SHORT

COMMUNICATIONS

A Technique for SalvagingAnatomical Material from Study Skins of Rare or Extinct

Birds

STORRSL. OLSON,J. PHILLIPANGLE,FREDERICK V. GRADY,AND HELEN F. JAMES

NationalMuseumof NaturalHistory,Smithsonian Institution, Washington, D.C. 20560USA Nearly one-third of all speciesof birds are still representedin collectionsonly by study ski•, with no skeletalor fluid-preservedmaterialavailable(Zusi et al. 1982). The development of alternative techniques for preservation of other organ systemssimultaneously with the integument (e.g. Norris 1961, Johnsonet al. 1984) arrived too late for the species that are now extinctand for which only studyskins exist. The lack of skeletal

material

of rare or extinct

to skin the wing to retrieve the radius, ulna, and carpometacarpus,although if in poor condition the skin maybe damagedwhen the remigesare separated from the underlying bone. The extractedwing bones are replaced with wire. The tibia and tarsometatarsus,even in small passetines,can be removed through a longitudinal slit in the podothecaand subsequentdissectionof the bones,leaving the toes in place. This leaves the po-

spedescan be a nearly i•urmountable impediment to systematicand paleontologicalinvestigations. Fortunately,mostof the skull wing, and leg bones remain in the standardmuseum study skin, so skin collectio• are potential osteologicalrepositories of speciesfor which skeletons can no longer be obtained. We have developed a technique for the recoveryof the bonesand attachedmusclesfrom study skinsthat preservesthe appearanceand scientificusefulness

of the skin.

When seriesare available, it is of primary concern to select the specimen in which the bones are best preserved.This will vary accordingto the preparation technique of individual collectors.The skin may be felt to determine

where

the back of the skull has been

cut or whether the mandibuiar articulation appears to be intact. Ideally, the specimenis x-rayed to determine exactly how much bone remains. There is a natural tendencyto selecta skin that appearspoorly made, but this may be false economy becausethe betterthe initial appearance,the betterthe final result will be (Fig. 1). Once a specimen has been selected,detailed measurementsshouldbe madeand recordedon a separate label, along with the date of repreparation and name of the preparator. The skin is then relaxed in a closed container above a layer of wet sand that has been laced with phenol (carbolic acid) to prevent mold. When the skin is supple,it is openedand the stuffed body removed. In most casesthe skin will be too delicateto sustainthe inversion of the skull through the neck, as in the usual skinning procedure,so incisionsare made along the inside of the mandibular rami to produce a "ular flap" through which the skull is extracted.The skin is then teasedaway from the fore part of the skull and cut away from the base of the bill so the entire

^ I

skull and bill can be removed.

At this stagethe remainingportionsof the humeri can be extractedeasily. Removal of the wing bones is facilitated by a longitudinal cut under the wing near the junction with the body. This is concealedby the wing in the finished specimen.It is also possible 510

Fig. 1. Harvard specimen of Ciridopsanna (Drepanidini) before (A) and after (B) repreparation.The bonesfrom this specimen(C) were in excellent condition and perfectly suitablefor comparativepurposes. Only five specimensof this extinct speciesexist.

July1987]

ShortCommunications

511

A

1

Fig.3. Sketchshowingthe degreeof preservation of jaw musclesin a drxed skull removed from a skin of the extinct Hawaiian endemic Loxopssagittirostris (Drepanidini).

into the castand a new body constructedaround it. When the castand body are replacedin the skin, the edgesof the skin aroundthe baseof the bill are glued to the castusing25%polyvinylacetate(PVA-AYAF or PVA-AYAF/AYAT)

dissolved in acetone. Adhesive

compoundscontaining cellulose nitrate, which can breakdown to producenitric acid,shouldnot be used. The skin is then sewn closedand the bill painted to simulatethe original color. The resulting specimen is as goodas the original (Fig. 1) for almostany purposeusing traditional study skins. Once the skull hasbeen cast,the jaw muscles,which may be well preserved(Fig. 3), should be dissected and the mandible C

D

E

Fig. 2. Cast(A) and two-piecemold (B) of the skull of Drq•i$ p•½i#½•(Drepanidiz•t).Cast(C) of the skull of P$•ttirostr•p•lm•r/ (Drepaz•dini) and subsequent casts(D) of the internal structureof the ramphotheca with an exampleof a one-piecemold (E). Bothof these speciesare extinct Hawaiian endemics.

the skull

epoxy). The cranial part of the finished cast should be hollowed out to reduceweight. A wire is inserted

from the skull. In the case

and mandible

should

be molded

and cast

again (Fig. 2) before the rhamphothecais removed and the skull cleanedcompletely. The scientific value of specimensof extinct birds treatedin the abovemanner is increasedgreatly by the amount of new information

available, whereas

practically no information is lost in the processother than minute

dothecaessentiallyintact, and when the bonesare replacedwith wire the substitutionis scarcelyperceptible.The skin, turned right side out and lightly filledwith cotton,isreturnedto therelaxingchamber while theskullismoldedandcast,followingthe basic proceduresfor vertebratefossils(Watersand Savage 1971).The processrequiresat least 2 days. For finchesand most other birds with relatively shortbills,a simpleone-piecemoldof the skull (Fig. 2) is made with silicone rubber (e.g. GE RTV 700 siliconewith Beta 1 curing agent). For specieswith long, decurved bills, a two-piece mold may be required (Fig. 2). After the mold has set, the skull is removedand a castmade with epoxyresin or similar plastic substances(Re Epoxy 103 and Tapox "4-1"

removed

of speciessuch as finches, which may have characteristic patterns of the horny palate (Sushkin 1924),

details of the external

nares and the area

of attachmentof the skin to the bill. We encourage museumcuratorsto permit irreplaceablespecimens in their care to be used to the maximum possible advantagewhenever circumstances dictate. We are especiallyindebted to the enlightened curators of the American Museum of Natural History, the Museum of ComparativeZoology,and the B. P. BishopMuseum, whose generousresponseallowed us to developthis technique.We are alsograteful to the staffof preparatorsin the vertebratepaleontology laboratoryof the Departmentof Paleobiology,SmithsonJanInstitution, especiallyArnold Lewisand Leroy Glenn,for assistance in castingandmolding.We thank CatherineA. Hawks for reading the manuscriptand suggestingimprovements in some of the materials used. Richard L. Zusi kindly altowed us to use one of his sketchesof jaw musclesto show the potential

512

ShortCommunications

of skin specimensfor myologicalstudies.The photographs are by Victor E. Krantz.

[Auk,Vol. 104

SUSHKIN,P. P. 1924. [Morphological studies of the Fringillidae and allied groups.] Bull. Brit. Ornithol. Club 45: 36-39.

WATERS, B. t., & D. E. SAVAGE.1971. Making duplicatesof small vertebratefossilsfor teaching and

LITERATURE CITED

JOHNSON,N. K., R. M. ZINK, G. F. BARROWCLOUGH, &

J. A. MARTEN. 1984. Suggestedtechniques for modern avian systematics.Wilson Bull. 96: 543560.

NORRIS,R.A.

for research collections.

Curator

14: 123-132.

Zu$i, R. L., D. S. WOOD, & M. A. JENKINSON. 1982.

Remarkson a world-wide inventory of avian anatomical specimens.Auk 99: 740-757.

1961. A new method of preserving

bird specimens.Auk 78: 436-440.

Received 26 September 1986,accepted 29 December 1986.

Brown-headed Cowbirds Learn Flight Whistles after the Juvenile Period STEPHEN I. ROTHSTEIN AND ROBERT C. FLEISCHER•

Departmentof Biological Sciences andMarine Science Institute,Universityof California,

SantaBarbara, California 93106USA' The ontogeny of songbirdvocalizationshasbecome

the ontogenyof the vocalizationknown as the cow-

a classicexampleof the interactionof genetic and

bird's song, learning plays a major role in altering songstructure(Westet al. 1981,King and West 1983).

environmental factorsin the development of complex behavior (Baptistaand Petrinovich 1984). In the best-

known study, Marlet and Tamura(1964) usedtaped playbacksof song as "tutors" and concluded that White-crownedSparrows(Zonotrichia leucophrys) learn only conspecificsongand will do soonly if they hear it between about 10-50 days of age. These findings do not apply to all songbirdsas many specieslearn new songsafter 50 days of age (Kroodsma1982). Moreover, by exposingbirds to live tutors, Baptista and Petrinovich (1984, 1986) found that White-crowns

will learn songs,evenof heterospecifics, after 50 days of age (but see Cunningham and Baker 1983, Baker and Cunningham 1985).Besidesits importancein basic ontogeny, vocal development in songbirdsis critical to understanding possible evolutionary consequencesof dialects,about which there hasbeen much

recentcontroversy(Kroodsmaet al. 1984,Bakerand Cunningham1985and accompanyingcritiques). We report on the flight whistles (hereafter FWs) developed by captive Brown-headed Cowbirds (Molothrusater) exposedto live, rather than taped, tutors.

The FW and song are given only by males,and both function in agonisticmale-male and sexualmalefemalecommunication(Rothsteinet al. in press).Critical differences exist between the two vocalizations,

however. Most FWs are given in flight. Becausevirtually all songsare given while malesare perchedor are standingon the ground (Friedmann 1929),we call this vocalizationperchedsong(hereafter PS). PSsalways begin with brief notes below 3 kHz and then rise rapidly to at least7 kHz (West et al. 1981, Dufty

1985).By contrast,FWs are mostly pure-tone vocalizations between 4 and 10 kHz (Rothstein and Flei-

scher 1987, Rothsteinet al. in press).Although the PS varies between two cowbird subspecies(King et al. 1980), localized dialects with discreteborders appear to be absent,at least in New York state (Dufty 1985). But discretedialectsoccur in the FW in parts of California (Rothstein et al. 1986, Rothstein and Fleischer 1987). Males have repertoires of 2-6 PSs (Dufty 1985,pers. obs.)but most have only one FW type, except that males with 2 FWs are common at

The behavioralontogenyof a brood-parasiticspecies borders between FW dialects (Rothstein and Fleischer suchas the cowbird is especiallyinteresting because 1987). Becausethe PS and FW have different acoustic the birds have no known contactwith their parents. This has led some (e.g. Mayr 1974) to suggestthat cowbirds have a closed developmentalprogram re-

structuresand patterns of variation, work on the on-

sistant to environmental

ontogeny of the FW.

influences

to ensure that vi-

tal species-specific behavior developsproperly. Although there are geneticallyprogrammedaspectsto

togenyof the former(Westet al. 1981,King and West 1983) cannot be used to make conclusions about the All birds cited herein

were

housed

in 3 outdoor

aviaries (A, B, C) at the University of California at SantaBarbara.The aviarieshad 3 adjoining and parallel cages,each measuring 5.3 x 1.0 x 2.9 m. The 5 • Presentaddress:Hawaiian EvolutionaryBiology subjectsof this study were divided into 2 groups. Program,1993East-WestRoad,University of Hawaii, Males 1-4 were captured on the east slope of the Honolulu, Hawaii 96822 USA.

Sierra Nevada at Mammoth Lakes, Mono Co., Call-