¢ G'OG OTS OTE Vv 6S OST eS Ost: OVP OEP O'6E 865 6S ZF sie) e\lee eit ee eee + GT ye) (a) lec Z Gl Sito.d 4 GOT ct: PIT occ" VOL cc: 4 vooeypAydoAreo e Bssouopoy yy voevsodod vuerdoyo 6 vuyn VION * eSTuyypR * gsruyye * pSTuyype P BILLOUISTTRA "+ &§ Buldoy P RUROLLOULe P Sliey][Oo 6 RoOLAU P BynsIpNy P B[LLeuL PAYAY eSodriqna zSodniqna Sod riq nt P vtoydour Ao * PsuoapLaqqs - & vnponb.onb P BynoR P SLOoStp seuy SNS] -DJAMOSA VE AT HOUT TIT hed TT Reid Ted Soryet JO UOTPRIAVP plepueysS pue UBT UIMLULL UL Y “UINULIX® JL ‘sopeul QT JO opdures Woda, Sole JO UOTZRIAOp plepuRys pur Uvo] ‘snaouny Jo Yysus, Jo Juss tod sev possoadxo snsavpeyouLOsany puR S}LTIp JO syysuo'T Z TIAV], No. 61 G8 9°¢ 8°9 89 6'T ceits 9°9 oT 9 ry quaa sad ov O'S 9E GT Ea Gs 9's GG S's LG CnC nC CCC aC aC esodiiqna Soohonvcoo000H zsodraqnt aie)ie\ a) bile sire) lejfel e)cei 0 cSodtiqna seuy DGREGGC DODD OO Ay reVyilaxal vuridoyo fy OSS EC TCU Oso OA a HT (ER BVIIN PAGS DOO ODES FO ETT e iad ah eT OUTS TEA Qon0GodoBenaeoob sHeKEEyI DEEL EES CERO ON 0.0 U2} sHsanas teaches eee ST ABT] OD sree eee reese tees paoThu BRO ODES OCR Sia) Bi eye 2AUAY 60 9's G0 61 GO LT O€ ol 6G quad sad 2a0R[[AYdo ALK BSsouOpoY yy eSodtaqna zsodtaiqna psodraiqna vaioydouer dd SUOLJLLOGqia vpnponb.azonb ou eGo “Ny Maye) "+ §LOOSIp seuy HHL ool © (ITI-AD) Ky pbhig uy) sab DTT SA uo) TIT 21d — ITT 001 ® (AI-ID) — 47 big uy) sabuo) [TT Vtbiq— AU IUIE Postilla Yale Peabody Museum FH rl “‘SoTRUL & “SOTRUL g “SOpTPul fos jo opdures jo uvoyy yo ofduies Jo ues “UINULIX® AL UUM UT AY jo ofdures jo uvoyy ‘S[PNPIAIPUL & UL AT JIZIP uUvYy} JoJL0ys pur s[peNpLArpul uRYy} toduop sea TT] PSIp (sopeur) suouoods QT Jo ssadiiqne spupr *yuoo aod 7g 0} YUod aod gE AG JoBuoT SALATR SVM AT ySip (sateur) suosuttoads g Jo @ aAlTavye§ Alu L UL AT PSIp mun viywhip TTT wsrp Jo yysuep vyy Jo yUued aod se possotdxs o1B SootatoyIp gSoy} SUUIN]OD quaovl'pe oy} UL post] oA AT PUP JIT SPSIp JO syyFuoy oy UL soousIopIp oT, “AT pure TI] S}4Ip Jo syysue, Jo uostreduio0y 1D The Pink-Headed Duck 62 May 10, 19 OLE OCF e'EG L'08 6'GG 0°89 6LL 966 © § voorypAydodreo BSSIUOPOY YY Fog Z'OP 6'LG 18 809 O'eL O'eL o'g0L “°°°' °° * ~Pvaesoded vuridoyo[w EPS 0'6P e118 9 Veh PSL ZOE. eee 6 BUND Ryo ZG O'SP 0'6¢ G18 G'8g 0°69 CPL GOO. eed ae ese s¢ G1¢ ro PS 0°E9 SL SL GeO eee eeece LUE cg SIFEL6r LOFG0O9 TI+res 9L+ 079 CaS PlsacoOL Cl -ee0hy says ae ase EG ZEG L'6¢ 9&8 C09 ey OLL GG ot ae P BI LOUISITRA 0G SSP LG 0°08 0°09 argh PPL COM no ae cee CO UULIO ag gg 0°29 9°18 8°29 PCL SOL ep Vow P BUROTLOUIR PES WAS s'19 ("G8 0'e9 O'EL rag pl) 600k 2 e siauioe OV Log €'6S 0'E8 9°29 Gai, 6PL Gg sna nO BOOMAUL 6'GG Vig 66S ©'E8 reo GZL GOL OCOls wes cee BUST Ty G'ZS 9'0G 119 G08 19 VSL TOL OO oo ee ay ee, BAYJAYV pp 9°63 8'ZG OSL PS G99 6'OL Gogm oo ee ees ouiacnd L'6P Lop e' 9G PIs 0°09 9°69 9'9L SOL °°°° °°? gsediaqna CL=VOF LLtoer Slop Gi+s6s RIG GOFG LO) Oca re, Sls 86 5 6. | sodtiqna Gg GOP Pog 18 619 GPL OLL e796 °°’ «6 B1aydoueAd 80 GLP [DS G08 Leg 9°19 ORL ZL6 TP suoaztraqqys OLY TOP seeks) O18 8'6¢ r69 GOL e146 <c°7 7° & Bfnponbasonb 2g CTP rss OSL roo Zz'69 oP GGG ne ae ae Oe EOP Ze peg S18 G'6G OSL OIL Ue eo Gore EOP EL(D seuy PAI SAI GA! dAl PITL 211 PII ayy AL Ord — pr nbrq — — jp nhiqg — ‘TIT }51p ‘Sopeul § jo soyduies Wodfy SOeL Jo uOT]RIAIp pte purys pue uvo[y “P UINULIUL TAL *& “UNULIXR ILS ‘SoyeuL QT Jo opdures morf SOLA Jo UOTPBIAIp piepuerys pue ueoW ‘1 p AIAV jo xuvpeyd peurxoad oy} JO YSusyT oy} JO YUoo aod sv possoadxo sosurpeyd Jo sqysuo'] 16 Postilla Yale Peabody Museum No. 61 reduction, the more proximal phalanges being least affected and the distal phalanges most affected. This is most apparent in digit III but is suggested in digit II. Digit IVY must have undergone the greatest reduction of all from the presumed ancestral pochard condition of having been longer than digit III. Miller (1937 :45) found in his studies of the feet of geese that ‘‘increase or decrease in toe length takes place to a greater degree in digits two and four than in three, and in- crease or decrease in phalangeal length takes place to a greater degree in distal (exclusive of ungual) than in proximal pha- langes.”’ ges, BEHAVIOR AND ECOLOGY There has been little recorded about Rhodonessa beyond the fact that it has always been observed uncommonly. The Pink- headed Duck appears to have been a solitary species, occurring as pairs in the breeding season or small groups in winter (Hume and Marshall, 1881, 3:176) on isolated marshy ponds or swampy lakes and not joining the large wintering concentra- tions of migrant waterfowl. Delacour (1956:198) notes that its behavior was much like that of dabbling ducks. The species nested in April in long grass (Andropogon) or grass tufts sometimes away from the water. The eggs were white and uniquely spherical (Finn, 1909:25) and the nests well-formed with no special lining. The male had a whizzy whistle like a Mallard but lower and weaker. The female had a low quack. Finn (1909 :86), however, speaks of the male’s call as quite unlike that of any other duck, resembling the syllables “wugh-ah.” Males when together (in captivity) were noted to display like Mallards but more simply. The head was drawn in between the shoulders while the head feathers were puffed out. Following this the neck was stretched upwards and the call uttered. How- ever, as Delacour (1956:198) has observed and described the display, it was so simple that it lacks any real relationship to the display of a dabbling duck. It resembled equally well the simplest forms of aythyine display such as that of Metopiana for example, lacking only the angular position of the bill pointed upwards at the climax of the neck-stretch. These May 10, 1962 The Pink-Headed Duck LK? observations, in captivity, may have been of birds insufficiently stimulated to have produced a diagnostic display. Flight of Rhodonessa is said to have been light and easy and the habits like those of a true surface-feeder (Finn, 1909: 86). Finn (1915:25) observed an individual dive ‘tas neatly and long as a pochard.” Perching was never observed. The Pink-headed Duck was a solitary, non-migratory species which was local in distribution: the center of its restricted distribution was the Gangetic area of the “terai” of northern Bihar. The last record of its occurrence in nature was in 19385 (C. M. Inglis). Jerdon (in Hume and Marshall, 1881, 3:176) reports that the birds remained out in the center of the pond during the day and so presumably fed at night. Shillingford (in Hume and Marshall) reports that a gizzard contained water weeds and small shells. The duck was said not to have been a palatable species, indicating an animal diet. It is unfortunate that so little is known of the habits and behavior of the Pink-headed Duck. The lack of information 1s particularly aggravating considering that this now extinct or nearly extinct species was at one time not uncommon in col- lections of European aviculturists where it displayed but never nested. All that can now be said is that the anatomy and pro- portions of the foot of Rhodonessa indicate feeding and loco- motor adaptations very much like those of dabbling ducks. The sparse behavioral information available provides some if not overwhelming support for this notion. DISCUSSION What sort of history of ecological changes can have pro- duced the Pink-headed Duck, a pochard with strikingly ana- tine modifications of the foot? The most likely sorts of environ- mental changes would have involved a shift from the more typical pochard habitat, that is, a marine or inland sea littoral environment, to the present one of marshy, fresh water ponds. It appears that just such a change took place in India during Tertiary times. The Neogene period, characterized by a general regression in the middle Miocene, brought an isolation of the eastern 18 Postilla Yale Peabody Museum No. 61 Mediterranean into an inland sea. This eastern Mediterranean province is described by Gignoux (1955 :554-5, 587-8, fig. 144). Characteristically beginning with the upper Vindobonian or upper Sarmatian, the inland sea thus formed began to show isolated genera of invertebrates responsive to decreased salinity with the outpourings of large fresh water rivers. The eastern- most of these lacustrine basins is the Aral-Caspian which stretched north to Samara and east far beyond the Aral Sea. This eastern basin escaped the reinvasion of the Mediterranean which occurred in the Quaternary with the opening of the Dardanelles and the extinguishing of the Levantine and western Caspian faunas. The southern parts of this area of trapped inland sea in the southern Caspian and south of the newly risen Caucasus re- present part of the old Mesogean or Tethys Sea basin, con- tinuous through to the Indian Ocean and cut at the beginning of the Miocene. Similarly in India, the Miocene system (Wadia, 1944 :256- 275) shows a series of clays and sandstones whose characters suggest deposition in an estuary or the broad mouth of a river (Gaj Series in Sind). This shows a regression of the sea border and its replacement by the wide basin of an estuary. A good example of this sequence is the Potwar trough which shows de- posits of nummulitic form over Mesozoic rocks, overlain by brackish-water sediments of Aquitanian and Burdigalian age (lower Oligocene and upper Miocene) followed by the fluvia- tile and sub-aerial Siwalik strata. How far southward the Siwalik system lies is unclear; much of it is probably buried under the recent alluvium of the Ganges. Siwalik birds are interesting in that Phalacrocoraa, Pelecanus, and Mergus are represented, genera found in India today, and characteristic of very large lakes or large river systems. The Pleistocene system in India reveals evidences of glacia- tion in the form of moraines and polished and grooved rock at low altitudes in the Himalayas, and also the desiccation of the Tibetan lakes consequent on the disappearance com- paratively recently of the glaciers of the ice age. The plains of India north of the peninsula reveal another aspect of the Pleistocene in the Indo-Gangetic depression filled May 10, 1962 The Pink-Headed Duck 19 with thousands of feet of alluvium (Wadia, 1944, fig. 33). The present river systems bear little relation to their past history even during historic times. This is a vast area of wandering rivers and repeated alterations of clays, sand and marls with recurring layers of peat, lignite and some forest beds. Huge deltaic areas as in Bengal cover parts cf the former Bay of Bengal, or in the west, as in the Rann of Kutch, remains of bays of the Arabian Sea. All of this intervening plain except for occasional projections of older rocks must represent the remains of an inland sea, perhaps formerly continuous with the Aral-Caspian basin of Miocene times. The whole recent history of this area would appear to be one of shift from marine to brackish to lacustrine to riverine conditions. Following this has been the sustained recent desiccation of central India in historic times (Salim Ali, 1927 :833-861). Seventeenth century records show the range of the great Indian rhinoceros for example as occurring up te Peshawar and the foothills of the northwest in what is now a semi-desert area. In view of the above it seems very likely that certain animals adapted for life in or about large bodies of water if capable of the necessary evolutionary modifications, would have been able to adjust to this radical change in environment and continue to persist. Of the four endemic species of Indian birds discussed by Ripley (1961: xxiii) two are endemic Indian genera whose normal habitat is aquatic or semi-aquatic. One is the bristled grass warbler, Chaetornis, a bird of the Gangetic drainage system, local in habitat, found in moist grassy places. The other is the Pink-headed Duck, which has been found only in lakes or ponds in the are from Bihar to eastern Assam, al- though there are scattered records of nearly a hundred years ago indicating a wider distribution throughout the Indian plains and spottily into the Peninsula. If this species indeed shows affinities with the estuarine diving ducks as we believe the anatomical evidence suggests, then it would appear to be descended from a form which was plastic enough to respond adaptively to long-term changes in the ecology of southern Asia. It evolved gradually into a fresh water surface feeding type of duck as the total water area diminished. Again, with the 20 Postilla Yale Peabody Museum No. 61 disintegration of the Aral-Caspian, it became isolated in the furthest reaches of the increasingly dessicated Gangetic system. The Pink-headed Duck ultimately became trapped, so to speak, in the evironment to which it had become adapted and as a sedentary relict species appears at present to have vanished almost, if not completely, from the scene. The evolution of the Pink-headed Duck appears to have been related to the major environmental changes which took place in India during Tertiary times. The principal occurrence in the history of this peculiar genus was the development of dabbling- duck-like locomotor adaptations, at least so we infer from the curious modifications of the foot. The trachea, plumage pat- tern and certain characteristics of the foot leave no doubt about the aythyine relationships of Rhodonessa. LIreRATURE CITED Ali, Salim, 1927. The Moghul Emperors of India as Naturalists and Sports- men. Jour. Bombay Nat. Hist. Soc., 31:833-861. Delacour, Jean, 1956. The Waterfowl of the World, Vol. 2. Country Life Limited. London. Delacour, Jean, and Ernst Mayr, 1945. The Family Anatidae. Wilson Bull., 5733-55. Finn, Frank, 1909. The Water Fowl of India and Asia. Thacker, Spink and Co., Caleutta. ix + 121 p. Finn, Frank, 1915. Indian Sporting Birds. F. Edwards, London. xi + 280 p. Garrod, Alfred Henry 1875. On the Form of the Lower Larynx in Certain Species of Ducks. Proc. Zool. Soc. London, 1875:151-156. Gignoux, Maurice, 1955. Stratigraphic Geology. W. H. Freeman, San Francisco. xvi + 682 p. Hume, A. O., and C. H. T. Marshall, 1881. The Game Birds of India, Burmah, and Ceylon, Vol. 3. Calcutta. Humphrey, Philip S., 1955. The Relationships of the Sea Ducks (‘Tribe Mergini). Ph.D. Dissertation, Univ. Mich. Johnsgard, Paul A., 1961. The Taxonomy of the Anatidae—A Behavioural Analysis. Ibis, 103a (1961) :71-85. Miller, Alden H., 1937. Structural Modifications in the Hawaiian Goose (WVesochen sandvicensis), a Study in Adaptive Evolution. Univ. Calif. Publ. Zool., 42:1-80. Peters, James Lee, 1931. Check-List of Birds of the World, Vol. 1. Harvard Univ. Press, Cambridge. Phillips, John C., 1922. A Natural History of the Ducks. Vol. 1. Houghton Mifflin. Boston. May 10, 1962 The Pink-Headed Duck 21 Ripley, S. Dillon, 1950. Vanishing and Extinct Bird Species of India. Jour. Bombay Nat. Hist. Soc., 50:902-906. Ripley, S. Dillon, 1961. A Synopsis of the Birds of India and Pakistan. Bombay Natural History Society. Salvadori, Tommaso, 1895. Catalogue of the Chenomorphae (Palamedeae, Phoenicopteri, Anseres), Crypturi, and Ratitae in the collection of the British Museum. Catalogue of the birds in the British Museum, Vol. 27. London. Verheyen, René, 1955. La Systématique des Ansériformes Basée sur l’Ostéologie Comparée (suite). Bull. Inst. Roy. Sci. Nat. Belgique, Vol. ol, no. 37:1-22. Wadia, D. N., 1944. Geology of India. London. Woolfenden, Glen E., 1959. A Pleistocene Avifauna from Rock Spring, Florida. Wilson Bull., 71:183-187. Woolfenden, Glen E., 1961. Postcranial Osteology of the Waterfowl. Bull. Fla. State Mus., Vol. 6, no. 1:1-129. fF (f ‘i is iy ‘ ‘f Aa. ~ any iN : ; 7 ‘tA iz ant \ + iy % =) i 4 ; 2 ' a: Ver 5 ] t a ' ‘4 A . L -_ i af i ' i ‘ i J \ iy es hy 4 f& ee ae} Date a4 h is + ig id = oan 4 = on ann — —- ast wT Date Due N74 MAR ——1976 AUS HB ee aL” es aa ar Nn es ms Se