Printer Friendly

Skeletal characters and the systematics of Estrildid finches (Aves: Estrildidae).

ABSTRACT. The family Estrildidae contains 108 to 140 species arranged in 15 to 40 genera and two to nine tribes or subfamilies. In this study 401 skeletons belonging to 103 species and 27 genera were compared, as well as 75 skeletons of 15 genera of possibly related forms; skeletal characters add to the existing characters of behavioral, external anatomical, and molecular features. Best skeletal characters for generic separation were shape of the caudal basibranchial and a combination (PCA) of mensural characters. Pholidornis is not an estrildid. The classification of Goodwin (1982) was affirmed with these changes: Tribes should be dropped. The genus Lepidopygia should be lumped in Lonchura. Additional taxonomic changes are cautiously suggested.

Keywords: Wax-billed finches, Africa, Australia, southern Asia, Indo-Pacific islands

**********

The finches studied here have variously been called waxbills, grass finches, manikins, munias, etc. They inhabit Africa, Australia, southern Asia, and the Indo-Pacific islands, and have been widely kept and studied in aviaries worldwide. From 108-140 species of Estrildidae, arranged in 15-40 genera, are recognized by recent authorities. Three tribes (or subfamilies) are usually recognized, although several genera have been moved back and forth amongst those tribes. External anatomy (including patterns of mouth spotting in nestlings), behavioral patterns, presence or absence of nessoptiles, appendicular myology, chromosomal configurations, and chemical analysis of proteins and DNA (both nuclear and ribosomal) have been used as characters for classification of species, genera, and tribes. Biochemical studies were those of Kakizawa & Watada (1985), Christidis (1987), Sorenson & Payne (2001, 2002), Payne & Sorenson (2003), Sorenson et al. (2004), and Van der Meij et al. (2005). The addition of a set of skeletal characters should help clarify relationships.

Complete classifications which may be called modern were these: Delacour (1943) recognized 15 genera and 108 species, grouping them into three tribes, Estrildae, Poephilae, and Lonchurae. (Hereafter, these tribal names are more correctly spelled Estrildini, Poephilini, and Lonchurini.) Wolters (1957) recognized 31 genera. Steiner (1960) recognized 9 tribes and 33 genera. Mayr (1968) recognized 25 genera, but did not list every species; he utilized Delacour's three tribes, but moved four genera to different tribes. Mayr, Paynter & Traylor (1968) recognized three tribes and 27 genera. Goodwin (1982) recognized 27 genera and 132 species--not quite the same genera as in Mayr et al., for he split one of their genera (Lonchura) and lumped another (Padda); he did not explicitly recognize tribes. Sibley & Monroe (1990) recognized 29 genera, although they mentioned as generic characters only the DNA work of Sibley & Ahlquist (1990) on three genera and four species and the protein electrophoresis work of Christidis on 16 genera. In a series of papers, Sorenson & Payne (2001, 2002), Payne & Sorenson (2003), and Sorenson et al. (2004), using mitrochondrial DNA, studied 74 species and 26 genera (also 12 more genera they recognized, but which Goodwin 1982 did not). Sorenson et al. (2004) grouped the family into two major clades, a primarily African one (= Estrildini) and a primarily Australasian one (= Poephelini and Lonchurini); the former they divided into three clades and the latter into two or three in their next two steps.

Systematic studies at the generic level but omitting several genera included: Wolters (1950, 1966) reviewed the African genera. Gtittinger (1970) revised Lonchura using behavioral characters. Ziswiler, Guttinger & Bregulla (1972) revised Erythrura using behavioral and external anatomic characters. Guttinger (1976) reviewed two genera using behavioral characters. Bentz (1979) studied 23 genera using appendicular muscle characters. (Bentz raised the taxonomic level of the three tribes to subfamilies.) Kakizawa & Watada (1985) studied protein electrophoresis of 20 genera; they reduced the number of tribes from three to two, lumping Poephilini with Lonchurini. Christidis (1986, 1987) studied the chromosomes and proteins of 16 genera; he recognized the three tribes but moved Aidemosyne to a different one. Baptista et al. (1999) studied seven genera using protein eletrophoresis, palate markings of nestlings, neossoptile distribution, bill shape, and six behavioral characters; they split Lonchura into two genera, lumped Padda into one of these, and moved Amadina from Lonchurini to Estrildini. Schodde & Mason (1999) reclassified the 20 Australian species into eight genera; they lumped Aegintha, Aidemosyne, Chloebia, and Padda into various genera, but split Emblema into two genera (Emblema and Stagonopleura) and Poephila into two genera (Poephila and Taeniopygia). Also, they split Heteromunia off from Lonchura. Van der Meij et al. (2005) studied 12 genera using mitochondrial and nuclear DNA.

Many workers have investigated the relationships of the Estrildidae to other major suprageneric groups. Sibley & Ahlquist (1990) gave a thorough summary. Sushkin (1927) concluded that the waxbills were most closely related to the whydahs (Vidua), but belonged as a subfamily in the Ploceidae. Tordoff (1954) on the basis of skull characters placed the Estrildinae in the Ploceidae, near the Carduelinae. Friedmann (1960, 1962) concurred with Sushkin. Ziswiler (1967) studied the structure of the alimentary canal of various seed-eating birds (including two species of whydahs and nine species of eight genera of estrildids). He found estrildids uniform among themselves and distant from Ploceidae (including Viduinae) and Fringillidae. Foelix (1970) studied the salivary glands of seed-eating birds (not including any whydahs); he found the estrildids uniform among themselves and distinct from Ploceidae and Fringillidae. Ziswiler (1979) from a study of the stiffening devices of the tongue confirmed his earlier position that the Emberizidae, Ploceidae, and Estrildidae had evolved the seed-eating adaptation separately. Christidis (1986, 1987) on the basis of chromosome and protein electrophoresis work placed the estrildids closer to the Ploceidae than to any other group. Sibley & Ahlquist (1990) and Sibley & Monroe (1990) on the basis of their DNA studies placed the estrildids and whydahs (two species studied) in one subfamily of Passeridae and the sparrows and weavers in other subfamilies. Cramp & Perrins (1994) characterized the Estrildidae, Passeridae, and Ploceidae, placing the Bubalornithinae in Ploceidae and Viduinae in Estrildidae. Schodde & Mason (1999) included several skeletal characters in their characterization of Estrildidae. Sorenson & Payne (2001, 2004) on the basis of mitochrondrial DNA studies, concluded that the estrildids and viduins had evolved together, separate from the ploceids and passerids. Lahti & Payne (2003), followed by Fry & Keith (2004) moved Anomalospiza from Ploceidae to Estrildidae, Viduinae, on the basis of anatomy and behavior, concluding that it was a fairly close relative of Vidua.

From all of the literature review above, one poorly-known genus, Pholidornis, which has sometimes been placed in Estrildidae, has been excluded (Traylor 1968, Vernon & Dean 1975, Sefc et al. 2003).

METHODS

Skeletons of 401 individuals belonging to 103 species and 27 genera of Estrildidae were examined. In addition, 23 skeletons of eight species of Viduinae and 52 skeletons of 22 species of 13 genera of Ploceidae (two genera of Bubalornithinae, five of Passerinae, and six of Ploceinae) were compared. The classification of Goodwin (1982) is used in this paper for Estrildidae except where otherwise stated. Outside Estrildidae, the classification of Moreau & Greenway (1962) in Check-list of Birds of the Worm is used, except for the genera Pholidornis and Anomalospiza.

Fifty-one characters (measurements, ratios between measurements, and shapes of bony parts) were tabulated for each specimen. Of course, as a result of broken or missing bones or my inadvertence, a few data on some specimens were not recorded. On arithmetical characters the figure used for a species is a mean of the specimens measured. Of these, 20 characters were arithmetical and complete enough to be used in a principal components analysis (PCA) ordination for the 127 species (Table 2 and Figs. 1-14). All data were converted to ranks to avoid problems associated with distributions of ratios and lack of normality and to give all variables equal weight. Analyses were performed using PC-ORD (MjM Software 2002).

The specimens and species examined are listed in Appendix 1. The 35 species and one genus not studied are listed in Appendix 2. One taxon (Estrilda melanotis quartinia) which was listed as a subspecies by Goodwin (1982) was analyzed separately because it was recognized as a full species by Sibley & Monroe (1990) and Fry & Keith (2004). Eight more taxa split as full species by Sibley & Monroe but lumped as subspecies by Goodwin (1982) were not examined, nor were five more species added by Payne & Sorenson (2004) and Fry & Keith (2004).

Ortygospiza (= Paludipasser) locustella could not be studied because no skeleton exists in the world's museums (Wood & Schnell 1986). The only one listed by Wood & Schnell could not be found in the Brodkorb collection.

GENERA OF ESTRILDIDAE

The following 15 characters provided the best separation of genera. See Table 1 for a convenient comparison. Padda is included for convenience, but it was reduced to a subgenus by Goodwin (1982). They are arranged below in approximately their order of usefulness--that is, the order in which they separate the most genera.

(1) Shape of caudal basibranchial: The basilbranchial (= urohyal) was flattened dorsoventrally (depressed) in some species, but slender, not depressed, in others.

(2) Shape of maxillary process of nasal: It was only broad and heavy or some broad and heavy and some moderately broad, in Pyrenestes, Spermophaga, Erythrura trichroa (one of the seven species in the genus), and two of the 29 species of Lonchura.

(3) Shape of transpalatine process: The transpalatine process, at the latero-caudal corner of the palatine bone, was measured as the ratio of its length over its width. It was elongated, 3.3 or more, in Parmoptila, two species of Nigrita, two species of Estrilda, Emblema oculaturn, and Erythrura prasina. It was blunt, under 1.6, in eight genera and parts of eight more.

(4) Palatine process of premaxilla: The palatine process of the premaxilla was present as a lateral flange (except in one specimen of Parmoptila), but with no suture evident between it and the palatine bone (Tordoff 1954; Bock 1960). Often an additional extension, or process, was present, extending caudally from the lateral edge of the process; sometimes this additional process was as long as 0.3 of the width of the entire lateral flange plus palatine at that point. The lateral flange was small or slight in Parmoptila (absent in one specimen), two species of Nigrita, Nesocharis shelleyi, and in one specimen of Clytospiza; in all other species it was broad. The additional process was lacking in three genera and some specimens or species of 19 more genera.

(5) Pneumo-tricipital fossa: The pneumotricipital fossa of the humerus was always double, and usually combined. The various shapes were categorized as they were for Icteridae (Webster 2003): A--Combined dorsal and ventral fossae equally deep, bottom translucent, separation by medial bar incomplete. F--Dorsal and ventral fossae equally deep, but a low, rounded ridge extends distally from the internal tuberosity clear across, separating the fossae. DT--Dorsal and ventral fossae equally deep and translucent-bottomed, but a prominent partition or medial bar extends distally from the internal tuberosity, separating the fossae. E--Dorsal fossa opaque-bottomed, 70-90% as deep as the translucent-bottomed ventral fossa; separation between the fossae a step-down. DO--As in E, but dorsal fossa only 40-60% as deep as the ventral fossa. Only two specimens (of Clytospiza and Emblema guttatum) were DT.

(6) Ventral surface of parasphenoidal rostrum: The base of the parasphenoidal rostrum bore a narrow ventral keel (Sushkin 1927) in some specimens.

(7) Tibiotarsus length.

(8) Skull length: Measurement was from the caudal end of the external nare. Maximum was 21.2 mm in Spermophaga ruficapilla; minimum was 13.0 in Amandava subflava.

(9) Interorbital septum.

(10) Ectethmoid foramina: The ectethmoid (= orbitonasal) foramina were single, pinched, double, or in a few cases triple. However, these divisions were rather erratic within genera or even within species. The relative size of these foramina was fairly consistent within species and genera; it was recorded as size of the lateral foramen over the medial foramen. This was not clear in individuals with a single foramen; if triple foramina were present, the two lateral ones were added together as the numerator.

(11) Interpalatine process: The interpalatine process varied from absent to well-developed. It was absent or only a slight bump in five genera and in three species of Estrilda, Poephila bichenovii, and eight species of Lonchura. All other species showed a well-developed process; the maximum of 2.5 was in Lagonosticta nitidula and Emblema oculatum.

(12) Tarsometatarsus shape: The torsometatarsus varied in stoutness.

(13) Compression of rostral basibranchial: The rostral basibranchial (basihyal) was invariably compressed, but the degree of compression (ratio of height to width at the midpoint) varied. This character was omitted from the PCA because so many specimens (all specimens of seven species) lacked the bone.

(14) Fusion of mediopalatines: The dorsal walls of the choanal fossa of the paired palatine bones (= mediopalatines) were fused or sutured together in the midline for part (usually) or all of their length, except in Pyrenestes and Hypargos and a few scattered specimens (no entire species) in other genera. The condition was recorded as the percentage of the length fused (or sutured) of the length where the paired palatines were close to the midline.

(15) Shape of premaxilla: The premaxillae varied in their taper, measured as length over width at the caudal end of the external nares. (Of course with the rhamphotheca removed.)

Principal component analysis.--A summary is in Table 2, showing factor loadings. Seven of the best intergeneric chracters are omitted because they are not quantitative (six) or lack complete data (one). On the PCA figures, axis 1 is heavily weighted to long bone ratios between wing and leg--positive correlation to humerus/femur and ulna/tarsometatarsus and negative correlation to tibiotarsus/ humerus and tibiotarsus/ulna. Thus species positive on this axis have relatively long wings and short legs; negative values are the opposite. Axis 2 is heavily weighted positively to more slender humerus and broader cranium and negatively to larger lateral ectethmoid foramen and longer postorbital process. Axis 3 is heavily weighted to skull length and tibiotarsus length (measurements of total size); negative values are bigger species.

The following 12 additional characters showed consistency within species, but individually provided only a little in the way of intergeneric variation, or generic characters (all of them were used in the PCA analysis, along with eight of the nine arithmetical characters already discussed): (1) ratio of cranium width-interorbital width, (2) shape of the zygomatic process of the squamosal bone, (3) relative length of the postorbital process, (4) ratio of tibiotarsus length-ulna length, (5) ratio of tibiotarsus length-humerus length, (6) ratio of tibiotarsus length-femur length, (7) ratio of ulna length-femur length, (8) ratio of ulna length-tarsomatatarsus length, (9) ratio of humerus length-femur length, (10) ratio of tarsometarsus length-femur length, (11) ratio of tarsometatarsus length-skull length, (112) ratio of humerus length-width.

The PCA analysis (Table 2 and Figs. 1-10) shows these points of generic similarities and differences: Nesocharis is very close to Parmoptila on both Figs. 1 and 2, but those genera are far from the close cluster of Nigrita in both figures. Cryptospiza, Mandingoa, and Spermophaga are the only genera of Estrildini which are separate from all others on all Figs. 1-6. Estrilda, Uraeginthus, and Lagonosticta are separate from one another on Fig. 4, but not on Fig. 3. Spermophaga is close to Pyrenestes, especially P. minor, on both Figs. 3, 4. The twinspots, Clytospiza, Hypargos, and Euschistospiza, but not Mandingoa, are close to one another in Figs. 5, 6. Ortygospiza is not close to Amandava on either Fig. 5 or 6, and the three species of Amandava are not well clustered on either figure. Pytilia is rather widely split, with P. melba far from the other three species in both Figs. 1, 2. Estrilda is well clustered, despite the large number of species, in both Figs. 3, 4.

The genera Aegintha, Emblema, Neochmia, and Poephila (= Poephilini) are all separate from one another in Fig. 8. In Fig. 7, Aegintha is within Emblema and Emblema partly overlaps Poephila. Emblema is radically split in both figures, with E. guttatum and E. pictum close to one another and Poephila and E. oculatum and E. bellum near or around Aegintha. All five species of Poephila are grouped together in both figures. Aidemosyne is close to Neochmia in both figures.

The genus Lonchura is well clustered for such a large group in Fig. 9, and Aidemosyne is within it; Lonchura, Padda, Erythrura, Chloebia, and Amadina are separate from one another, although Chloebia is near Erythrura. In Fig. 10 of the same genera, Chloebia and Amadina are separate from the others and each other, but Aidemosyne, Padda and Erythrura overlap Lonchura; Amadina is outside all the other genera of Estrildidae and not particularly close to any of them on both Figs. 9, 10 and the complete figures (Figs. 11 and 12, where not labeled).

Unused characters.--Twenty-four more characters were tabulated for each specimen, but then abandoned. Of these, nine were too inconsistent within species to be useful: Shape of the lateral margin of the ectethmoid plate, size of the supraotic fenestra, relative lengths of the metatarsal trochleae, shape of the maxillopalatine bones, shape of the internal (= medial) process of the mandible, presence and nature of the Pocock foramina (Pocock 1966), the degree of the distal bend of the tarsometatarsus, degree of the medial curve of the rostral palatine, and presence and shape of a manubrium-sternum bridge.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Fifteen characters showed too little variation within the family--were too consistent--to be useful: Degree of ossification of the nasal septum, degree of ossification of the nasal conchs, presence or absence of a horizontal flange of the nasal septum (= floor of the nasal capsule), rostral palatines twisted or not, shape of orbital process of quadrate bone, shape of retroarticular process of mandible, completeness of rostral palate, shape of process 7b of mandible, shape of pseudotemporal process of mandible, presence or absence of a free lacrimal bone, degree of fusion of pterygoid-palatine junction, ratio of ulna length: humerus length, ratio of tibiotarsus length: tarsometatarsus length, degree of squamosal inflation, shape and size of the rostral end of the vomer.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

[FIGURE 10 OMITTED]

[FIGURE 11 OMITTED]

[FIGURE 12 OMITTED]

[FIGURE 13 OMITTED]

[FIGURE 14 OMITTED]

[FIGURE 15 OMITTED]

[FIGURE 16 OMITTED]

[FIGURE 17 OMITTED]

Boundary of Lonchura.--Because there have been so many taxonomists who have split the large genus Lonchura, but in a variety of ways, it was analyzed in detail for the 51 tabulated osteological characters (including Lepidopygia). For these the maximum and minimum one or two species within the genus were recorded; if three or more tied, it was recorded as no extreme. Altogether, 24 of the 29 examined species were extreme for one or more characters. There were no extremes for 15 characters; there was either no minimum or no maximum extreme for 12 more characters.

Results were: Extreme on nine characters--Lonchura cantans. Extreme on seven characters--L. vana, L. monticola. Extreme on six characters--L. melaena. Extreme on four characters--L. striata. Extreme on three characters--L, cucullata, L. nevermanni, L. teerinki, L. pectoralis. Extreme on two characters--L. griseicapilla, L. bicolor, L. fringilloides, L. grandis, L. hunsteini, L. leucosticta, L. fuscans, L. spectabilis, L. flaviprymna. Extreme on one character--Lepidopygia nana, L. leucogastroides, L. tristissima, L. malacca, L. caniceps.

The genus Padda has been lumped with Lonchura by several workers (Delacour 1943; Goodwin 1982; Baptista et al. 1999; Schodde & Mason 1999). When compared with the 29 species of Lonchura for the 51 osteological characters, there were these results: both species of Padda were beyond the variation in Lonchura on three characters. Padda fuscata was beyond the variation in Lonchura when Padda oryzivora was not in three more characters. Padda oryzivora was beyond the variation in Lonchura when Padda fuscata was not in three more characters. The three characters in which both species of Padda were outside Lonchura were ratio of humerus length to femur length, ratio of ulna length to femur length and caudal basibranchial shape. It is concluded that osteological characters indicate the same as external structural characters--Padda may either be retained as a weakly characterized genus or lumped as a subgenus within Lonchura.

In the same way, Euodice, a subgenus in the Goodwin (1982) classification, was compared with the variation in Lonchura. The two species examined, L. cantans and L. griseicapilla, were not both outside the variation in the rest of Lonchura on any character. Similarly, Spermestes, another subgenus in the Goodwin (1982) classification, was compared. The three species, Lonchura cucullata, L. bicolor, and L. fringilloides were outside the variation in the rest of Lonchura on only one character (shape of the zygomatic process). On two other characters, all but fringilloides were outside. As noted above, Lepidopygia nana was outside Lonchura variation on only one character, length of tibiotarsus. Lonchura pectoralis has been split from the large genus as a montypic genus Heteromunia by several workers. On osteological characters, then, Lepidopygia, Spermestes and Euodice are not valid genera; and Heteromunia is but weakly separate.

Other genera which are weakly characterized on the basis of their skeletons are Aegintha (from Emblema), Chloebia (from Erythrura), and Mandingoa (from Pytilia). Two genera, Pytilia and Emblema, seem composite osteologically. There is some osteological evidence for the lumping of Clytospiza, Hypargos, and Euschistospiza as Goodwin (1982) suggested.

SUPRAGENERIC TAXA

Tribes.--The Family Estrildidae was divided into three tribes by Delacour (1943) and these were recognized by most subsequent workers. Osteological data don't support a division of the family into three tribes or into two tribes, much less their elevation into subfamilies. In fact, the most distinct groups osteologically (See Table 1) are Pyrenestes (with Spermophaga) and Parmoptila (with Nesocharis and Nigrita). The PCA (Figs. 11, 12) shows on Fig. 11 about 1/3 of Estrildini over lapping 1/3 of Lonchurini, with both of those tribes mostly overlapping Poephilini. In Fig. 12 there is nearly complete overlap of all three tribes. Different tribal placement of Amadina and Aidemosyne does not change this point.

Subfamilies.--Osteological data are equivocal as to whether the Viduinae belong in Estrildidae or Ploceidae or in a separate family by themselves. Seven species (21 specimens) of Vidua (the whydahs and indigo birds) and two specimens of Anomalospiza imberbis (the Cuckoo Finch) were examined, with the results in comparison to the Estrildidae presented in Table 3.

Only the first two characters in Table 3 show consistent differences between Estrildidae and the other two genera. The next five characters show Anomalospiza within Estrildid variation, but some or all Vidua outside. The last three characters show Vidua within Estrildid variation but Anomalospiza outside, as noted by Lahti & Payne (2003). Those authors also noted a large median groove in the premaxilla. This observation in Anomalospiza is confirmed but not enough Estrildidae were compared with rhamphotheca removed. (Most of the specimens of Estrildidae had been returned to their owners before this comparison.) Not a large enough series of skeletons of any species was examined to speak to their character of delayed pneumatization of the skull.

In summary, Vidua and Anomalospiza are each more distinct from the 26 genera of estrildids on osteological characters than is any genus within that group except, perhaps, Pyrenestes. However, Vidua and Anomalospiza bear about the same weight of osteological distinction from the other subfamilies of Ploceidae as they do from Estrildidae.

Families.--Sushkin (1927) listed skeletal differences between the Estrildidae (excluding Viduinae) and Ploceidae (Bubalornitinae, Passerinae, and Ploceinae). Findings in this more extensive study are listed in Table 4, with an asterisk (*) marking that mentioned by Sushkin. Only the first character is consistent; the other 11 show some overlap. Presumably the large vomer in Estrildidae functions in seed hulling (Figs. 15-17; cf Ziswiler 1979).

Eight more family-differentiating characters mentioned by Sushkin (1927) were found useless. In seven of these Sushkin did not realize the variability within Estrildidae because he lacked specimens of Parmoptila, Nigrita, Nesocharis, and several other genera. These characters were: position of the club of the maxillopalatine, medial curvature of the rostral palatine, length of the interpalatine space, position of the palatal crests, shape of the rostral end of the vomer in ventral view, nature of the pterygoid-palatine joint, and size or presence of an interorbital fontanel. One character mentioned by Sushkin, the constriction of the rostral palatine in its caudal region in a direct ventral view is better expressed as the twist of the bone at that point, as described by Tordoff (1954) and above. The actual relative width of the palatine at that point varies little from species to species.

Schodde & Mason (1999) included five skeletal characters in their characterization of the family Estrildidae. Of these, three are evidently the same as those listed above, although the terminology is different. One was noted, but not tabulated--"vestigially toothed mandibular angle." One, "vestigial maxillopalatine process," was too variable within Estrildidae, even within species, to be useful.

The PCA figures show a broad overlap between Estrildidae and Ploceidae (four subfamilies, with Anomalospiza in Viduinae) in Fig. 14 and a lesser overlap between Estrildidae and Ploceidae of three of these subfamilies (Passerinae, Viduinae, and Bubalornithinae) in Fig. 13.

In five more tabulated characters there was some difference in a few species between Estrildidae and Ploceidae, but there were large areas of overlap: size of the retroarticular process of the mandible, shape of the caudal basibranchial, presence and size of a manubriumsternum bridge, ratio of tarsometatarsus length to skull length, and ratio of humerus length to femur length.

DISCUSSION

Traylor (1968) put Pholidornis rushiae (Cassin) in Estrildidae as a genus incertae sedis. Apparently there is no complete skeleton in the world's museums (Wood & Schnell 1986), but the lone skull in the British Museum was examined. It is certainly not an estrildid; it exceeds the variation in that family on 11 of the 26 tabulated osteologic characters of the skull. The premaxilla is too long and slender, there is no palatine process of the premaxilla, there is no bone in the nasal septum, the vomer bears no horns, and the rostral palatines aren't twisted, etc. However, with neither a hyoid or a humerus a more specific family identification is unwise. Sefc et al. (2003) placed Pholidornis in the Sylvioidea, close to Cisticolidae.

The small, slender, rostral tongue bone figured by Ziswiler (1979) and called hypentoglossum in Estrildidae was not clear in most dried skeletons. It was clear, however, in a few specimens of four different genera. No doubt its presence or absence must be ascertained in fresh or spirit-preserved material for an analysis.

Schodde & Mason (1999) revised the genera of Australian Estrildidae. They put Aegintha into Neochmia, segregated Emblema bellum, E. oculatum, and E. guttatum into Stagonopleura, and left E. pictum as the only species in Emblema. Also they moved Aidemosyne into Neochima and split Poephila into Taeniopygia with guttata and bichenovii, leaving Poephila including only P. personata, P. acuticauda, and P. cincta. Chloebia they lumped into Erythrura. Of these changes, osteologic characters do not agree with their disposition of Aegintha--rather it should be put in Stagonopleura. Emblerna guttatum should be left in Emblema, which would then require changing the name Stagonopleura to Zonaeginthus. The placement of Aidemosyne in Neochmia agrees with osteology. The split of Poephila is quite unnecessary--the five species form a close group on osteologic evidence. The placement of Chloebia within Erythrara is equivocal on osteologic characters. Moving of the genus Amadina to a position within the almost strictly African Estrildini as advocated by Waiters (1957), Goodwin (1982), Baptista et al. (1999), and Fry & Keith (2004) is neither supported nor denied by osteologic data.

Evolutionary history within Estrildidae has been studied by several workers. Delacour (1943) and Steiner (1960) nominated Clytospiza as most like an ancestral form. Mayr (1968) was not explicit, but seemed inclined toward either Clytospiza or Parmoptila as nearest an ancestral form. Goodwin (1982) suggested that Amandava was probably nearest to an ancestral estrildid. Of the several molecular studies (Kakizawa & Watada 1985; Christidis 1987; Baptista et al. 1999; Sorenson & Payne 2001, 2002; Payne & Sorenson 2003; Sorenson et al. 2004; Van der Meij et al. 2005) only Sorenson & Payne studied nearly all of the genera and a majority of the species. All of these workers found an ancient divergence between the African Estrildini and the primarily Australasian Lonchurini plus Poephelini. However, Kakizawa & Watada found Australasian Erythrura and Chloebia within the African group, and Christidis found African Amadina within the Australasian group, near Lonchura. Sorenson & Payne (2001, 2002) found Ortygospiza locustella (= Paludipasser locustella) distinct and representing a group ancestral to all other estrildids. (That species was examined by none of the other molecular workers nor was a skeleton available for this study.)

The recent treatment of Estrildidae by Sorenson & Payne (2001, 2002), Payne & Sorenson (2003), and Sorenson et al. (2004) should be compared with this work. Actually, there are only these distinct differences: (1) There are no skeletal data on Paludipasser (= Ortygospiza) locustella to compare. (2) They recognize 13 more genera than this work, including a split of Lonchura into six genera; most of these were subgenera in the Goodwin (1982) scheme.

Relationships of Vidua and Anomalospiza to Estrildidae have been variously hypothesized. Most systematists who have studied the question--Chapin (1917, 1929, 1954), Sushkin (1927), Delacour (1943), Tordoff (1954), Wolters (1957, 1960), Friedman (1960, 1962), Bentz (1979), Sibley & Ahlquist (1990), Lahti & Payne (2003), and Fry & Keith (2004)--placed the whydahs closer to their hosts, the waxbills, than to any other systematic group. On the other hand, Beecher (1953), Steiner (1960), Nicolai (1964), Ziswiler (1967), Mayr (1968), Mayr et al. (1968), and Goodwin (1982) placed the whydahs closer to one or another ploceid group than to the waxbills. Recent studies of Anomalospiza by Sorenson & Payne (2001) and Lahti & Payne (2003) placed that genus close to Vidua. Cramp & Perrins (1994) and Fry & Keith (2004) separated Passeridae as a separate family from Ploceidae.

The family Estrildidae is a relatively homogenous group on osteologic characters, contrasting with Emberizinae (Webster & Webster 1999) and Icteridae (Webster 2003) in this respect. No taxonomic changes in the arrangement in Goodwin (1982) above the species level are recommended except the deletion of Pholidornis, the dropping of tribes, and the lumping of Lepidopygia with Lonchura. Probably, also, the genera Clytospiza and Euschistospiza should be united with Hypargos following the suggestion by Goodwin (1982). Probably the genus Zonaeginthus should be recognized to include Emblema bellure, Emblema oculatum, and Aegintha temporalis. Probably Aidernosyne modesta should be transferred to Neochmia. Recognition of the genera Padda and Heteromunia (splits from Lonchura not recognized by Goodwin (1982) are equivocal, as is the lumping of Chloebia with Erythrura. Whether the subfamily Viduinae should be transferred from Ploceidae to Estrildidae or segregated as a separate family is not clarified by osteology.

Appendix 1.

Specimens examined. Museum designations: AM, American Museum of Natural History. CAS, California Academy of Sciences. CM, Carnegie Museum of Natural History. FL, Florida State Museum. FM, Field Museum of Natural History. HA, Hanover College. KS, University of Kansas Museum of Natural History. LAC, Los Angeles County Museum. LSU, Louisiana State Museum of Science. MI, University of Michigan Museum of Zoology. RO, Royal Ontario Museum. BM, The Natural History of Museum of the United Kingdom. US, United States National Museum. YP, Peabody Museum of Natural History of Yale University.

Estrildidae:

Parmoptila woodhousei, 5 FM385327, US292451, US292452, BM S/1980. 6.2, AM24935

Nigrita fusconota, 4 FM313270, FM3835328, US322542, US292481

Nigrita bicolor, 3 US 347578, US292479, US292480

Nigrita canicapila, 5 FM356422, KS70888, US291136, US292478, YP9410

Nesocharis shelleyi, 2 US318625, US318626

Nesocharis capistrata, 1 MI201774

Pytilia phoenicoptera, 3 AM12715, AM12829, BM S/1966. 39.20

Pytilia hypogrammica, 3 AM12268, AM14084, AM 16495

Pytilia afra, 3 AM12346, AM12660, AM12347

Pytilia melba, 5 US430316, US429100, US558781, LAC89770, LAC89768

Mandingoa nititula, 6 AM13700, AM13698, AM13701, MI1214380, MI207779, BM1981.82.2

Cryptospiza reichenovii, 2 BM S/1992.27.1, S/ 992.27.2

Cryptospiza salvadorii, 5 FM356455, FM356459, FM356458, KS70941, KS70942

Pyrenestes sanguineus, 4 AM13696, AM13759, AM13697, BM S/1993.38:1

Pyrenestes ostrinus, 4 AM12686, AM17105, AM17104, MI232532

Pyrenestes minor, 2 M1214379, M1214378

Spermophaga haematina, 6 US347593, US322435, US322434, CAS71723, MI2221317, MI1221316

Spermophaga ruficapilla, 4 MI223525, MI208342, US499894, YP13765

Clytospiza monteiri, 4 MI201775, MI201776, MI232533, AM24947

Hypargos niveoguttatus, 4 CAS71724, AM11819, KS80654, CAS71724

Euschistospiza dybowskii, 3 AM14260, AM14261, AM14263

Lagonostica rara, 3 CAS71727, CAS71726, CAS71727

Lagonostica rufopicta, 4 AM14304, AM14046, MI221300, MI221308

Lagonostica nitidula, 2 MI203301, BM S/ 958.19.1

Lagonosticta senegala, 4 US430652, US502062, CAS83508, CAS71730

Lagonostica rubricata, 6 CAS84857, AM14082, AM13205, AM13204, BMS/1986.75.26, BM S/ 5.24

Lagonosticta rhodopareia, 3 US430676, US430807, BM S/1982.77.3

Lagonosticta larvata, 3 AM14135, MI221311, MI221312

Uraeginthus angolensis, 5 US558782, US431594, US430001, BM S/1982.136.5, BM S/ 982.136.2.

Uraeginthus bengalus, 5 CAS71738, CAS85089, LAC89787, LAC89790, YP9272

Uraeginthus cyanocephalus, 4 AM8759, AM8809, AM8767, CAS90061

Uraeginthus granatina, 6 CAS71734, MI222633, MI220363, MI233565, BM S/1991.1.27, BMS/ 991.1.28

Uraeginthus ianthinogaster, 4 AM11915, AM'12086, AM12087, BM S/1987.23.15

Estrilda caeruleseens, 4 CAS42861, US347286, US557890, CAS42861

Estrilda perreini, 1 BM S/1962.18.1

Estrilda melanotis, 3 MI218669, CAS90031, CAS90030

Estrilda melanotis quartinia, 1 MI218670

Estrilda paludicola, 3 FM362222, FM362221 FM362220

Estrilda melpoda, 4 CAS71735, MI223530, MI223531, BM 1891.7.20.188

Estrilda rhodopyga, 3 AM11812, AM11832, CAS90074

Estrilda troglodytes, 5 AM11829, AM11818, AM12173, BM S/1966.39.10, BM S/1966.39.8

Estrilda astrild, 5 US490778, US490874, US490779, BM S/1990.3.3, BM S/1990.3.2

Estrilda nonnula, 3 LAC89798, LAC90692, LAC89800

Estrilda atricapilla, 3 FM385352, FM358204, FM313281

Estrilda erythronota, 4 CAS71422, MI222628, MI233264, CAS90073

Amandava amandava, 6 US556164, US5556156, FM96103, CAS90068, CAS90026, BM S/ 1962.12.2

Amandava formosa, 3 LAC90694, FM106437, BM S/1986.75.38

Amandava subflava, 5 US322525, US432034, US322505, KS31150, CAS90027

Ortygospiza atricollis, 5 AM1410, AM14141, AM14138, BM S/1998.48.73, CAS90105

Ortygospiza species, 2 BM S/1967.5.1, FL31619 (incomplete data, measured by Auth)

Aegintha temporalis, 5 MI204457, MI205437, MI205436, FM105467, FM105492

Emblema pictum, 6 CAS57594, CAS60370, AM12831, AM11814, BM S/1954.7.1, CAS60371

Emblema bellum, 3 RO186380, RO136166, RO136378

Emblema oculatum, 2 FM104083, FL31621 (incomplete data, measured by Auth)

Emblema guttatum, 6 CAS84574, CAS57637, AM10737, AM10331, BM1867.7.8.16, BM 1865.5.20.13

Neochmia phaeton, 4 CAS60372, AM12964, AM12966, BM S/1998.48.52

Neochmia ruficauda, 4 CAS83891, CAS71740, BM S/1984.28.1, CAS71738

Poephila guttata, 8 MI220494, MI205109, MI219070, KS83369, KS83322, KS80636, CAS90110, CAS90058

Poephila bichenovii, 4 US620269, US343821, US620270, BM S/1983.129.1

Poephila personata, 5 MI205599, MI205598, MI119233, FM104058, FM96083

Poephila acuticauda, 5 US345241, US354284, US344878, BM S/1984.28.10, BM S/ 1984.28.11

Poephila cincta, 5 CAS43603, MI119234, FM96152, FM105362, KS80879

Erythrura hyperthra, 4 CAS85132, AM11846, AM11844, BM S/1984.27.1

Erythrura prasina, 6 US344594, US344497, US614130, CAS71753, BM S/1989.2543, BM S/1987.23.20

Erythrura viridifacies, 3 BM S/1982.104.1, BM S/1983.133.1, BM S/1984.30.1

Erythrura trichroa, 5 MI210799, MI209458, MI210798, FM104454, FM104172

Erythrura psittacea, 5 CAS48970, MI221643, MI220930, KS80964, KS80963

Erythrura tricolor, 2 CAS90129, CAS No #

Erythrura pealii, 4 CAS42707, CAS90118, BM S/1993.16.4, FL31650 (incomplete data, measured by Auth)

Chloebia gouldiae, 7 US321161, US501540, CAS71477, CAS84820, BM S/1954.1.1, BM S/1981.91.2, CAS60949

Aidemosyne modesta, 3 CAS57865, MI205612, MI205613

Lepidopygia nana, 2 MI208433, BM 1897.5.10.48

Lonchura cantans, 3 AM12913, AM11994, AM12263

Lonchura griseicapilla, 5 CAS85726, FM362202, FM362201, KS70556, KS70524

Lonchura cucullata, 3 US556170, US556174, US556076

Lonchura bicolor, 3 US347586, US347587, US347585

Lonchurafringilloides, 3 LSU160452, LSU111948, AM16453

Lonchura striata, 5 CAS84020, CAS85461, FM315110, FM104099, CAS71540

Lonchura leucogastroides, 3 CAS85105, CAS85712, BM S/1985.32.1

Lonchura fuscans, 2 KS41000, KS41118

Lonchura molucca, 5 US557491, US557484, US557485, US557489, US557483

Lonchura punctulata, 4 CAS85727, MI235135, MI207365, LAC89840

Lonchura leucogastra, 3 MI206402, MI206399, MI206401

Lonchura tristissima, 5 MI21585, LSUI01032, YP9000, YP8999, YP8998

Lonchura leucosticta, 3 AM12832, AM11796, AM12834

Lonchura quinticolor, 4 CAS85108, CAS90081, CAS90050, CAS90049

Lonchura malacca, 4 KS31144, KS41357, CAS84904, CAS No #

Lonchura maja, 4 CAS85109, CAS85728, US345468, US344591

Lonchura grandis, 3 LSUI01033, YP8977, YP8978

Lonchura vana, 1 KS41113

Lonchura caniceps, 4 LAC89850, LAC89849, FM290944, YP6190

Lonchura nevermanni, 3 AM12948, AM12657, AM11835

Lonchura spectabilis, 7 CAS85111, KS81180, US489252, US289250, YP8987, YP8991, YP8985

Lonchura hunsteini, 3 LSU85929, LSU86921, LSU86928

Lonchuraflaviprymna, 3 FM105853, FM105901, FM314884

Lonchura castaneothorax, 4 CAS85309, FM343031, FM96143, CAS90080

Lonchura teerinki, 2 CAS90147, CAS90159

Lonchura monticola, 1 AM5820

Lonchura melaena, 1 BM S/1961.11.48

Lonchura pectoralis, 5 AM 12906, MI214327, KS83207, CM7240, CM7242

(Padda) fuscata, 3 CAS85729, CAS90085, CAS90059

(Padda) oryzivora, 4 CAS58654, CAS71539, CAS85747, CAS85314

Amadina erythrocephala, 4 US429108, US429109, US429107, BM S/2003.4.3

Amadina fasciata, 8 CAS85625, CAS85536, AM11963, AM11965, LAC89852, LAC102979, YP102898, YP103284

Hyliidae:

Pholidornis rushiae, 1 BM 1930.12.3.11

Ploceidae, Viduinae:

Vidua chalybeata, 3 MI222621, MI218102, MI222620

Vidua funerea, 2 CM9573, CM9571

Vidua funerea purpurascens, 2 MI223739, MI223738

Vidua fischeri, 3 BM S/1993.30.1, MI217526, MI212960

Vidua regia, 2 MI204760, MI222615

Vidua macroura, 2 MI212961, MI136012

Vidua paradisaea, 5 MI222619, MI224501, RONI49481, RON126721, BM 1982.54.1

Vidua orientalis, 2 CAR9570, CAR9569

Anomalospiza imberbis, 2 MI217511, MI219072

Ploceidae, Bubalornithinae:

Bubalornis albirostris, 2 RON156913, US430787

Dinemiellia dinemelli, 2 KS70745, KS70744

Ploceidae, Passerinae: Plocepasser mahali, 1 KS70632

Pseudonigrita arnaudi, 2 CM7266, CM7267

Passer domesticus, 9 HA3056, HA3059, HA3205, records lost of 6

Passer montanus, 1 MIl19071

Petronia xanthocoUis, 2 KS70447, KS70773

Sporopipes frontalis, 4 KS70628, YP13245, YP13250, US490719

Ploceidae, Ploceinae:

Amblyospiza albifrons, 3 CM789, CM785, US322509

Ploceus ocularis, 2 RO114496, RO114466

Ploceus melanogaster, 1 KS71003

Ploceus cucullatus, 5 YP9413, YP7500, MI221231, MI221191, MI221216

Ploceus philippinus, 1 YP9828

Malimbus rubricollis, 2 YP9412, YP9411

Malimbus malimbicus, 1 US291132

Malimbus rubriceps, 2 RO114571, RO114577

Quelea quelea, 7 RO12076, RO120766, MI219934, MI219943, MI219949, MI219953, MI219958

Foudia madagascariensis, 1 US432205

Euplectes orix, 2 CM1575, CM16514

Euplectes ardens, 2 KS71020, KS71019

Appendix 2.

Species of Estrildidae whose skeletons were not studied. An asterisk (*) means that no skeleton exists in the world's museums, according to Wood & Schnell (1986). Thirteen species recognized by various recent workers but not by Goodwin (1982) are included.

Parmoptila rubrifrons

Nigrita luteifrons *

Nesocharis ansorgei *

Pytilia lineata

Cryptospiza jacksoni

Cryptospiza shelleyi *

Spermophaga poliogenys

Hypargos margaritatus

Euschistospiza cinereovinacea

Lagonsticta landanae *

Lagonosticta virata

Lagonosticta vinacea

Estrilda thomensis

Estrilda rufibarba

Estrilda nigriloris *

Estrilda charmosyna *

Estrilda polioparia

Estrilda kandti

Ortygospiza gabonensis *

Ortygospiza locustella *

Oreostruthus fuliginosus *

Erythrura coloria *

Erythrura papuana

Erythrura regia

Erythrura kleinschmidti *

Erythrura cyaneovirens

Lonchura malabarica

Lonchura nigriceps

Lonchura kelaarti

Lonchura ferruginosa

Lonchura pallida *

Lonchura forbesi

Lonchura nigerrima

Lonchura stygia

Lonchura montana *

ACKNOWLEDGMENTS

In 2000, shortly before his death, Luis Baptista requested and encouraged me to study the skeletons of Estrildidae. This paper is the result of that encouragement, but without his help. I thank the curators of these museums for allowing me to study specimens under their care: American Museum of Natural History, California Academy of Sciences, Carnegie Museum of Natural History, Field Museum of Natural History, University of Kansas Museum of Natural History, Los Angeles County Museum, Louisiana State University Museum of Science, University of Michigan Museum of Zoology, Royal Ontario Museum, The Natural History Museum of the United Kingdom, United States National Museum, Peabody Museum of Natural History of Yale University. I thank David L. Auth for measuring three skeletons at the Florida Museum of Natural History. Richard L. Zusi kindly explained some old, obscure skeletal terminology for me. Jackson R. Webster performed the statistical analysis. The manuscript was helpfully reviewed by Michael A. Patten and three anonymous reviewers.

Manuscript received 12 April 2006, revised 5 October 2006.

LITERATURE CITED

Baptista, L.F., R. Lawson, E. Visser, & D.A. Bell. 1999. Relationships of some mannikins and waxbills in the Estrildidae. Journal fur Ornithologie 140:179-192.

Beecher, W.J. 1953. A phylogeny of the oscines. Auk 70:270-333.

Bentz, G.D. 1979. The appendicular myology and phylogenetic relationships of the Ploceidae and Estrildidae (Aves: Passeriformes). Bulletin of the Carnegie Museum of Natural History 15:1-25.

Bock, W.J. 1960. The palatine process of the premaxilla in the Passeres. Bulletin of the Museum of Comparative Zoology 122(3):361-488.

Chapin, J.P. 1917. The classification of the weaverbirds. Bulletin of the American Museum of Natural History 37:243-280.

Chapin, J.P. 1929. Nomenclature and systematic position of the paradise whydahs. Auk 4:474-484.

Christidis, L. 1986. Chromosomal evolution within the family Estrildidae (Aves) I, the Poephilae. Genetica 71:81-97.

Christidis, L. 1987. Biochemical systematics within Paleotropic finches (Aves: Estrildidae). Auk 104:380-391.

Cramp, S. & C.M. Perrins (eds.). 1994. The Birds of the Western Palearctic. Vol. 8. Oxford University Press, Oxford. 899 pp.

Delacour, J. 1943. A revision of the subfamily Estrildidae of the family Ploceidae. Zoologica 28: 69-86.

Foelix, R.F. 1970. Vergleichend-morphologische Unterschungen an der Speicheldrusen kornerfressender Singvogel. Zoologischen Jahrbuch Anatomie 87:523-587.

Friedmann, H. 1960. The Parasitic Weaverbirds. Bulletin of the United States National Museum 223:1-196.

Friedmann, H. 1962. The problem of the Viduinae in the light of recent publications. Smithsonian Miscellaneous Collections 145(3):1-10.

Fry, C.H. & S. Keith (eds.). 2004. The Birds of Africa. Vol. VII. Princeton Univ. Press, Princeton, New Jersey. 666 pp. (Estrildidae by various authors: C.H. Fry, S. Keith, C.J.H. Hines, T.B. Smith, & R.B. Payne).

Goodwin, D. 1982. Estrildid Finches of the World. British Museum (Natural History). Ithaca, New York. 328 pp.

Guttinger, H.R. 1970. Zur Evolution von Verhaltensweisen und Lautaussernngen bei Prachtfinken (Estrildidae). Zoologishen Tierpsychologie 27:1011-1075.

Guttinger, H.R. 1976. Zu systematischen Stellung der Gattung Amadina, Lepidopygia, und Lonchura (Aves, Estrildidae). Bonn Zoologischen Beitrag 27:218-244.

Kakizawa, R. & M. Watada. 1985. The evolutionary genetics of the Estrildidae. Journal of the Yamashina Institute of Ornithology 17:143-158.

Lahti, D.C. & R.B. Payne. 2003. Morphological and behavioral evidence of relationships of the Cuckoo Finch Anomalospiza imberbis. Bulletin of the British Ornithologists' Club 123.113-125.

Mayr, E. 1968. The sequence of genera in the Estrildidae (Aves). Breviora of the Museum of Comparative Zoology 287:1-14.

Mayr, E., R.A. Paynter, Jr. & M.A. Traylor. 1968. Family Estrildidae, waxbills, grass finches, and mannikins. Pp. 306-390, In Check-list of Birds of the World. (R.A. Paynter, Jr., ed.). Vol. 14. Museum of comparative zoology, Cambridge, Massachusetts. 433 pp.

MjM Software. 2002. Glenenden Beach, Oregon.

Moreau, R.E. & J.C. Greenway, Jr. 1962. Family Ploceidae, weaver birds. Pp. 3-75, In Check-list of Birds of the World. E. Mayr & J.C. Greenway, Jr., eds.). Vol. 15. Museum of Comparative Zoology, Cambridge, Massachusetts. 315 pp.

Nicolai, J. 1964. Der Brutparasitismus der Viduinae als ethoiogisches Problem. Zeitschrift fur Tierpsychologie 21:129-204.

Payne, R.B. & M.D. Sorenson. 2003. Museum collections as sources of genetic data. Bonner Zoologische Beitrage. 2/3:97-104.

Pocock, T.N. 1966. Contributions to the osteology of African birds. Ostrich, Supplement 6:83-94.

Schoddle, R. & I.J. Mason. 1999. The Directory of Australian Birds, Passerines. CSIRO, Canberra, Australia. 851 pp.

Sefc, K.M., R.B. Payne & M.D. Sorenson. 2003. Phylogentic relationships of African sunbird-like Warblers: Moho (Hypergerus atriceps), Green Hylia (Hylia prasina) and Tit-hylia (Pholidornis rushiae). Ostrich 74:8-17.

Sibley, C.G. & J.E. Ahlquist. 1990. Phylogeny and classification of birds. Yale University Press. New Haven, Connecticut. 976 pp.

Sibley, C.G. & B.L. Monroe. 1990. Distribution and taxonomy of birds of the world. Yale University Press. New Haven, Connecticut. 1111 pp.

Sorenson, M.D. & R.B. Payne. 2001. A single ancient origin of brood parasitism in African finches: Implications for host-parasite coevolution. Evolution 55:2550-2567.

Sorenson, M.D. & R.B. Payne. 2002. Molecular genetic perspectives on avian brood parasitism. Integrative and Comparative Biology 42:388-400.

Sorenson, M.D., C.M. Balakrishnan & R.B. Payne. 2004. Clade-limited colonization in brood parasitic finches (Vidua spp.). Systematic Biology 53: 140-153.

Steiner, H. 1960. Klassification der Prachtfinken, Spermestidae, auf Grund der Rachenzeichnungen ihrer Nestlinge. Journal fur Ornithologie 101:92-112.

Sushkin, P. 1927. On the anatomy and classification of the weaver-birds. Bulletin of the American Museum of Natural History 57:1-32.

Tordoff, H.B. 1954. A systematic study of the avian family Fringillidae based on the structure of the skull. Miscellaneous publications, Museum of Zoology. University of Michigan 81:1-63.

Traylor, M.A. 1968. Subfamily Viduinae, Indigo-birds and Whydahs. Pp. 390-397, In Check-list of Birds of the World. Vol. 14. (R.A. Paynter, Jr., ed.). Museum of Comparative Zoology, Cambridge, Massachusetts. 433 pp.

Van der Meij, M.A.A., M.A.G. deBakker & R.G. Bout. 2005. Phylogenetic relationships of finches and allies based on nuclear and mitochondrial DNA. Molecular Phylogenetics and Evolution 34:97-105.

Vernon, C.J. & W.R.J. Dean. 1975. On the systematic position of Pholidornis rushiae. Bulletin of the British Ornithologists' Club 95:20.

Webster, J.D. 1993. The manubrium-sternum bridge in songbirds (Oscines). Proceedings of the Indiana Academy of Science 101:299-308.

Webster, J.D. 2003. Skeletal characters and the genera of blackbirds (Icteridae). Condor 105: 239-257.

Webster, J.D. & J.R. Webster. 1999. Skeletons and the genera of sparrows (Emberizinae). Auk 116: 1054-1074.

Wolters, H.E. 1950. Uber einige Gattungen der Estrildidae. Bonn Zoologische Beitrag 1:31-38.

Wolters, H.E.. 1957. Die Klassifikation der Webefinken (Estrildidae). Bonn Zoologische Beitrag 8:90-129.

Wolters, H.E. 1966. On the relationships and genetic limits of African Estrildinae. Ostrich, Supplement 6:75-81.

Wood, D.S. & G.D. Schnell. 1986. Revised world inventory of avian skeletal specimens. American Ornithologists' Union, Norman, Oklahoma. 296 PP.

Ziswiler, V. 1967. Vergleichend morphologische Untersachungen am Verdaungstraht korner-fressendender Singvogel zur Abklarung ihrer systematischen Stellung. Zoologischen Jahrbuch Systematik 94:427-520.

Ziswiler, V. 1979. Zungenfunktionen und Zungenvesteifung bei granivoren Singvogeln. Revue Suisse Zoologie 86:823-831.

Ziswiler, V., H.R. Guttinger & H. Bregulla. 1972. Monographie tier Gattung Erythrura Swainson 1837 (Aves, Passeres, Estrildidae). Bonner Zoologische Monographien 2:1-158.

J. Dan Webster: Department of Biology, Hanover College, Hanover, Indiana 47243 USA
Table 1--Best generic characters observed in Estrildidae. For
explanation, see text, pages 92-94.

 Maxillary
 process of
 nasal shape:
 A = broad &
 Caudal heavy, B =
 basibranchial moderately
 shape: D broad, C =
 = depressed, fairly Transpalatine
 N = not slender, D = process shape:
 Genus depressed slender length/width

Parmoptila N C 3.80
Nigrita D or N B&C 3.0-7.3
Nesocharis Not B&C&D 2.8-3.0
Pytilia D or N B&C 1.3-1.8
Mandingoa N A&B&C 2.20
Cryptospiza N B&C 2.0-2.6
Pyrenestes N A 1.0-1.6
Spermophaga N A&B 1.2-1.5
Clytospiza N C 1.30
Hypargos D or N B&C 1.30
Euschistospiza D or N C&D 1.60
Lagonosticta N B&C 1.2-1.4
Uraeginthus D or N C&D 1.2-1.8
Estrilda N B&C&D 1.3-3.8
Amandava N B&C&D 1.2-2.1
Ortygospiza D or N C&D 1.5-2.4
Aegintha N C 1.60
Emblema D or N A&B&C&D 0.8-3.5
Neochmia D or N A&C 1.0-1.2
Poephila D or N B&C 0.4-0.9
Erythrura D or N A&B&C 2.0-4.5
Chloebia D or N A&B&C 2.20
Aidemosyne D C 1.00
Lonchura D or N A&B&C&D 0.7-3.0
(Padda) D A&B&C 1.6-1.8
Lepidopygia N B&C 2.20
Amadina N C 1.1-1.3

 Palatine
 pro cess
 pre-max illa:
 O = absent,
 sf = slight
 flange, f = Parasphenoidal
 broad flange, rostrum keel:
 --extra P = prominent,
 process: 0 = Pneumotricipital S = slight, R =
 absent, number fossa: See text absent, N
 Genus = width page 94 = not observed

Parmoptila O & sf--0 A&E&F S&P
Nigrita sf&f--0-0.3 A&F&DO S
Nesocharis sf&f--0-0.3 A&E&F N
Pytilia f--0-0.2 A&F S
Mandingoa f--0.2 A&F N
Cryptospiza f--0.05-0.2 A&F P
Pyrenestes f--0.1-0.2 A&E&F R
Spermophaga f--0.05-0.2 A&E&F&DO S&R
Clytospiza f--0 A&E&DT S
Hypargos f--0 A&E S
Euschistospiza f--0.1 A N
Lagonosticta f--0-0.1 A&E&F S&P
Uraeginthus f--0-0.2 A&E&F S&P
Estrilda f--0-0.2 A&E&F&DO S
Amandava f--0-0.1 A&E&F S&P
Ortygospiza f--0.2 A&E&F S&R&P
Aegintha f--0.05 A N
Emblema f--0-0.2 A&F&DT S&R
Neochmia f--0.1-0.2 A&E&F S&P
Poephila f--0.1-0.2 A&F&DO S&P
Erythrura f--0.1-0.2 A&E&F&DO S&R&P
Chloebia f--0.2 E&F S&P
Aidemosyne f--0.2 A P
Lonchura f--0-0.2 A&E&F P
(Padda) f--0-0.1 A&E&F P
Lepidopygia f--0.1 A&F P
Amadina f--0.1 A&E&F P

 Interorbit al
 septum:
 C = completely
 ossified, S =
 Tibiotarsus small gap,
 length Skull length L = large
 measured: measured: gap = N =
 Genus (mm) (mm) served not observed

Parmoptila 20.3 15.6 C&S&L
Nigrita 21.8-25.0 16.0-18.4 C&S&L
Nesocharis 19.0-23.8 13.8-18.4 S
Pytilia 22.9-24.3 16.2-17.0 C&S
Mandingoa 22.5 16.6 C
Cryptospiza 25.0-25.9 17.5-17.6 C&S&L
Pyrenestes 25.0-27.8 18.6-21.0 C
Spermophaga 28.9-29.8 19.6-21.2 C&S&L
Clytospiza 25.0 17.2 C&L
Hypargos 24.1 17.3 C
Euschistospiza 23.0 16.8 N
Lagonosticta 18.2-21.1 14.5-16.2 C&S&L
Uraeginthus 21.0-24.8 15.3-17.0 C&S&L
Estrilda 18.9-21.3 14.0-15.4 C&S&L
Amandava 17.4-21.6 13.0-15.2 C&S&L
Ortygospiza 20.2-20.6 13.8-15.1 C&S&L
Aegintha 22.4 15.5 C&S
Emblema 20.7-25.7 15.6-17.4 C&S&L
Neochmia 20.9-21.9 15.8-16.2 C&S&L
Poephila 20.2-22.7 15.0-16.4 C&S&L
Erythrura 21.7-24.6 16.6-18.0 C&S&L
Chloebia 22.9 15.9 C&S
Aidemosyne 22.0 15.6 L
Lonchura 19.0-26.7 13.3-18.8 C&S&L
(Padda) 26.2-27.8 18.3-19.5 C&S&L
Lepidopygia 18.4 13.8 C
Amadina 23.2-24.6 17.2-18.3 C&L

 Ectethmoid
 foraminoa: D =
 double, P =
 pinched, S = Interpalatine
 single, T = process: O = Tarsometatarsus
 triple. absent length/ shape: length/
 Genus Lateral/medial width width

Parmoptila D&P 1.1 0.2 9.6
Nigrita D&P&T 1.3-1.8 0.3-1.7 8.6-9.1
Nesocharis D&P 2.0 0-1.0 9.6-9.9
Pytilia D&P 1.3-3.3 0.2-1.2 8.2-9.4
Mandingoa D 2.1 0.2 8.3
Cryptospiza D 1.8-1.9 0 10.0-10.2
Pyrenestes P&S 6.0-14.0 0 8.2-9.4
Spermophaga D&P&S&T 3.7-4.5 0.2-0.3 9.4-9.8
Clytospiza D&P 1.6 0.8 9.2
Hypargos D&P 1.3 0.2 8.9
Euschistospiza P&S 1.3 0.8 8.9
Lagonosticta MS 1.0-2.0 0.5-2.5 8.5-9.2
Uraeginthus D&P&S 1.0-1.2 0.2-2.0 9.1-9.8
Estrilda D&P 1.6-5.7 0-1.0 9.5-11.1
Amandava D&P&S&T 1.3-1.7 0.4-1.2 8.6-10.1
Ortygospiza D&P&S 1.0-1.7 0.7-1.0 9.0-9.4
Aegintha D&P&T 2.3 0 9.8
Emblema D&P 1.5-4.0 0.4-2.5 8.7-9.6
Neochmia D&T 4.5-7.0 0.4-1.8 8.6-8.7
Poephila D&P 1.9-4.2 0-0.4 8.6-9.8
Erythrura D 1.1-2.3 0 7.6-9.3
Chloebia D 2.0 0.1 7.8
Aidemosyne D 1.9 0.3 9.1
Lonchura D&P&S 1.3-10.0 0-1.6 8.1-10.9
(Padda) D&P 4.0-6.8 0.8-1.8 83-9.0
Lepidopygia P 2.5 0 9.3
Amadina D&P&T 2.7-3.3 0.6-1.5 7.8

 Compression of Premaxilla
 rostral Fusion of shape: length/
 basibranchial: mediopalatines width
 Genus height/width % fused

Parmoptila 3.0 40 1.9
Nigrita 2.2-3.0 27-62 1.4-1.6
Nesocharis 3.1-3.2 25-100 1.3-1.6
Pytilia 2.2-2.7 25-58 1.5-1.7
Mandingoa 3.2 90 1.6
Cryptospiza 3.5 29-42 1.6-1.7
Pyrenestes 4.0-4.9 0 1.1-1.2
Spermophaga 3.8-4.2 0-5 1.4-1.5
Clytospiza 3.0 66 1.5
Hypargos 3.6 0 1.4
Euschistospiza 2.8 23 1.8
Lagonosticta 2.1-3.5 10-50 1.3-1.7
Uraeginthus 2.1-3.3 15-75 1.4-1.6
Estrilda 2.3-3.4 15-100 1.4-1.7
Amandava 2.5-2.8 33-58 1.4-1.5
Ortygospiza 2.5-3.0 18-25 1.2-1.4
Aegintha 3.3 86 1.3
Emblema 3.0-3.4 16-50 1.2-1.7
Neochmia 3.1-3.7 22-61 1.3-1.4
Poephila 2.3-3.7 33-94 1.3-1.4
Erythrura 2.5-4.1 30-88 1.3-1.6
Chloebia 3.5 60 1.2
Aidemosyne 3.2 65 1.3
Lonchura 2.7-4.6 30-100 1.0-1.7
(Padda) 4.1-4.5 50-60 1.4-1.6
Lepidopygia 3.3 62 1.20
Amadina 2.9-3.3 10-43 1.3-1.4

Table 2. --Principal component analysis of rank transformed data.

 PC PC PC
Variable Axis 1 Axis 2 Axis 3

Skull length 0.071 0.094 -0.554
Premaxilla length/
 width 0.020 0.293 -0.041
Cranium width/
 inter-orbital width 0.061 0.368 0.132
Tibiotarsus length/
 ulna length -0.392 -0.04 -0.100
Tibiotarsus length/
 humerus length -0.381 -0.127 -0.026
Tibiotarsus length/
 femur length -0.234 0.016 0.124
Ulna length/femur
 length 0.297 0.064 0.186
Humerus length/
 femur length 0.305 0.193 0.183
Length/width of
 inter-palatine process 0.050 0.080 -0.344
Length/width of
 transpalatine process -0.058 0.095 0.158
Length/width of
 zygomatic process 0.139 -0.245 0.008
Tarscnnetatarsus
 length/femur length -0.265 0.218 0.155
Length/width of
 tarsometatarsus -0.337 0.059 0.164
Tarsometatarsus
 length/skull length -0.231 0.296 -0.042
Ectethmoid foramina -0.117 -0.38 -0.083
Ulna length/
 tarsometatarsus length 0.394 -0.097 0.028
Postorbital process -0.08 -0.312 -0.13
Mediopalatines
 joining -0.083 -0.255 0.213
Tibiotarsus length 0.006 0.107 -0.558
Humerus length/width -0.108 0.411 -0.027
Eigenvalue 5.668 3.682 2.294
Variance explained 0.283 0.184 0.115
Cumulative variance
 explained 0.283 0.468 0.582

Table 3.--Subfamily characters.

 Viduinae
 Estrildinae
Character (or Estrildidae) Anomalospiza

Vomer (Figs. 15-17) Large, horns broad Small, horns narrow

Pseudotemporal process More ventral More dorsal
 of mandible
Humerus length: femur 0.9-1.1 1.1
 length
Ectethmoid foramina, 1.0-14.0 --
 lateral: dorsal
Tibiotarsus length: ulna 1.2-1.6 1.3
 length
Tibiotarsus length: 1.4-1.9 1.5
 humerus length
Process 7b of mandible Smaller, sloping Smaller, sloping
 except 3%
 flat-topped
Manubrium-sternum Absent, minute, or Very large
 bridge small
Rostral end of pterygoid Not expanded except Large, heavy,
 slightly in expanded
 Pyrenestes
Caudal end of jugal Not expanded Expanded

 Viduinae

Character Viclua

Vomer (Figs. 15-17) Small, horns broad or
 narrow
Pseudotemporal process More dorsal
 of mandible
Humerus length: femur 1.2
 length
Ectethmoid foramina, 0.4-3.8
 lateral: dorsal
Tibiotarsus length: ulna 1.1-1.2
 length
Tibiotarsus length: 1.3-1.4
 humerus length
Process 7b of mandible Smaller, sloping
 except 19%
 flat-topped and
 larger in one
Manubrium-sternum Absent or minute
 bridge
Rostral end of pterygoid Not expanded
Caudal end of jugal Not expanded

Table 4.--Family characters. Viduinae excluded. * Noted in Suskin
(1927).

 Character Estrilidac

Vomer (Figs. 15-17) Large; horns long and broad
Rostral palatines Twisted 90[degrees]-120[degrees]
Palatine process of premaxilla Lateral flange, usually prominent
 (Tordoff 1954, Bock 1960
Lateral wall of nasal capsule * Not ossified
Lateral ectethmoid foramen: 1-14, usually high
 medial foramen
Interpalatine process, length: 0-2.5, often absent
 width
Postorbital process of squamosal 0.3-0.7 of way to jugal arch
Pseudotemporal process of 0.4-1.8, usually below 1.0
 mandible
Pneumotricipital fossa of humerus A, E, F DO, or DT; 3% DO or DT
Tibiotarsus length: 1.3-1.7
 tarsometatarsus length
Tibiotarsus length 17.4-29.8 mm
Skull length 13.0-21.2 mm

 Ploceidae (Bubalornithinae,
 Character Passerinae, and Ploceinae)

Vomer (Figs. 15-17) Small; horns usually short and
 narrow
Rostral palatines Not twisted (12 species) or
 twisted (eight species)
Palatine process of premaxilla Not a flange, often separated by
 (Tordoff 1954, Bock 1960 a suture (12 species) or slight
 flange (seven species) or
 prominent flange (one species)
Lateral wall of nasal capsule * Bone in seven species of four
 genera
Lateral ectethmoid foramen: 0.3-9.5, usually low
 medial foramen
Interpalatine process, length: 0.8-5.2, usually long
 width
Postorbital process of squamosal 0.1-0.3 except Euplectes 0.4
Pseudotemporal process of 0.8-4.7, usually over 1.0
 mandible
Pneumotricipital fossa of humerus A, E, F, DO, DT, or DS; 31% DO or
 DT or DS
Tibiotarsus length: 1.3-1.5
 tarsometatarsus length
Tibiotarsus length 23.7-46.0 mm (eight of 20 species
 over 29.8)
Skull length 17.3-27.4 mm (10 of 20 species
 over 21.2)
COPYRIGHT 2007 Indiana Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Webster, J. Dan
Publication:Proceedings of the Indiana Academy of Science
Article Type:Report
Geographic Code:1USA
Date:Aug 9, 2007
Words:9520
Previous Article:Terrestrial mammals of the Naval Support Activity (NSA Crane) in Martin County, and of the Crane Division of the Naval Surface Warfare Center...
Next Article:Point-centered-quarter analysis of the upland forest at Yuhas Woods in Randolph County, Indiana.
Topics:


Related Articles
Stinking decorations protect nests.
Aircraft sales company begins operations in LR.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters