256 •A•, TheConnecticut Valley. [AAp•rril

2. An Adriatic-Tunisian route, along the coastsof the Adri- atic, Sicily to Tunis. ByLaridae (Larusridibundus); Scolopacidae; Gruidae;many small songbirds. 3. An Italian-Spanish route from Austria-Hungary via North Italy, Po Valley to Corsica,Sardinia, Baleuric Islands to South France and Spain:By Larldae; Charadriidae;Scolopacidae, etc. From 1903-1919, 7,778 birds were banded at Rossitten and, besides 123,569 bands were distributed to outsiders. Of these 7,778 banded birds, 2,011 have been accountedfor by reports received at the station. Observationsnow extending over 19 years have provedconclusively, that bandingis not injuriousto the birds and that it doesnot disturb or changetheir habits. All attacks made upon scienceby over-conscientiousbird protectors have been disprovenas based either upon misapprehensionor upon wilful misrepresentation. Berlin, Germany

THE VALLEY--A HIG/-15VAY FOR BIRD MIGRATION. 1

BY AARON C. BAGG.

Plates. XVI-XVII TrtAILSare perpetuallyfascinating. From the dawn of history the wanderingsof men and later, certain of their trade routes continuallyattract our attention. Justhow a groupbecame estab- lished in a given locality; by what route various hordespoured over a mountain-barrieror acrossa body of water; why another tribe did not remainsettled or stationarybut persistedin continual travel--all theseare problemswhich the studentof history likes to puzzleover. Soin the studyof ornithologythe routestraversed by birds of passageequally charm the studentor the layman. Long beforewhite men bravedthe mad Atlantic to explorea new world or the warriors of the Six Nations established the now

• Paper read before the Allen Bird Club of Springfield, Mass., Feb. 6, 1922. My grateful acknowledgments are due Mr. J. A. Farley for his painstaking re- vision of this paper and for a number of contributions in data. A.C.B. Vol.1923 XI,']J BAGG,The Connecticut Valtey. 257 famousMohawk Trail into the ConnecticutValley; yes, prior to the time that the Abenaquisfirst descendedthe Valley from Mem- phremagogand the headwatersof the river to its mouthfor their winter'ssupply of pe•nmican,this Valley, like others,was used for centuries,if not ages,by the birds in their annual springinvasion and autumnexodus. The origin and causeof this habit of regular advanceand retreat was alwaysin the past an unsolvedmystery. Consensusof opiniontoday, however,points to the Great Ice- Sheet,or better to the declineof the age of Ice, as being largely instrmnentalin bringingabout this remarkableand regularphe- no•nenon. Bird migration is now authoritatively consideredto have begunby only short changesof location,corresponding to the first slight recessionsof the ice cap from which must have co•ne benefit in some form. So migration became gradually a habit; and more gradually still, the range lengthenedwith the shrinkageof the ice-sheetuntil the present migration routes have resultedfrom what the late W. W. Cookeof the Biological Surveycalls: "the innumerableexperiments as to the bestway to travel from the winter to the summer home and return." The threechief routes of North Americanbird migrationare the eastern and the westerncoastal plains, and that great jugular vein of our continent, the "Father of •Vaters." The Atlantic coast-lineis a "short-cut" to the greatest breedingground of the continent. Crossingor encirclingthe Gulf of Mexico, the main northward flight followsthe seaboardas far as Long Island. Here a triple divisionoccurs; one group of birdssweeps up the Hudson,another crosse•'•the ,Soundand ascendsthe ConnecticutValley, while the relnaindercontinues on up the coast. It is with the secondof theseroutes that our subject deals. The ConnecticutRiver, rising near the watersheddividing Canadafrom northernNew Hampshireand runningfor muchof its lengthin a generalsoutherly direction, furnishes the mostdirect and easily followedroute for birds into easternCanada. The river includingthe lakes,extends 360 milesto the sea,and is, thus the principalwaterway of New England. Its old Indian name means"Long River." As the bird flies,the routewould probably be closeto 300 miles. A natural basin among heavily wooded hills and mountainsof the ancientAppalachian system gives birth 258 •3•,oo,The Connecticut Valley. [April to this famousriver. Dashingmountain streams in the northern- most cornerof New Hampshireand lessthan one-halfmile from the Canadianborder, convergeinto Fourth Lake, a limpid moun- tain-pool,the uppermostof the quartetknown as the . It coversbut a few squareacres at an altitude of nearly 2500 feet, and is within a few score feet of the summit of Prospect. Still within half a mile of the border the outlet of Fourth Lake entem Third Lake at an altitude somewhat over 2000 feet. This lake has an area of three-fourths of a square mile and like the Fourth Lake is set in a densemixed forest, made up for the most part of small spruceand fir. SecondLake is reachedsix and one-half miles below Third lmke by meansof a fair-sizedstream, which having gone five miles.is joined by a streamequal in size,coming in anotherdirection from the Canadian border. The Lakes increase in size and decrease in altitude as we proceed. SecondLake is two and three-fourthsmiles long by a mile or morewide at onepoint, and is more than a thousandfeet lower than Third Lake. Four miles further south and the First or "ConnecticutLake" is reached,the largest of all and the body of water from whichour river directly flows. This Lake is 1680 feet above sea-level,covers nearly three square miles and measuresfour mileslong by two and three-fourthsmiles in extreme width. The surroundinghills are coveredwith deciduoustrees, sprinkledhere and therewith evergreen;but the fieldsadjoining the Lake and for a considerabledistance back have long since been cleared and summer cottagesare becomingmore and more nu- merous. Regardingthis country, Mr. CharlesL. Whittle, the geologist,after visitingthe Lake regionin 1918says:'"In the matter of the generalcharacter of the coontry outsidethe zones surroundingthe ConnecticutLakes, I will state that with the ex- ceptionof a few squaremiles of farming land about First Lake and stretchingsouthwest to Canaan,, along the roads, the entire countryis an nn13rokenwilderness covered in the main by a virgin forestof deciduoustrees from which nearly all the conifershave beenremoved by the lumberman." With a rapid, sizablecurrent the river leavesConnecticut Lake and continuesswiftly for two or threemiles with an averagewidth of 90 feet. Severalotherstreams meantime contriLute their quotas. ¾oi.1923 XL'IJ BAGG•, The Connecticut Valley. 259

Eighteenmiles from First Lake the river comesto the dividing line of Canada,New Hampshireand Vermontand in this distance descends 583 feet. It is now 1035 feet above sea-level. For the next 50 milesthe drop is but 204 feet. Toward the end of this lap the proximityof theWhite Mountainspushes the river twentymiles to the ;vest to where it joins the Passumpsicriver, a good-sized tributary comingdown from the regionof Willoughbylake to the north in Vermont. From this janetion to the Massachusettsline it is 118 miles as the crow fliesand 137 milesfollowing the course of the river. The descentis about two feet per mile. The Valley takes on now its characteristicappearance of a seriesof terraces. In the river bottom the rich soilsnourish the prosperousfarms all the way along, and with each successiveterrace, sometimesfive in number, fertility givesplace to outcropsof variousrock-forma- tionsor to remnantsof ancientflood-plains which the glacialriver overflowedeach year. The highestterrace may rise as much as 200 feet, above the river. Where the Valley is narrow, as for muchof its lengthon the line of cleavagebetween and Vermont, the terracesrise closeand steep;for the river is not as yet the broad tortuouscurrent that it becomesthrough Massa- chusettsand much of Connecticut. Contributing streamsenter all alongthe way and here and there mountain peaksstand guard. East of Haverhill, New Hampshire, towers Moosilauke,4790 feet abovethe sea,the outpostof the southwestextension of the White Mountains. Not far to the south in Orford is Mt. Cube, 2927 feet above the sea, while farther down on the Vermont side lies Mr. Ascutney3168 feet, the highestelevation lying wholly within the Valley. At Junction enters White River the largest stream in Vermont east of the Green Mountains, while in the next twenty-five miles a half-dozenmore tributaries join. Again the hills convergeand hem in the river until at Bellows Falls it becomesa foamingtorrent, the steephills on either side rising to 1200 feet above sea-level. During the seventeen-mile run to Brattleboro, Vt., the river again broadensout and becomes placidand two tributariesjoin. Twenty milesaway on the eastern edg• of the Valley, Monadnoeklifts his head to a height of more than 3000 feet. The Ashuelot,last of the Nev½Hampshire rivers, enters the Connecticut within four miles of the 260 B•, TheConnec•icug Valley. [AA7il border. At this junction the river has already come208 miles with a descentof 1412 feet, but is yet 206 feet above the sea and has 152 miles to go before entering Long Island Sound. Its total length in New Hampshirefollowing all bendsand turns is approximately236 miles,but as the bird flies it is not more than two hundred miles. Acrossthe fifty-mile extent of Massachusettsthe Valley varies in width, but averagesabout twenty miles. At the southernend, as at the north, it narrowsbetween fringing hills. Miller's River, first of Massachusetts tributaries, enters from the northeast not far from Greenfield. West of here a groupof hills appearwhich, turning southward,enter the Valley, extend along its center and twice crossthe river as the Mr. Toby and Mt. Holyoke Ranges. A unique combinationof ranges,hills and terraces,together with a great area of fertile fields through which the river meandem, sometimeseven forming ox-bows, gives the Valley throughMassa- chasettsits unequalledcharm. Near Greenfieldtwo other streams enterthe main current. The DeerfieldRiver by way of a consider- able valley flowsfrom the GreenMountains, while the Falls River, a smallertributary, comesdown from the north. In SouthDeer- field,the Deerfieldchain of hills endsin a sandstonebluff, "Sugar- loaf" by name,which overlooks the river; while directly opposite the groupof hills formingMr. Toby loom up. From this point the Valley widensout considerably,and the river flowson in great curvesand bows, the greatestof thesebeing the ox-bowat North- ampton,opposite which lies the Mr. Holyoke Range. The river runs between this and the Mr. Tom Range. Mr. Tom itself, althoughbut 1218 feet in height, riseshead and shouldersabove the surroundingcountry so that for long distancesit can readily be seenup and down the Valley, particularly on a clear day or at night when its beacon-lightshines. For the next fifty-there miles,or as far as Middletown,Conn., the Valley variesin width from three to ten miles. The river averages 1200 feet in width as far as the Connecticut State Line, and it reachesits maximum breadthat Longmeadowwhere for a mile it spreadsto 2100feet. On the eastbank, alonghalf of this mile, growsa wild rice, planted about 1880 by Ro%ertO. Morris, Springfield'sveteran ornitholo- gist. ¾ol.1923 J BAGG,The Connecticut Valley. 261

Just below the Connecticut State Line the river runs over a rocky bed for five and one-halfmiles to WindsorLocks, where becauseof rapids it seldom,if ever, entirely freezesin winter, particularlywhere the railroad crosses.For the next ten miles to Hartford the river again slowsdown, gainsdepth and averages 1500feet in width, runningin a comparativelystraight coursedue south. From this point wherea daily tide of two feet is registered, it becomesirregular again. During this interval severalstreams join forces with the main current. The chief of these is the Tunxis or , which gains accessto the Valley through its western ridge here known as the Talcott Range. From Hartford to Portland, where it is a half-mile wide, the Con- necticutpresents the curiousphenomenon of a constantlychanging course. Six miles l•elow Hartford, for instance, in a little more than a decade,it has movedits entirewidth eastward.At Middletown there remainsbut thirty-eight miles to the end of the long ran. The hills now rapidly closein until for about the distanceof a mile,known as the "Straits," onlyforty rodsseparate shore from shore. In the short distance to the sea it broadens for the last time and enters the Soundat Saybrookwith quite a wide mouth. "Thus does the Long River," accordingto Oliver Wendell Holmes, "loiter down like a great lord, swallowingup the small proprietaryrivulets very quietly as it goes." To sum up: The Valleifs easternand westernboundaries are the moun- tains of the Appalachiansystem, some in greater, bat more in lesserheight. Betweenthe two main rangesof this systemthe watershedreaches a maximum of nearly fifty miles. The River by meansof more than twenty tributaries,drains about three- tenths of New Hampshireand four-tenthsof Vermont,or a total of 6900 squaremiles in both Statesand 11,300square miles in its entire area. On the other sideof thesenorthern rangesrise rivers flowingdirectly into Canadaand northernNew England. With suchdirectness, with so many prominentlandmarks, with an un- equalednumber of connectingtributaries, does the Valley appeal to the birdsof passageand becomethe highwaythat it is to the northern network of routes. The highlands of western Con- necticutand Massachusetts including the Berkshireand Hampshire hills, the Green Mountains, the White Mountains, western and 262 •A•, TheC•nnecticut Valley. [AA• northernNew England,eastern Canadaand the Provinces---all are directlylinked up with this great Valley. Threelife zonesare representedin the Valley. At the southern extremitythe Carolinianfauna of the upper Australzone after comingeastward from New York along the Sound shoreline, extendsnorth as far as Hartford. More specificallyit approxi- mates the MassachusettsState Line, althoaghcertain species seldom,if ever, occur in the Valley north of the Connecticut cap;tal,s.ach for instanceas the CarolinaWren, Blue-winged, Hooded and Kentucky Warblers, Tufted Titmouse and Blue- gray Gnatcatcher. On the other hand such representativesof this zone as wander north either occasionallyor else in small numbersinto the Valley in Massachusettsare the Fish Crow, White-eyedVireo, OrchardOriole, Cardinal, Chat, Mockingbird and Golden-winged,Prothonotary, Worm-eating and Prairie Warblers. A notablesouthern kird that hasoccurred in Spring- field is the Blue Grosl•cak,which was seenby variousobservers incladingthe writer on May 11, 12 and 13, 1922. Recentlyit has been establishedthat the Rough-wingedSwallow, while it is wellknown to ke more or lessa k:reederin westernMassachusetts, ascendsinto New Hampshireto nest. Portland, Conn., seemsto be the dividingline in the Valley of the Parpie and the Bronzed Grackles,as the former rarely wznders farther n•rth. Inter- mediatesare found in the vicinity. From the Connecticutline well ivto the valleysof the Green and White Mountains, birds of the Transition zone are found. Most of our common birds come in this group. Just as this sectionof the Valley, particularlyin Massgchusetts,is the mecca of geologists,so it is important ornithologically,not only as a highwaybut becauseit has been made famous for generationsby the breedingof Duck Hawks. Year after year on suchcliff-sided summits as Talcott Mountain near Hartford, Mts. Tom, Holyoke and Sugarloaf, and others farther north, Faleoperegrinus ariaturn has nested. Most notedof theseis the Mr. Tom eyrie fully describedby Dr. J. A. Allen who publishedin November,1864, the first authenticdescription of the nestingand eggsof the AmericanPeregine Falcon. ! In the early

Proceedings Essex Institute, volume IV, page 153. THEAUK, VqL. XLo PL•-TEXVI.

THE CONNECTICUT VALLEY. View from west sideof Mt. Tom Range, lookingnorth, showingwest bendof Ox Bow. This is the cliff of the Duck I-Iawk eyrie. •rol.1923 J ]•AGG,The Connecticut Valley. 263 summerof 1921, R. L. Cofi•n of Amherst located in this sectionof the ConnecticutValley five of thesenests with either eggsor young. Where the more northern climate and higher altitude are encoun- tered, where spruce and fir supplant white-oak and chestnut, birds of the Canadianzone may be lookedfor. In addition to the regionroundabout the ConnecticutLakes, such peaks as Ascutney, Moosilauke and Monadnock may be called isolated zones of Canadian faunal life; at least their summitsmay. Here one will find nesting the Golden-crownedKinglet, Red-brcastedNuthatch and Myrtle Warbler. Moosilaukeis famousfor its nestinghaunts of t]icknell's Thrush. This bird as well as the more northern breedingWarblers, the HudsonianChickadee, Philadelphia Vireo, PineGrosbeak, Crossbills, Canada Jay and SprucePartridge, are foundnesting in the vicinity of the CanadianL•kes, rarely farther to the south. During the winter, such northern birds as Three- toed Woodpeckers,Snow Buntings, Redpolls, Tree Sparrows, Acadian Chickadees, Pine Grosbeaks, Crossbills, Goshawks, Rough-leggedHawks and Snowy Owls descendthe Valley in searchof food, often well into Confiecticutand sometimesbeyond.' Several decadesago large incursionsof Rough-leggedHawks oc- curred. Sixtywere taken in onewinter on the Hadleymeadows in the vicinity of Mt. Holyoke--most of which were mountedfor different museums. In another winter forty were collectednear East WindsorHill, Connecticut. The Northampton and Hadley meadowsproved formerly a great attraction likewise for Short- caredOwls. Mr. Morris reportsthat duringthe eighties,colonies of them were found there in winter. Mr. Edward O. Damon of Northamptononce encountered a flight which he estimatedto be at least one hundredbirds which were flying low down in the twilight. The only record we have of the Gyrfalcon is of one takenby Mr. Damon many yearsago in this vicinity. Dr. J. A. Allen recordsa Raven taken in 1859, at Springfield. With the cutting off of more and more timber such rare birds are in late yearsseldom, if ever, observedeven in the northernmostpart of the Valley. q¾o speciesthat originallywere a commonsight when the set- tlers first came to the Valley are now extinct, namely,the Wild Turkey and the PassengerPigeon--chiefly because they formedso 264 BAGø,The Connecticut Valley. [April[Auk

large a part of the people'sfood. Talcoft Mountain had Wild Turkeysduring the earlypart of the nineteenthcentury; but the Mr. Tom and Mr. HolyokeRanges seem to haveafforded the best protectionto thespecies in theValley for it washere in theearly fiftiesthat thelast Turkey was captured. PassengerPigeons, when the flightswere on, camein an unendingprocession to and from the large foreststo the north and west. They were abundant until the seventieswhen their ranksbegan to thin out,and the last Pigeonswere observedin the springof 1884 at SouthwickPonds and in the spring and fall of 1887, at Portland, Connecticut. JohnH. Sageshot a youngmale October1, 1889,at Portland, which constitutesthe last Valley record so far as we are aware. T•vo other speciesthe Purple Martin and the Dickcissel,while not extinct, might as well be so far as our Valley is concerned. Nestingabundantly in formeryears throughout most of its length, but a few scatteringPurple Martin coloniesremain today. As with the Wild Pigeons, the Martins began to decreasein the eighties,largely due to the rapidincrease of the EnglishSparrows, which drove them from their nestingplaces. A few colonieswere breedingin 1912,in Portlandand Middletown, but it seemsonly a questionof time whenwe shallno longernumber these attractive Swallowsamong the migrantsof the Valley. Prior to 1840 the Dickcisselwas a commonbird in summerin the Valley. Dr. J. A. Al.lenrecords one taken in May 1866, in Holyoke, which seems to be the last authentic record. But notedas the Valley is in thesephysical and faunal aspects, it is primarily famousas a pathwayfor migration. Every month of the year witnessesflights of variousbands or flocksof birds; at timesthere may be but a few stragglersor only a solitary stran- ger. Even in the interim, before the spring and the autumn migrations,birds are movingto and fro; indeedit is difficultusually to tell when one migrationleaves off and the other begins. Nor do the cold days and nights of winter, supposedlybirdless, drive all bird-life from the Valley. Particularly is this true the nearer the river reachesthe Sound. During most winters many birds remainin the vicinityof Saybrookand other shoretowns waiting to ascendthe river with the first signsof spring. For the most part suchwintering birds are Grebes,Loons, Mergansers, Golden- 'Vol.1923 XL] J BA(•(•,The Conn½cticu: Valley. 265 eyedand Black Ducks,Canada Geese, Great Blue Herons,Mourn- ing Doves, Red-tailed, Red-shouldered,Roughdegged, Marsh and Duck Hawks, Kingfishers, Flickers, Cowbirds, Blackbirds, Meadowlarks,Vesper, Fox, White-throated, Song and Swamp Sparrows,Juneos, Cedar Waxwings, Myrtle Warblers, Pipits, Catbirds, Winter Wrens, Hermit Thrushes and Robins. All through the winter wheneverthere is open water, Herring Gulls may be seenflying abovethe river as far north as Holyoke; while Mergansersand Golden-eyedDucks ascendthe river at times to northern Vermont. In the bitter winter of 1918-1919, when a little open water still persistedin the Connecticut,at Newbury, Vt., where comesin, a few of the rare Barrow's Golden- eyesflocked with a larger numberof the commonspecies. In early spring one might well comparethe oncomingbirds to the column in a thermometer. With the incessantchange of New England weather the vanguardadvance and retreat at first much as does the quicksilver. Birds that have bred for several seasonsin a particular spot seemmost anxiousto escapethe in- activity of the long winter months and to return to household duties again. Bluebirds are quick to return to the old nest and use the Valley in large numbers. I have in mind, likewise,a Phoebethat nestedeach year over the front piazza doorwayof a South Deerfield farmhouse, ' the owner of which sometimes re- ported the return of his Phoebe in the spring as much as a week before one had }:eenseen or reported in either Spring- field or Holyoke. Northern nestingRobins are similarlyanx- ious to be the initial claimants for the "early worm." With the advent of a warm wave Robins by the score will appear in the Valley and push rapidly northward. Then, next day, with the. veering of the wind, the Valley may be oncemore in the grip of winter. Snow or sleet with much w:nd de- scendsand sendsthe eager birds to cover. Hundredsand some- times thousandsof Robins may be seennow in large flockseither retreating down the Valley or scurrying back' from the hill- country. Such storms bring about a retrograde movement, a goodexample of which occurredin the first day or two of April, 1919, when large numbersof Redpolls, Siskinsand Robins drifted down the Valley before'the storms. At Hatfield an observer 266 BAoo,The Connecticut Valley. [AA• stated that Robins by thousandswere taking refuge in the long stubbleand grassof the fields. When startingup in simultaneous flight, they presenteda tremendousarray. In the vleinlty of Holyoke,with the storm,came numbers of Tree and Fox Sparrows and Juneos--birdswhose ranks were beginningto thin out--while Bluebirds,Red-winged Blackbirds and Grackles largely disap- pearedin retreatbut put in appearancepromptly again on April 3rd when the snow and ice had melted. A like movement of even greaterproportions occurred with the severestorm of the last day of March, 1922. Seven or eight days of warm weather brought the early birdsin eonslderableforce. m•ny of them ascendingthe Valley into Vermont. The unexpectedsnow storm caughtthem unawares;and with it unabatedon the secondday. retreat was their only salvation. Juneosby the hundreds,Fox and Song Sparrowsand Robins by dozenswere now to be found in the vicinity of 5•t. Tom. Hawks and Woodcockwhich were ob- serveda week beforemaking their way northward could now be seengoing in the oppositedirection. Phoebesseemed sadly out of placeamid the falling flakes. Somethingmuch out of the or- dinary occurrednear Brattleboro,when flocksof Canada Geese were observedcoming back down the river where several days prior they had ascended.Yards, gardensand highwayswere suddenlyfilled with birdsthat we'had not expectedto seeagain till fall. 1Vfanythat came to our trap proved to be in a weak, starvingcondition, with breastbone prominent. Birds that we had beentrapping completely disappeared, and the forty birdswe bandedduring thesethree days (23 Song Sparrows,10 Juncos, 4 Tree Sparrows,1 Fox Sparrow,1 Chickadeeand 1 Bluebird) were all newcomers. April bringsmore and more new arrivals. Immenseflights of Sparrowbands, one after another in rapid succession,pass north duringthis month,filling our gardensand fields,our thicketsand hedgerows. Of this processionthe Juncos are in the van.. many having passedthe winter in the southern end of the Valley or vicinity. Like leavesbefore the October wind I have seenthem billowpast, twittering as they flew. Song Sparrowsaie heretoo, as are the Fox Sparrows, Savannah, Vesper, Field and Chipping,but especiallythe White-throats,with a sprinklingof the White-crowned.The real processionof passing 'Vol.1923 XL] J BAG%The Connecticut Valley. 267 birds,however, may well be saidto reachon the averageits climax duringthe first half of 5Iay. Then onemay well gain an idea of what an importanthighway the ConnecticutValley is by going afieldwell nigh anywherettu'oughout its lengthand breadth. In groupsit may be, or in companies,singly or in immenseflocks, the headlong stampede wings its way 'continually northward. Brightest and best of all the flying hostsare the Warblers. I recall a certainspring day not many yearsago, when seatedon a favorite stump, I was watchingwave after wave of thesesprites descendupon a clump of white birches above the western bank of the bend in the river, just north of Holyoke. In a scoreof minutes, I observedas many or more of the various Warblers, including the rarer Wilson's, Cape May, Tennesseeand Bay- breasted. The never-to-be-forgottenweek in May, 1917, readily comesto mind. A sudden turn for the cold not only held the Warblers in check but brought them to our very door-yards. Retardedlikewise was all insectlife and hundreds,possibly thous- ands,of Warblersmust have perished. Schoolchildren brought in many in starvingcondition and sofeeble that they werereadily captin'ed. One farmer told of dozensof theselittle birds settling down all over him, his horsesand load as he drew manure to his field. That same week a similar condition existed in most of the Valley as far north as Vermont. Mr. E. C. Bliss states that severalcarloads of manure happenedto be piled in one place at South Deerfield ready for the tobacco fields. The heat from the compostbrought out a smallcrop of insectswhich in turn attracted hunch'edsof Warblers. They completelycovered the big pile measm'ingmany cords, and had to be well nigh brushed off, so hungryand weak were they, beforethe men couldload their carts. For most of that week a Yellow-bellied Flycatcher and several TennesseeWarblers remained in our front yard, and 'subsistedon barberries.Sometimes a Golden-wingedWarbler, oce/•sionally the Mourning, and rarely the Cerulean, accompanythe main flight. For an entire week in May, 1921, R. L. Coffin reports that a Cerulean Warbler could be observedalmost any time of day in a certain forsythia bush of a neighbor'syard in Amherst. The principalpleasm'e in witnessinga springWarbler-flight is due to the fact that they usually precedethe fully-developedfoliage. 268 BACG,The Connecticut Valley. [AAp•rrtl

A much easiermatter is it then to observeand identify the ever- movingbirds and quite oftento registertheir songsas well. May of 1921,brought a disappointmentin this respect. The stagewas all set for a bright pageantbut no Warblers! The foliagepushed out to full bloom, and still they did not come. Finally there arrived a very hot morning when it seemedas if the birchesand bush-borderedstreams were delugedwith birds. Not only were most of the Warblers representedbut Flycatchers, Vireos and Thrusheswere aplenty. Thinking we could augmentthe morn- ing'slist--at noonwell upwardsof one hundred--I visitedin the afternoon Chesterfieldand the hill-country. But not a Warbler anywhere;birds were extremely scarce. The Swamp Sparrow provedto be the only new bird we couldlocate. Obviouslythe flight was pouringup the Valley, but it had not penetratedthe higher hill-country to the east or west. The heightswere as free of birds as the Valley had beena few daysbefore. The Warblers bring the springmigration to a dramatic climax, and when they have passedthey leave one with a feeling of somethinglost. With the greatspring burst of birddife;with all its songand color, all flights to the novicemust seemdiurnal. Many species,how- ever, fly by night. While canoeingup the ConnecticutRiver, near Mr. Tom Junction, early in May, 1921, our attention was called to the loud call-noteof the Greater Yellow-legs. The high river water of springhad overflownthe northern bank and left there a long lagoon-likepool of smallwidth and depth. In this shallow water we discovered with the aid of the binoculars seven Yellow-legs,their headsbobbing and their call notescoming clear and rapid. :Keepingunder the lee of the high bank, we quietly paddledso near that the lengthof the canoebarely separatedus. Very cautiouslyapproaching, we obtainedexcellent observations at closerange. Finally, to watch their flight we let our presence be known, when all flew off a short distanceto a sandy beach of the river. Beingunmolested they presentlyfell to preeningtheir feathemand not long after one or two pulled up a foot and took a noon-daysiesta, with headstucked under wings. Most of the others followed suit and so we left them. Returning late that afternoon,I headedthe canoefor the spotwhere we first discovered them. My companionimpetuously broke the stillnessto say •Yol.1923 XL] .I BA(•G,The Connecticul Valley. 269 that it was uselessto cometo the high bank a secondtime; the birds surelyhad flown away. But he had scarcelysaid the last word whena chorusof alarm notesclose by camefrom the lagoon, and presentlythe Yellow-legsrose in air. Their flight this time was different. Risinggradually they flew with strongwing-beat in closearray into the westerntwilight until they were well up over Mr. Tom Junction. As they wheeledto the right high in air, we dimly discernedthem as they fadedout of sightin a direct courseup the Valley, flyingapparently straight over Northampton insteadof followingthe meanderingriver. Doubtlesswhen re- turningdaylight came, it foundthem beyond the Canadianborder, well on their longjourney to northernLabrador or Ungava. On anotherMay day I recalla singleNighthawk that lay lengthwise on a limb of an elm tree in our front yard, twenty-five or thirty feet abovethe sidewalk. Both eyeswere closed, and he seemedto be sleepingundisturbed although swayedwith the breeze. His perchwas within twenty feet of the house. Not till nearly dusk did he take leave. Suchnocturnal flights are com- parativelyeasy on a brightmoonlit night, for the riversstand out like silver threadsas the birds wing their way on high. Like sleepingsentinels the taller hills or peakslie revealed,and from this greatheight the mostdirect route is readilydiscernible. It is, however,in the fall whenthe great host,now voicelessand with many birdsstill in molt, makesthe return journeythat this path- lesscourse of the upperair is employedto the largestextent. The late Willia•n Brewsterafter his intensivestudy of migration at Point Lepreauxon the Bay of Fundy, cameto the conclusionthat birdsmigrate either singly or in stragglingbands; that the different bandsare connectedwith othersso as to form a practicallycon- tinuousstream of birds; that adults precedeor accompanythe first flightsof the young. Ho'• oftenon a starlightnight in late Septemberor Octobercan one hear the lispingnotes of the Warblers,some faint, otherslow down and distinct as the multitude pass southward. Brewster foundnot onlyat PointLepreaux but elsewherethat on favorable nightsthe sky is literally "alive with birds." (Suchnights are not infrequentin our Valleywhen the flightis on.) He felt, moreover,that youngbirds whoseparents had departedearlier, 270 Baoo,The Connecticut Valley. [AprilFAuk fell in with the generalprocession; that becauseof a strongsense of localityand directionthey "quicklylearn to followa path, even for thousandsof miles,along which .they have beenin the first instanceled by older companions,and the youngbird after a few seasonsbecomes an experiencedguide." This must be the case likewisehere in the Valley, where, becauseof the guidingpeaks and the north and south direction for so much of its distance,it must be a simplematter for even young birds to becomeaccus- tomedreadily to any portionof it. In the fall, then, we find the spring flocks now augmentedwith large numbersof immature birds, and distinct flights occur. In August for instance the Nighthawkspass down the Valley in someseasons in greatnumbers. Mr. James Mackaye, of Boston, tells of a flight of this kind in August, 1881, near Brattleboro when he observedhundreds flying overhead in one large flock. September brings by night the Warbler host accompaniedby Vireos and Thrushes. By day the Swallowsand Swifts take leave and the Hawkflights occur. In 1921, the Hawks came on September18th, when seventy-fiveor eighty, mostly Red-tailed and Red-shoulderedHawks, were countedin oneflight. In someyears these flights vary according to the latenessof the season. In 1910for exampleit did not occur until October 10th. This flight, however,was one long to be rememberedfor four or five hundred Hawks made up the lengthy line that drifted at noon over Holyoke. They were travelling at rapid rate althoughfew wing beats were noted; all were taking advantageof the strongnortherly wind whichcarried them along with so little effort. The older Hawks seem to await such a favor- able wind beforestarting a flight of this kind down the Valley. October,like April, is the specialmonth for the Sparrows--this time on the return trip. One would imagine it was the same flockin one'sgarden staying the month out; but bird bandinghas proventhat a flockwill remainseveral days until a stormarrives when they go on for the next lap. The samestorm, however, bringsin a newgroup which remain until the followingstorm, and so on. About the middle of Octoberthe Woodcockflight arrives. Then it is that the alder thicketsharbor a hostof thesenocturnal travellers,particularly with the comingof the hunter's moon. Night after nightthey continueto arriveand moveon, andmany

Vol.1923 XL] J B•-oG,The Connectic• Valley. 271 lingervery late in the fall. In 1908one lingered until December 8, but was picked up in Agawanhaving died only a short time previous. A Woodcockhas been capturedin January, not far [rom Saybrook,with six inchesof snowon the groundand a gale blowing. It was found near an open spring-hole,and was in an emaciated condition. Another was observedas early as March 8 in 1919, in the vicinity of Holyoke. With the coming of colder days and nightsin Novemberthe Robinsand Bluebirds retreat. All throughthe fall they had beengoing south in scatteringbands or mixed flocks,but now even the rear guard disappear,although a few Robins occasionallywinter especiallyin the vicinityof the Mt. Tom and Mt. Holyoke Ranges. During the winter of 1921-1922 a flock of eleven Meadowlarks remained near the Amherst Agricultural College Campus, and three or four more near the Mt. Holyoke CollegeCampus. Novemberwitnesses the gatheringof immenseflocks of Crowsfrom the woodedhill country to the•eastand west. Theycome into the Valley as severe weather sets in. SometimesCrow flights occur. Mr. Mackaye records one of theseflights in the late fall of 1881 near Brattleboro. He first noticedthe Crowsin a long cloud-likeformation comingdown the Valley and gradually settling and comingto rest on a hill where,even though more than a mile away, the babbleof somany voicesmade a great noise. With the comingof inclementweather many Crowsdesert the valley of the Connecticutand its tribu- taries north of Massachusetts. Upon the return of the species with the passingof winter, the Crowin manyplaces (as in Haver- hill, N.H.) isregarded as a harbingerof springas much as the Robin or the SongSparrow. Early in November,1919, a similargather- ing of severalthousand Crows was notedat Worthington,Massa- chusetts. With the initial snowfall such flocks retreat into the Valley to pass the winter in the vicinity of Mt. Tom and Mt. Holyoke where favorable roosts are found. When the snow becomesdeep, however, they retire still further south. The seasoncomes to a closewith the flight of Canada Geesewhich occursusually in late November or sometimesearly December. Again, Brewsterbelieved that along every route couldbe found "stations or points of departure"--favorable placeswhere the "migratory tide hesitatesor halts" either for rest or food. Let r•e mentionseveral in our Valley. 272 BAGG,The Connecticut Valley. [Apri[Auk

A few hours flight from the Sound brings most birds to the sectionof the Valley, where there are numerousponds, streams, swampsor tributary valleys, ideal natural locationsfor every sort of refuge and food. I have in mind partleularly the region extendingroughly from the Conneetleutstate line to the vicinity of Mt. Tom. Mr. John H. Sage tells of several hundred acres, kinownas the Little River Meadows,on the oppositeside of the from his home at Portland, Conn. From boyhooddays he annuallywitnessed there great Swallowflights in August. There were Barn, Eave and Tree Swallows,principally the last, all of whom took refugefor the night in the water-oats whlehgrew in abundance. In closeranks one above the otherthey clungto the standingstalks. They first beganto gather,he noted, early in August,and eontlnueduntil late in September,when the majority departed. He has found a few there until the last week of Octoberand states that it is an excellentplace likewisefor "Rail, Marsh-Wrens and many other birds" (' The Auk,' Volume XII, p. 83). Thesemarshes are likewisea favoriteroosting place of Grackles,thousands of the Purple and the Bronzed varieties flockingthere in the late summer,and aeeordlngto Mr. Sage,out- numberingin early November all other species. I have already mentionedthe wild flee fringing the river for half a mile in Long- meadow, just above the Connecticutstate line. The expenseof securingthe wild flee was sharedby Mr. M. B. L. Bradford and Mr. Morris. They securedthe seed from Minnesota. /vlindful of theproper time in thefall for plantingthey earfled it downthe river by rowboat, and were successfulin starting a strip about two hundred yards long closeto the shore. Each succeedingyear saw the growth of this rice widen and lengthen until today it is about half a mile long and fifty yards wide. It has attained a height of six feet or more. Here and in the fieldsadjoining Mr. Morris has collectedmany of his unusualrecords including Bona- parte's Gull, the CommonTern, the King Rail and the Yellow Rail. It is here, too, that nearlyevery fall, late in August,he has recorded the Black Tern. The Black Rall has also been observed here. In the fall the wild rice provesan attraction for not only Sandpipers,Snipes and Ducks but for large flocksof Blackbirds and of Savannahand SwampSparrows which with an occasional Vol.192a XL] J BXGG,The Connecticut Valley. 273

Sharptailcongregate there. The onlyrecord we have of a Savannah Sparrowwintering is of one whichMr. Morris foundnear here in 1895-96. A dozenmiles directly west of Springfieldlie SouthwickPonds or CongamondLakes. They comprisea group of four narrow, connectedlakes, two and one-half miles long by one-half mile wide. Gulls, Ducks, Geeseand otherwaterfowl visit theseponds frequently. Mr. Morris for a long time was of the opinionthat LaughingGulls came occasionallyup the Valley as far as Massa- chusetts.While rowing on the Ponds early in October,1920, he noteda Gull restingalone on the water. On comingwithintwenty feet he discoveredit to be a LaughingGull which seemednot at all disturbed by his proximity. Rarely the White Egret visits theseponds. In mid-July, 1920, early in the mornihg,Mr. E. S. Smith of Holyoke, after an all-night fishingtrip, came upon an Egret in the shallow water near shore. It appeared unusually xvhitein the mistsof early dawn. It was here too, that a large flight of Snow Geese,the first in many years, were reportedin late Nove•nberor early December,1921. Three were shot on ThanksgivingDay, in Westfield Little River, the Springfield Museum specimen,an immature Greater Snow Goose,being one of them. One was taken at Portland, Conn.; another at East Windsor Hill; and a small flock was observedabout the same time at Glastonbury. All undoubtedlywere part of the main flock estimated at upwards of one hundred which visited Southwick Ponds.t In the days of the first settlers these Geesenumbered thousands,.and every year in thefall flightin particularthey used the ConnecticutValley as a highway. Today we have their kin --the Canada Geese--whosespring and fall flights usherin and bringto a closethe mainmigration season. Their V-shapedforma- tion and loudhonkings are familiar occurrencesin mid-March and late November. The last two seasonshave brought the largest flights recordedin many years. Sometimesthey arrive much hamperedby sleet and ice storms. The writer's grandfatherin his early days caughtby hand severalgeese so coatedwere they with ice. This occurredabout a century ago near his farm in West Springfield,Massachusetts. SeeA•k, vol.XXXIX, Apr.,1922, p. 251. 274 Bao,The Connecticut Valley. l'AukkApril

A scoreof miles north of Southwick,the confluenceof several small valleysin what is known as Fomer bringsabout conditions especiallyfavorable to migrating birds. During the winter of 1918-19 came a large influx of Downy and Hairy Woodpeckers. On Lincoln'sbirthday of that winter, which was an unusually open one,the writer saw at Fomer the Arctic Three-toed Wood- pecker reported wintering there sinceNovember. We found him peelingbark and boringfor timber beetles(Buprestidae) in a burnt- over patch of white pine. That same day we observednot far from there a dozenRobins in one flock. On November 29, 1921, an immenseflock of Pine Grosbeakswas noted here, spreadout over half an acre and numberingit was estimated, from three to four hundredbirds. There are many more suchstopping places equallyfavorable that we might tell of. HampdenPonds, Ashley Ponds,the Island in the river at Holyoke, Aldrich Lake, Forge Pond, Carver's Pond and others in the vicinity of Holyoke all come within such a group. The Hadley meadowsmentioned earlierin the paper equalperhaps any other suchplace we could mention. It was here through the seventiesand eightiesthat 3•lr. E. O. Damon of Northampton collected among others, such unusual birds for the Valley as Knot, Buff-breasted Sand- piper, HudsoninnCurlew, Red-breasted Merganser, Prothonotary Warbler, Gyrfalcon, Great Grey Owl, AmericanHawk Owl, and Northern Raven. Clerk of Courts Chilsonof Northampton was an expert in collectingRough-legged Hawks here. "If you went afoot with a gun," he relates, "your hawk was sure to seeyou and take flight a long way off, but a passinghorse and buggy they would not suspect. By driving my horseat first in large circlesand gradually cutting thesedown until within range, and then gallopingrapidly toward the tree, I could usually secure my specimen." Various sea birds driven inland by stormsare taken from time to time in the Valley. A Red-throatedLoon at Hartford caught itself in the electricstreet wires during a fog and was capturedon November 20, 1895. A Brunnich's Murre was taken by Mr. Morris, November 30, 1899, in Longmeadow and two more were taken near there--one on December 19, 1897, following a severe gale,the otherin January1901; still anotherwas taken February 1, Vol.1923 XL] J B.•,NCROFT•T AL., Blue Feathers. 275

1902, in Ware. The Dovekie has been taken at Greenfield and Belchertown where in 1872, large numbers were secured. The Leach'sPetrel was taken September2, 1900, at Agawan; October 11, 1905, at Hampden Ponds; and in October, 1908, at Smith's Ferry. A Black Skimmer was found exhaustedin the latter part of August, 1893, in West Springfield. In 1877 or 1878, a number of Double-crested Cormorants were taken from small flocks near Springfield. Red-breastedMergansers have occurred--oneApril 28, 1908, in Longmeadowand one prior to that in Northampton; also Gadwalls--one October 14, 1904, at Glastonbury, Conn., and oneNovember. 5, 1883,at East Hartford. Old-squaws,American, Surf and White-w/nged Scoters, and Brants have been either identified or taken near Springfield. Two speciesof Phalaropes have beenknown to usethe Valley in the fall as well as the Purple Sandpiper,Red-backed Sandpiper and Willet. By the keeping of migration dates up and down the Valley, museums, scientists and interested bird students have collected much valuable data relative to arrivals and migrations. It is hopedto establisha seriesof bandingstations througho. ut the length of the Valley. Bird bandingmore than any other agency shouldreveal all and more than all that can be written emphasizing the importanceof the ConnecticutValley as a highway for migra- tion; and of its constantuse, year in and year out, by thousandsof migrants in both directions. 70 FairfieldAve., Holyoke, Mass.

BLUE FEATHERS •

BY WILDER D. BANCROFT, EMILE M. CHAMOT, ERNEST MERRITT, AND CLYDE 'W. MASON.

PHYSICISTSdistinguish between pigment and structural colors. Pigment colorsdepend on the chemicalnature of the material and are due to the absorption of certain wave-lengths by the mole-

2 The investigation upon which this article is based was supported by a grant from the Heckscher Foundation for the Advancement of Research, established by AugusE Heckscher at Cornell University. The paper, of which this is an abstract, will be published in full in the Journal of Physical Chemistry.