THE :

A QUARTERLY JOURNAL OF

ORNITHOLOGY.

Vo.. xxxx. JunY, 1912. No. 3.

BIRD GENEALOGY. •

BY CHARLES W. TOWNSEND, M.D.

ARCHAgOPTERYXhad teeth in its jaws, separatehip bones,bi- concavevertebrae, claws on its front limbs and a vertebrated tail, all marks of the , in which group it might still be.placed by some, were it not for the fact that the impressionof the feathers has been preservedto us and stampsits essential nature. The links betweenbirds and their reptilian predecessors are very perfect. Now if birdsare descendedfrom ,one may perhapsstill find sometraces of this lowly originin the infantileperiod of bird life, just as thereare variousear-marks of the savageof the jungle in the infancy of the most gilded city dweller,not to mentionthe transient and permanentreversions often found amongadults of the race. Thus the Hoatzin of the Orinocowhen young, has clawson the wingsand scramblesabout the branchesin a truly reptilianstyle, a mode of progressionthat, accordingto Beebe,is still usedby the adults. One need not go sofar as the Orinoco,however, to find evidences of the quadrupedalreptilian mode of progressionin , as witnessthe actionsof youngHerons before they learn to fly, when with wingsand legsthey climb about their family tree almost as gracefully,I dare say, as did someof the ancientwinged reptiles.

Read before the Nuttall Ornithological Club, March 4, 1912.

285 286 Towr•sEN•),Bird Genealogy. [Auk[July

The extensionof the so-calledthumb or bastardwing in the Pigeon and other birdsas they approachtheir perchmay in the sameway hark back to the time when the reptilian ancestorgrasped with its fore feet its goalon the tree tops. Both youngGreen and Night Heronselevate the bastardwing at ti•nesas they climb about the trees,but I have never seenthem attempt to useit for grasping. Left with a coupleof young English Sparrowson my hands owing to the destructionof their nest causedby the closingof a blind, I droppedthem into a basin of tepid water, expectingthe inert massesto sink or at least that their wabbly headswould fall below the surface. On the contrary they becameendowed with life and vigor as if upon their native heath, and, with a combination of rapid strokesand leg action,and with necksoutstretched they scuddedacross the surfaceof the miniaturepond. Blood will out, the reptilian ancestrywas working! To make surethat this was not an accidentI droppeda youngRed-winged Blackbird into the pool below his nest. He too performedin ex- actly the samemanner and safelyreached some reeds up which he scrambled,and wasthere well taken careof by hisexcited parents. It is probablethat many a passerincbird nestingover the water hasbeen saved from destructionby thisreturn to primitivemethods. Further experimentationwith youngCrows and BronzedGrack- les showedme that very youngbirds generally moved the alternately,while older onesalways flapped both togetheras in flight. From this one would infer that the primitive reptilian scramblewas naturally an alternate methodwhile the simultaneous methodwas simply the moreadvanced style used in flight. In the easeof the Chimney Swift the method of flight has been thought to be an alternate flapping of the wings. Let anyone watch thesecurious birds as they dart with amazingspeed through the air, and I am sure he will agreethat the wingsare usedalter- nately with great rapidity. Steady flight by this method is, I believe,mechanically possible. Onemight argue therefore that the Swifts retain the more primitive or reptilian methodof moving the front limbs and are thereforemembers of a very early branch on the arian tree, althoughit is possiblethat through extreme speeiali'zationthey have returned to this form of flight. If this pronemethod of propulsionon the water on all foursis a VoL 1912XXIX] J TOWNSEND,BirdGenealogy. 287 primitive one, as indeed it must be, then birds that swim in an erect -like manner must have advanced beyond this stage and becomespecialized. [ have se;•eraltimes seenyoung Spotted Sandpipersthat were unable to fly, swim with ease like little , although when very young and much frightened they return to the prinfitive reptilian scramble on all fours. All of the membersof the ShoreBird family swi•nnaturally if they find themselvesin water beyondtheir depths. Phalaropeshabitually disport themselveson the surface of the water as gracefully as miniature . It would seem to be a natural inference, there- fore, that the ancestors of Shore Birds were swimmers and that the art of swimmingis inherited and not developedby this group, and that the Phalarope is a case of reversion. The awkward actionof a youngseal at its first bath is an exampleof a easewhere the art of swimminghas been recently acquiredby a group and whereit is not one of long inheritance. In the classificationof birds proposedby Hans Gadowthe order Charadriiformes,or Plover-like birds, includesthe Shore Birds, , and Pigeons. The Shore Birds, we have just seen, showevidence of a swimmingancestry, although, with the excep- tion of the Phalaropes,they habitually preferthe shoreunder their feet, even if it is wet and partly coveredwith water, to the deep sea. The presenceof partial webs, as in the Willet and in the SemipalmatedPlover and ,point to the former existence of the swimminghabit, for these birds like other ShoreBirds do not swim exceptwhen unexpectedlyforced to it. If the partial web in the foot of the adult ShoreBird showedthe beginningof the swimminghabit in birds of land ancestry we shouldfind the young birds like young sealsvery inexpert in the water.. As the reverse of this is the ease our conclusion that these birds are of water ancestry must be correct. Gulls and have fully webbedfeet but their habits at the presentday hardly justify them in this possession.Webbed feet are of great advantageto the rapidly swimmingbird and to the diving bird that dependson its feet. Now Terns rarely rest on the water or swim and Gulls do not often swim rapidly, in fact they rarely swim at all, but drift about, while if either bird descends belowthe surface,it is as a result of the velocityof its plungefrom 288 TOWNSl•ND,BirdGenealogy. [July['Auk the air, and its feet are probably not used. In fact the web, althoughuseful, is largely wastedon thesebirds, and it is evident that it is ancientand pointsto a swimmingancestry. That this ancestryis lessremote than in the ShoreBirds is perhapsshown by the fact that a wing-tippedGull, fallingon the beachwill take to the water,and swim vigorously out to sea,while a similarlycrippled ShoreBird falling into the water will swimto the beachand endea- vor to run inland to hide. Before they are able to fly, young Skimmersare said to seek safety by running into the water, another evidenceof their water ancestry. Chapmanin his "Campsand Cruisesof an Ornitholo- gist," speakingof youngCommon Terns a few days old, says: "severalwere seen to enteran inflowingcreek, drink repeatedly of the salt water and swimactively, in evidentenjoyment of their natatorialpowers, while the parents,who rarely alighton the water, watchedthem from the shore. Possiblyhere was an explanation of the value to Terns of webbed toes. Functionless in the adult they are of serviceto the young,before the powerof flight is ac- quire&" In this suppositionhe is probablyright, althoughthis serviceto the youngis not the reasonfor the existenceof the webs, but the observationpoints very dearly to the swimmingancestry of the birds. We couldnot have strongerproof of it. That the Auks are out and out water birdsthere needsno defence, but one is at first sight puzzledby the presenceof the Pigeonsin this group. The oldersystematists placed the Pigeonswith the Partridgeand DomesticFowl tribe, but Pigeonsmay be seen wadingin puddlesin a mannerthat wouldalarm the Barnyard Cock. I have beentold by a Pigeonfancier that youngPigeons are much attracted by water, and fond of bathing therein, and that youngbirds are liable to drownthemselves in tanksor troughs if theseare accessibleto Pigeonlofts. I recentlyplaced a half- grownDomestic Pigeon in a wash-tubof tepidwater. With head and neckerect, the bird swamrapidly with alternatestrokes of the feet to the sideof the tub. The wingswere arched up and waved slightly,--not stretchedout and flappedin the water as in the case of young Passerinebirds. Its position was like that of a Duck but low in the water. Progresswas much more rapid than on land wherethe bird stumbledawkwardly along. Indeedit had Vol. 1912XXIX'] J TOWNSEND,Bird Genealogy. 289 never before left the nest. ! repeated the experiment several times with the same result. A fact of considerable interest in this connectionis that "A Pigeonwith a perfectlywebbed foot [was] evolvedat Cambridgeby only three years' selectedcrossings." • This may be lookedupon as a caseof reversion. The throwing of somersaultsin the air similar to thoseof the Tumbler Pigeon has beenreported in the caseof the Black-belliedPlover. The Sheathbill, Chionis,is so ancestraland generalizedin its type that it suggestsall the groupswe have just been considering. Anatomicallyit is allied to the Oyster-catchersand the Gulls. It is often classedamong the Plovers,but it is as marinein its haunts as are the Auks,and in flight it resemblesthe Gulls. Its appearance on land, and mannerof courtingare very muchlike thoseof a Pigeon,and it goesby the nameof ' Kelp Pigeon.' While youngTerns take to the water, youngCormorants when pursuedtake to the shore. This would suggesta terrestrial an- cestryof thesebirds, and accordingto Gadow, strik- ingly resemblethe New World Vultures, and the habit of both these birds of sitting with their wings spread is suggestiveof kinship. The fact that Cormorantson risinginto the air hop with the feet together,although their usual gait is a waddle, suggests a former arboreallife, and many Cormorantsstill nest in trees. Tree dwellersnaturally hop from branch to branch, and it is probable that the earliest birds were arboreal. When the tree- dwellingbird descendsto the ground it naturally hopsthere also, but hopping is not a satisfactorymethod of progressionfor a ground-feeder,-- it doesnot per,nit of cautiousapproach, and it is decidedlyjarring. A walking gait, therefore,may be understood to indicate a long custom of feeding or dwelling on the ground. Although the Flicker is frequently seenon the ground,the ground habit is probably but recently acquired,for it has not learned to walk, while the Robin for exampleis able to run, and doesso much more often than it hops. Young Robins show, however, their arborealancestry by hoppingmore than they run. Pipits,Horned Larks and Ipswich Sparrowshave so completely departed from arborealhabits, that they run easilyand walk with grace. Walking

• T. Digby Pigoit, "London Birds and other Sketches". London. 1902, p. 239. 290 TowNsram),BirdGenealogy. [July,Auk appearsto be acquiredlater than running. It is a very interesting fact that the Savannah Sparrow, frequenter of meadows and marshy pastures,generally hops even when on smooth ground, althoughit is alsoa goodrunner, while its nearrelation the Ipswich Sparrow,frequenter of sandy wastes,almost never hops and is a goodwalker. as far as I know, althoughconstantly in the water very rarely swim, but that they come of a swimmingancestry seems probablefrom the behaviorof a youngGreen not old enough to fly that I put in the water. It sat erecton the surfaceand swam off with a graceand easethat contrastedforcibly with its awkward movementson land. Not only was its poisegraceful and - like, but the speedwith which it swam, the practicedmanner in which it featheredits ungainlytoes, the easewith which it threaded its way amongthe grassstalks, and dabbedevery now and then at the water with its bill, all pointed to an inherited instinct, an instinct, however,that is largely if not entirely lost in adult life. This young Heron had never practicedthe art of swimmingbefore --it had probably never left the nesting tree, which was on a marsh island somedistance from even the highesttides. Adult Heronslike someShore Birds showtheir swimmingancestry by a distinct web between the middle and outer toes. The use of the wingsunder water in somediving birds and the significanceof this fact I have alreadydiscussed in anotherplace2 One is apt to think of evolution as a thing of the past, an ac- complishedfact, and to forget that at the present period of time this great law is still as existent as it has been since the world began. With changein environment,there comes through natural selectionacting on slight variations and occasionalmutations a change in the structure to fit the new environment,and in time a new speciesis developed. As new speciesarose in the past, so they must be in variousstages of formation at the presenttime. The greatgroup of AmericanWarblers are for the mostpart slender- billed, insect-eatingbirds, that go southwith the approachof cold weather. One of them, however,is enabled to spend the winter on the bleak New England coast by a changefrom an insectiv-

Auk, XX¾I, 1909, pp. 234 to 248. Vol. •o•XXIX] J TO•VNSgND,Bird Genealogy. 291 orousto a seed-eatinghabit. The Myrtle Warbler thrivesthrough the cold winters chiefly on a diet of bayberries,while all the other membersof this family seek more genial climes,where they may continue to live on insects. Not only this, but a large number of its own speciesgo south, and winter in the Greater Antilles, Mexico and Panama, where insect food is of course abundant. The New Englandbirds eat not only bayberries,but alsothe seeds of grassand weedsthat extend above the snow, and they glean the bark of trees like Titmice. Now birds like men are clannish; in fact there is a remarkable similarity betweenanimal and human nature,--which is not so surprisingwhen one considers our origin and relationships. Among savagesslight differencesdue to different environment,set apart one group or race from another. Each race considersitself the people,and despises,fights and refusesto mix with the other. The Eskimo and the Indian, although both manifestly of Eastern origin, so dislike each other that intermarriage,except under the influenceof civilization, is rare. This tendency makes of course for differentiation; without this tendency the constant mixture of raceswould make the productionof new speciesmore difficult. While this clannishnessis most marked amongsavages, it is also so pronouncedin civilizedraces that eachnation classesall foreign- ers, especiallythose that speaka different tongue,as their inferiors with whom intermarriage is not to be thought of. The more ignorant the individuals, that is to say the more primitive or animal-like, the more intense is this clannishness,and, its boun- dariesmay be limited,not by the nationor state,but evenby the village in which the individuals live. Mr. Punch's collier who proposed' 'caving'alfa brick' at the strangerin town is an in- stancein point. The element of home also enters into this exclusiveness which favors the formation of races, and hence of new species. This factor is stronglyshown in the humanspecies unless the individual has becomecosmopolitan by travel and education; and the in~ habitants of what appears to an outsider to be a most desolate regionregard their homeas superiorto any other countryon the globe,and pine if taken away from it. Now the seed-eatingMyrtle Warbler that spendsits winters in the cold and stormy north is undoubtedly as clannish as the 292 Tow•sr,N•), Bird Genealogy. [JulyAuk

Eskimo,and considersitself superiorto the south-seekingMyrtle Warbler, and it would probably pine for its northern home if transplantedby forcewith the restof the speciesto tropicalregions. In addition,its elannishnessprobably impels it to chosea summer home apart from its southernrelations. At present man cannot distinguishthe northern from the southernMyrtle Warbler,just as in the remotepast, it is probable that the Eskimo could not be distinguishedfrom the Indian. In time, however, aided by this inherent elannishnessand love of home, one might predict that a larger race of northern Myrtle Warblers would be formed with thicker, strongerbills and more musculargizzards. Indeed I have endeavoredto investigatethese three points in order to discoverwhether a beginninghad been made in the evolutionof this new species,but I have not as yet examinedenough material to throwany light on the subject. One can easilysee how important the elementof elannishnessis, for withoutit interbreedingmight for a longtime, if not indefinitely delay the birth of a new species. The importanceof elannishness in the evolutionof racesand species,has I believenever beengiven due weight. As among men so among birds there are striking differencesin ambition and ability to succeed. Somemen, somefamilies, some nations are progressive,--they are always reachingout for new opportunitiesand taking advantageof them. Othersare retiring, unambitiousand contentedto remainwhere they are. One of the most remarkablyprogressive birds is the Horned Lark which has spread to nearly every part of the continent,and has made each part somuch its homethat it hasadapted itself to the environment to the extent of changingits own form and plumage. There are now recognizedfourteen different North American races or sub- speciesof the Horned Lark. The pushingcharacter of the bird is shown in the recent extensionof the breedingrange of the Prairie HornedLark from the centralpart of the continentto New England. In 1889 it was first recordedas breedingin Vermont, and the same year in central Massachusetts. In 1903 it reached the seaand bred at Ipswichand has comethere to raiseits young ever since,meanwhile increasing in numbersthroughout the New England states.•

Auk, XXI, 1904, p. 81. Vol. 1912XXIX] .1 TOWNSEND,Bird Genealogy. 293

The Song Sparrow has adapted itself in twenty different forms to all parts of the continent,and is abundantalmost everywhere. Incidentally it is interestingto comparea map of North America showingthe variouslingual races of Indianswith oneshowing the various races of Song Sparrows. Both maps show an extensive race in the more uniform east- the AlgonquinIndians, and the melodiasparrow,--while both showin the diversifiedsurface of the extreme West numerous races of both man and bird. What a contrast is the enterpriseshown by the SongSparrow to the lack of enterprisein the caseof sucha bird as the Swamp Sparrow,for instance. Althoughfirst cousinto the SongSparrow and althoughit is spreadover a largeterritory, the SwampSparrow limits itself to the almost uniform environmentof swamps,and has thereforenever developedany races. Anotherbird whichis showinggreat developmentalor evolution- ary possibilitiesis the Grackle both Purple and Bronzed. This bird insteadof shunningman has beenbright enoughto appreciate the fact that it is safestfrom persecutionwhen in most intimate relations with him. It has come into his towns and cities, and it does not hesitate to build its nests on his houses. In Boston, although there had been a few previousrecords, it was not until 1900 that the Bronzed Grackle began to breed regularly in the Public Garden,and the numbersincreased so that thirty-two nests werecounted there by Mr. H. W. Wright in 1906. In 1907 they first beganto build nestsin the vines on my Ipswich house,and two pair have nestedthere every summersince, when I permitted. In the matter of food they are not particular, or rather their appetite is a catholicone, and they can adapt themselvesto cir- cumstances. They are able to pick eggsout of a Robin'snest and peasfrom podsin the garden,and they undoubtedlyserve a useful purposein towns and cities by diminishingthe English Sparrow nuisance. I have seenone hold downa strugglingEnglish Sparrow with its foot while it deliberatelypeeked out its brains. While the English Sparrowsfollow Robins hunting wormson the lawn, and saucilysnatch the worm away from their very mouths,they keep at a safe distancefrom the Grackle, and if he so much as stopsto look at them, they fly off in terror. In fact Gracklesput to flight the innocentRobins. I have seena Gracklepartly run 294 Tow•sEm),Bird Genealogy. [.July[Auk and partly hop with wings extended toward a Robin that was diggingworms near by, making the Robin desertthe spot on which the Grackle then dug. But the most interestingdevelopment of the Grackle, one that showsits great adaptability and intelligence,is a habit it has of pickingup food from the water, after the mannerof the Herring . A Grackle will hover closeto the water its head to the wind, and then suddenlydrop, and with its bill pick up from the surface somemorsel as gracefullyas a Gull. This they do at times with- out wetting their plumage; at other times the bill, feet and tail are immersed,while I once saw a Grackle splashhis whole body into the water and entirely immersehis head, to emergewithout difficulty,carrying in his bill what appearedto be a small silvery fish. I have seenthem after sailing and hovering over the water in a high wind with the spraydashing about them, skilfullypick up food from the topsof the waves. It is easyto picture an islandcommunity of Gracklesbecoming more and more addicted to a maritime life, owing perhapsto the shrinkingof their terrestrialfood supplydue to a changeof elimate or to land subsidence. Would not these habits become in time as much inherited as are similar habits in the Gulls? Or, to put the questionin another way, were not the inherited traits of the Gulls originally acquired? The IpswichSparrow is the only strictlydune dweller among the birds. Its summer home is on Sable Island, an island of sand dunes off Nova Scotia,and it spendsits wintersalong the sandyportions of the Atlantic coast. It is evidently a near relation of the Savan- nah Sparrow, which is somewhatsmaller and darker, and lives chieflyin marshesand openfields from Labradorto New Jersey. As the glaciersreceded we can picturethe gradualpushing north of the SavannahSparrows, and their extensionto the great sandy wastesthat fringedthe coastfor miles. As the land sank and the waters rose restrictingthese regions of sand, the strugglefor life amongthe elan that preferredthe sanddunes must have beenan intenseone, and it is probablethat the larger and strongerbirds, as well as thosethat more nearly matchedin color their surround- ings were the more likely to survive. Isolation from the main land finally aidedin the work, and at last a distinctlynew species ¾ol. 1912XXIX] J PHILLIPS,The American Black Ducks. 295 was evolved,a bird larger than the SavannahSparrow of the main land, and of a gray or sandy, rather than a black and brown color, so that when it squatted in terror on the sand the sailing Hawk was more apt to passit by. It seemsto me, therefore, that the evolution of the Ipswich Sparrowis comparativelyrecent, and that the age of this species may be countedby the paltry fifty thousandyears or so that have elapsedsince the last glacial period.

A RECONSIDERATION OF THE AMERICAN BLACK DUCKS WITH SPECIAL REFERENCE TO CERTAIN VARIATIONS.

BY JOHN C. PHILLIPS.

THERE are severalspecies of primitive ducks which for many reasonsare of peculiar interest, on account of their remarkable geographical distribution and mutual interrelationship. This group of speciesis composedof AnasfuIvigula, Anas tristis,Anas diazi, A•as wyviIliana and A•as Iaysancnsis. Most of these are poorly representedin collectionsand this fact has led to certain ntisconceptions.It is the purposeof the followingnotes to point out someof thesentistakes, and to say a few wordsabout individual and sexual variation. To begin with I wish to call attention to the principal difficulty in the properunderstanding of theselocal races; this is the presence of a sexual differencein plumage,increasing probably with age, and comparable,with that of the Hawaiianduck (Anas wyvilliar•a). In this way all these related speciescan be separatedfrom A. tristis in which the sexesare similar. At first I thought that this sex difference was confined to A. diazi and I started to limit these notes to the latter species,but as more specimensturned up I thought it better to considerall the AmericanBlack Ducks.