The Identity of Townsend’s Bunting

Mathew Louis

June 2, 2015

The name Emberiza townsendii was applied by Audubon to a single specimen collected by John Kirk Townsend in 1833 in New Garden, Pennsylvania. It is often referred as “Townsend’s Bunting,” and the details of the collection of this bird and its enigmatic status are well known.

A copy of Extinct Birds (Fuller, 2001; rev. ed.) in my library includes a color illustration of Audubon’s type (296) alongside a hypothetically-colored reconstruction by the artist (Julian Hume) of how that type may have appeared when first collected. While I had become familiar with Fuller’s book, and the illustration, since about the time of its publication, a notion occurred to me only recently which I feel is worth disseminating.

This is, very likely, a first-year or immature male Dickcissel x Lark Bunting hybrid (Spiza americana x Calamospiza melanocorys).

The illustration in Fuller is in copyright (but see www.extinction-website.com). Audubon figured the type (U.S. National Museum, #A10282) as plate 400, figure 4 in Birds of America [plate 157, 1841 edition.].

The most striking characteristic is the pale area along the folded wing, visible here in this, the decrepit specimen. The Lark Bunting’s tertiary and wing-coverts are white, and it is in Hume’s illustration that this character appears most obvious, whereas Audubon had not incorporated it in his portrait.

Comparing the above Townsend’s Bunting specimen to a male Lark Bunting specimen also conveys another peculiarity. Though the Lark Bunting’s undertail feathers are patterned with white panels on their outer extremities, the outer margins themselves are white. The image of the Townsend’s type likewise shows a margin to the outer right rectrix of a decidedly paler tone in contrast with the undertail itself. Audubon (Birds of America, 3: 62) had described the tail as “wood-brown, slightly edged with paler,…” This is quite different from the Dickcissel, which has a tail—an undertail—of a uniform brown without pale edges.

(Image source: www.miriameaglemon.com)

It is the Lark Bunting which, in immature or female plumages, has dark streaks along its crown (Byers, Sparrows and Buntings, pl. 17). As in our mystery bird, which is similarly marked (depictions of Audubon, Hume), the streaks appear as irregular flecks along the crown. The Dickcissel also bears a marked crown, but this is only true among immature or juvenile birds (Farrand, Western Birds (464)), and these marks tend to appear as evenly-arranged furrows. Though the streaks of the latter can also appear speckled if its crown is raised, there is a tendency for these rows of marks to be, nonetheless, nicely separated, and that aspect of this character also subtly distinguishes the Dickcissel from the other two. (See http://little-buffalo.com/tag/county/).

In the recent Extinct Birds (Hume & Walters, 2012, p. 346), the authors treat it as a hypothetical species combined in Spiza and provide these details of its description: “…eye-stripe, chin, throat, central line of underparts and edge of wing [emphasis mine] white; black-spotted line from lower corner of mandible down the side of throat, connecting with a crescent of streaks on upper edge of slate-blue breast.”

In ascribing the edge of wing as white, Hume and Walters are treating the type from a reconstructive perspective. Sharpe (Catalogue of the Birds in the British Museum, 12: 774 (1888)), likely working from American sources, described it in its actual state: “…the edge of the wing, and a gloss on the breast and the middle of belly, yellow.”

The Lark Bunting in immature or female plumages has discontinuous malar streaks, whereas in the Dickcissel the malar is a penciled line, with hardly an exception among individuals in the consistency of the marking. With respect to that character alone in the above image of the Townsend’s type, it may be an equivocal matter as to which species is represented by it.

The streaks on the breasts of the Townsend’s may actually be a correlating character of the Dickcissel, which has less streaking compared to the Lark Bunting, or limited streaking in younger birds.

Measurements [mm.] of Ridgway (Birds of North and Middle America, Part 1 (1901; 167, 171—175) on the Townsend’s and a series of the other two species also add to the argument.

Spiza townsendi [prevalent spelling of the Townsend’s]

Length: 146.05

Wing: 73.15

Tail: 53.59

Exposed culmen: 12.70

Depth of bill at base: 9.65

Tarsus: 20.07

Middle toe: 16.51

Ratio of exposed culmen to length—1: 11.50

Spiza americana

Length: 140.97—160.27 (148.08); female 139.70—145.80 (143.26)

Wing: 78.99—85.85

Tail: 55.12—61.47

Exposed culmen: 14.73—15.49 (14.99); female 12.70—14.22 (13.46)

Depth of bill at base: 10.41—11.43; female 9.91—10.67

Tarsus: 22.86—24.13

Middle toe: 16.76—18.03

Ratio of average female exposed culmen to average female length—1: 10.64

Calamospiza melanocorys

Length: 154.94—184.15 (163.32); female 144.78—165.10 (157.23)

Wing: 85.09—91.95

Tail: 65.53—71.12

Exposed culmen: 13.21—14.73 (13.97); female 12.70—13.21 (12.70)

Depth of bill at base: 10.67—12.19; female 10.16—11.94

Tarsus: 22.86—25.91

Middle toe: 16.76—18.03

Ratio of average female exposed culmen to average female length—1: 12.38

From this I can infer that the Townsend’s Bunting is appreciably larger than adult female Dickcissels, but as an immature or possibly adult individual (as the literature treats it as an adult), it averages much smaller than adult males of either species. For a (presumably) younger bird, it had already grown beyond the size of the largest female Dickcissels and in length is approaching that of the female Lark Bunting.

The most significant inference from the above is the comparative length of exposed culmen, the character where the Dickcissel averages greater than the Lark Bunting, as the former has a longer beak. The Townsend’s is on par with the smallest female Dickcissels, but this same figure is actually the mean of the female Lark Buntings. Even more compelling, culmen length is greatest in relation to body length in the Dickcissel, but the Townsend’s in this respect represents a significant departure from what would be expected if it was a pure Dickcissel, a figure midway between the two proposed parent species and even one approaching that of the Lark Bunting. [Female measurements are given as a comparative ratio on the basis of the small Townsend’s approaching females in size.]

Both the Lark Bunting and Dickcissel breed in the Great Plains region of the United States, the former a rare bird in Pennsylvania and the latter having been known to breed in the southern region of the state. Both are represented in rare records from Prince Edward Point National Wildlife Area (Prince Edward County, Ontario). From my own observations of the Lark Bunting, it is an irregular species. It will sometimes occur singly among flocks of other species (as I have observed one or small numbers among gambelii White-crowned Sparrows, Brewer’s Sparrows, or House Finches). The Dickcissel, too, is often an irruptive species and one which will turn up in single or small numbers among flocks of other species, notably House Sparrows. Given consideration of these arguments, it is plausible to argue that perhaps once in a great while the two may interbreed. Hybridization, despite the two species being in different respective subfamilies (Emberizinae, Cardinalinae) of the Emberizidae, these being treated in more recent works as separate families.

I am not aware if the literature has discussed hybridization between these species and assume no formal record has been published. Thus, I am arguing that each indiviual in question—Audubon’s type and the bird photographed in Ontario—represent legitimate hybrids and that further discussion on these two individuals should appropriate such a designation. Implying that there is but a singular instance of known hybridization of the Dickcissel, Denis Lepage (see link below for “Open Mic: The Townsend’s Bunting Story”) refers to a Dickcissel x Blue Grosbeak hybrid “with the name Spiza townsendii also attached to it.”

The discovery last year of a Dickcissel similar to the famed Townsend’s Bunting and having aberrant plumage characters deserves attention —

http://www.kyleblaney.com/photo-blog/

http://blog.aba.org/2014/09/open-mic-the-townsends-bunting-story.html

Reflection on this case gives me reason to present the arguments in this essay as more than speculation. This new bird not only adds to the discussion, it makes it very likely that it represents an instance—another—of hybridization. The superb photographs by Kyle Blaney appear in his blog, where he, and later Denis Lepage of the American Birding Association, argue for it being an aberrant color variant of the Dickcissel, and suggest the same is true for the Townsend’s. This argument I reject, and in a major irony, it is these very images which more than reinforce my own arguments, not only of this individual but of the original Townsend’s Bunting.

These photographs reveal a bird that is similar to the Townsend’s Bunting, but with obvious differences. No streaks on crown, a white mark below the eye, brown tertiaries only edged white, a continuous malar streak, gray of lower back the same color as the head. The tail appears to be of a uniform brown color, though the images do not reveal the acute detail of the edge of the underside.

These images also reveal a number of stunning aspects about the bird which show a definite correlation to Lark Bunting characters—

*While the tertiaries are brown in this bird, they are edged pale whitish, typical of immature Lark Buntings. To clarify, the Lark Bunting has white panels at the outer edge of the tertiaries which extend along the bend of the wing to the outer wing-coverts and alula feathers. The inner tertiaries are brown in young or female birds, edged paler.

*The wing-coverts of this individual are white with black chevrons in their centers—this an obvious character of immature Lark Buntings. The last of the images in sequence shows how this white area appears as panels on the outstretched wing.

*The beak is dark blackish above, and of a weak bluish tone on the lower mandible. This beak appears bicolored, and that tends to favor suggesting the Lark Bunting, as it is distinctively bicolored in all stages, each mandible a separate hue. While the shape of this bird seems lengthy and pointed like the Dickcissel’s, the immature Dickcissel has a more pale or flesh-colored beak and the adult’s is not truly bicolored—coloration is similar to the Lark Bunting but the blue almost always leads beyond the lower mandible to the base of the upper, with only the area at and approaching the culmen dark or blackish.

The lower eye is marked white with a faint whitish suffusion leading along the lores, and this is a character of the Dickcissel, along with the gray color of the back.

Reinforcing these arguments also is a belatedly published description of the Townsend’s Bunting type by John Kirk Townsend himself. The record of Townsend’s original notes were not published until 1909 (Auk, 26: 262—272). He carefully described something approaching a Lark Bunting beak: “[u]pper mandible black, middle edge white, lower light blue with a longitudinal line of black extending from the point half way to the base;…” He described it having “shoulders yellowish white,” a reference to the aforementioned panels.

Its song was compared to that of an Indigo Bunting Passerina cyanea (“a succession of lively notes somewhat resembling…but louder and more varied”), which is significant as the Dickcissel has a distinct harsh call from which its vernacular name is derived. A melodic bird, as conveyed by Townsend, would be an Indigo Bunting, but the characters of that species, as also with the Blue Grosbeak as Denis Lepage has argued, do not at all allow for a correlation to the characters of the mystery birds in question. So it is to the vocalizations of the Lark Bunting which consideration is afforded, and it is plausible to argue that Townsend may have been hearing the whistled phrases of one in song. Just years later, he would be first to describe the Lark Bunting as Fringilla bicolor, but in describing it (from flocks he and Thomas Nuttall watched) no recollection or comparison to the earlier specimen, particularly its vocalizations, was made. In his notes, he compared the earlier mystery in its plumage to the Gray-headed Sparrow Passer griseus, a species which was originally and erroneously believed to have been first collected in the United States.

Another character deserving note, if only that, are the stipples below the eye in Julian Hume’s reconstructive illustration of the type. Though Audubon’s portrait does not show this, he having it fresh, it is another trait common in the Lark Bunting. The specimen may be too worn to be able to examine this point further.

I feel that a hybrid Dickcissel x Lark Bunting is a better argument than that of a simple aberrantly-plumaged individual for both cases. Each bird represents, in its plumage characters, examples of both species, and for the sake of my arguments, it is the Lark Bunting which is adequately represented as a parent and most likely the one species from which crosses such as these would originate. Despite the assertions made, a Blue Grosbeak hybrid is unlikely as that species does not adequately betray characters in these mystery individuals, and an aberrantly-plumaged individual is unlikely as this is a type of bird with remarkable proportions unlike those of a pure Dickcissel. It is possible that the Ontario bird may have different proportions (or even be an adult), but in regards to an actual identity of what these individuals are, I feel that what holds in the identity in the original Townsend’s Bunting holds for this more recent example, one and the same. A hybrid argument is also reinforced in that there is considerable variation between both individuals in question. A DNA analysis of the Townsend’s Bunting type would ultimately rectify this mystery, and this essay presupposes a course of further action.

"The Sage Sparrow, like the Lark Bunting and other related sparrows, usually reveals a panel of white wing-coverts with dark chevrons in the center of each. Comparison from this mystery bird to the Sage seems plausible from that, but it is important to note the distinction between the panels of the Sage and Lark Buntings, which also reduces the question of the former representing a hybrid parent. In Blaney's bird, these markings are unequivocally black, as night/day against the white areas--exactly like that of the Lark Bunting; whereas, in the Sage Sparrow, the dark markings on its wing-coverts are never as black, but a dark grayish-brown or blackish-brown and somewhat dull. There is proportionately a much lesser amount of white."

Having run search queries on the net of “Emberiza townsendi” in conjunction with the following terms: hybrid, Calamospiza, “Fringilla bicolor,” I was unable to find precedent on the Lark Bunting being associated with the discussion on this form.

(Also see Audubon Society Master Guide to Birding, 3: 214—215, 246 and Threatened Birds (ICBP, 2000).