Vol. IX May, 1936 No. 2 BULLETIN OF The New York State College of Forestry At Syracuse University Samuel N. Spring, Dean Roosevelt Wild Life Annals VOLUME 4 NUMBER 2 OF THE Roosevelt Wild Life Forest Experiment Station STUDIES ON THE BANK SWALLOW, Riparia tiparia vipavia (Linnaeus) DAYTON STONER CONTENTS OF RECENT ROOSEVELT WILD LIFE BULLETINS AND ANNALS BULLETINS Roosevelt Wild Life Bulletin, Vol. 4, No. 1. October, 1926. 1. The Relation of Birds to Woodlots in New York State Waldo L. McAtee. 2. Current Station Notes Charles C. Adams. (Out of print) Roosevelt Wild Life Bulletin, Vol. 4, No. 2. June, 1927. 1. The Predatory and Fur-bearing Animals of the Yellowstone National Park. Milton P. Skinner. 2. Current Station Notes Charles C. Adams. Roosevelt Wild Life Bulletin, Vol. 4, No. 3. July, 1927. 1. A Trout Survey of the Allegany State Park in 1922. William C. Kendall and Wilford A. Dence. 2. A Preliminary Survey of the Fish Life of Allegany State Park in 1921. Thomas L. Hankinson. 3. Current Station Notes Charles C. Adams. Roosevelt Wild Life Bulletin, Vol. 4, No. 4. July, 1927. 1. The Beaver in the Adirondacks : Its Economics and Natural History. Charles E. Johnson. Roosevelt Wild Life Bulletin, Vol. 5, No. 1. March, 1928. 1. A Preliminary Wild Life and Forest Survey of Southwestern Cattaraugus Co., N. Y Victor H. Cahalane. 2. A Preliminary Report on the Trout Streams of Southwestern Cattaraugus Co., N. Y Wilford A. Dence. (Out of print) Roosevelt Wild Life Bulletin, Vol. 5, No. 2. February, 1929. 1. The Fishes of the Cranberry Lake Region W. C. Kendall and W. A. Dence. 2. The Story of King's Pond F. A Lucas. 3. Its Fish Cultural Significance W. C. Kendall. Roosevelt Wild Life Bulletin, Vol. 5, No. 3. September, 1929. 1. The Summer Birds of the Northern Adirondack Mountains Aretas A. Saunders. 2. The Summer Birds of the Adirondacks in Franklin County, N. Y. Theodore Roosevelt, Jr., and H. D. MinoL (Reprinted. Original date of publication, 1877) Roosevelt Wild Life Bulletin, Vol. 5, No. 4. August, 1930. 1. The Biology of the Voles of New York Robert T. Hatt. 2. The Relation of Mammals to the Harvard Forest Robert T. Hatt. Roosevelt Wild Life Bulletin, Vol. 6, No. 1. March,. 1931. 1. A Biological Reconnaissance of the Peterboro swamp and the Labrador Pond Areas. Charles J. Spiker. Vol. IX May, 1936 No. 2 BULLETIN OF The New York State College of Forestry At Syracuse University Samuel N. Spring, Dean Roosevelt Wild Life Annals VOLUME 4 NUMBER 2 OF THE Roosevelt Wild Life Forest Experiment Station Entered as second-class matter October 18, 1927, at the Post Office at Syracuse, N. Y., under the Act of August 24, 1912 ANNOUNCEMENT The serial publications of the Roosevelt Wild Eife Forest Experiment Station consist of the following: 1. Roosevelt Wild Life Bulletin. 2. Roosevelt Wild Life Annals. The Bulletin is intended to include papers of general and popular interest on the various phases of forest wild life, and the Annals those of a more technical nature or having a less widespread interest. The editions of these publications are limited and do not permit of general free distribution. Exchanges are invited. Sale prices for the Station publications are based on the actual cost of printing and distribution in accordance with Chapter 220 of the Laws of 1933. Price lists will be furnished on request. All communi- cations concerning publications should be addressed to The Director and Editor. Roosevelt Wild Life Forest Experiment Station. Syracuse, New York Copyright, 1936, by Roosevelt Wild Life Forest Experiment Station TRUSTEES OF THE NEW YORK STATE COLLEGE OF FORESTRY Ex Officio Dr. Charles VV. Flint, Chancellor Syracuse University Dr Frank P. Gravf.s, Commissioner of Education Albany, N. Y. Hon. Lithgow Osborne, Conservation Commissioner Albany, N. Y. Hon. M. William Bray, Lieutenant-Governor Albany, N. Y. Appointed by the Governor Hon. George W. Driscoll* Syracuse, N. Y. Hon. Alfred E. Smith New York City Hon. William H. Kelley Syracuse, N. Y. Hon. Edward H. O'Hara** Syracuse, N. Y. Hon. Charles A. Upson Lockport, N. Y. Hon. J. Henry Walters New York City Hon. George W. Sisson, Jr Potsdam, N. Y. Officers of the Board Hon. Alfred E. Smith President Hon. WiLLiAm H. Kelley Vice-President HONORARY ADVISORY COUNCIL OF THE ROOSEVELT WILD LIFE STATION Hon. Theodore Roosevelt New York City Mr. Kermit Roosevelt New York City Dr. George Bird Grinnell New York City Hon. Gifford Pinchot Harrisburg, Pa. Mr. Chauncey J. Hamlin Buffalo, N. Y. Dr. George Shiras, 3rd W ashington, D. C. Dr. Frank M. Chapman New York City Dean Henry S. Graves New Haven, Conn. * Deceased, January 31, 1936. * Deceased, February 10, 1936. [123] ROOSEVELT WILD LIFE STATION STAFF Samuel N. Spring, M.F Dean of the College Charles E. Johnson, A.M., Ph.D Wilford A. Dence, B.S H. Ruth Merrill Director of the Station Ichthyologist and Ass't Director Secretary Temporary Appointments * Myron T. Townsend, Ph.D Field Naturalist A. Sidney Hyde, M.A Field Naturalist Murvel R. Garner, Ph.D Field Naturalist Dayton Stoner, Ph.D Field Ornithologist Charles J. Spiker, M.A Field Naturalist * Including only those who have made field investigations and whose reports are now in preparation. 1124] Digitized by the Internet Archive in 2015 https://archive.org/details/rooseveltwildlif04unse_4 Map 5. Map showing territory at the east end of Oneida Lake and the location of the principal field stations where most of the work discussed in this report was done. Adapted from U. S. Geological Survey quadrangles. [126] STUDIES ON THE BANK SWALLOW RIPARIA RIPARIA RIPARIA (Linnaeus) IN THE ONEIDA LAKE REGION By Dayton Stoner Field Ornithologist , Roosevelt Wild Life Forest Experiment Station, Syracuse, N. Y* CONTENTS PAGE 1. Introduction and Acknowledgments 129 2. The Oneida Lake Region 131 Seasonal Data on Temperature, Humidity and Precipitation 131 3. Descriptions of Field Stations Representing Typical Bank Swallow Habitats in the Oneida Lake Region 135 Station 1. Herder Farm, South Bank of Fish Creek 135 Station II. Mott Farm, North Bank of Fish Creek 136 Station III. Fox Farm, North Bank of Fish Creek 137 Station IV. J. Cook Farm, South Bank of Fish Creek 138 Station V. Daniels Farm, North Bank of Fish Creek 138 Stations VI and VII. Chas. Edell Farm, South Bank of Fish Creek 141 Station VIII. E. H. Onderdonk Farm, South Bank of Fish Creek. . 142 Station IX. W. P. Graham Farm, South Side of Fish Creek 142 Station X. Grems' Sand Pit, One-fourth Mile South of Mc- Connellsville ■ 143 Stations XI and XII. Bryant's Sand Pits 143 Station XIII. Humaston Sand Pit 144 Delahunt Sand Pit 144 Gravel Pit, Oak Orchard District 147 Dutcher Farm Sand Pit, Oak Orchard District 147 4. The Local Abundance, Distribution and Seasonal Occurrence of the Bank Swallow in the Oneida Lake Region 147 5. Weight Studies of Bank Swallows 153 Weights of Adults 153 Growth Rate in Young 158 6. Body Temperature of Bank Swallows 159 Body Temperature of Adults 159 Body Temperature of Young 176 7. Breeding Habits of Bank Swallows 180 Burrows 180 Nests and Nest Building 193 Eggs and Egg Laying 194 Young 195 ■■ Appointed State Zoologist, New York State Museum, Albany, New York, October 15, 1932. [127] 128 Contents PAGE 8. Feeding Habits of Adults and Young iOl Remarks on the Arthropod Groups Represented in the Food of Hank Swallows 207 Summary ill 9. Destructive Agents ili Man ili The Skunk ili Slipping Banks 215 Accidents 215 10. Parasites and Disease 215 Mites 215 Fleas 215 Biting Lice 216 Protocalliphora Larvae 216 Foot Disease 219 11. Miscellaneous 221 12. Banding Studies 2ii Method of Capture 223 Behavior of Captured Birds 224 Recoveries of Captured Birds 2i4 Remarks 228 13. Bibliography 230 Studies on the Hank Swallow 129 INTRODUCTION AND ACKNOWLEDGMENTS Among the more than 170 forms of birds which are known to occur in the Oneida Lake region during late spring and summer, the bank swallow is one of the most abundant breeding species. Its unusual nesting sites, its colonial life and its remarkable ability on the wing combine in drawing to public attention this interesting and valuable bird. Although more or less circumscribed in its local distribution its habit of nesting in colonies tends to bring considerable numbers of the bank swallow together in favorable situations. Of the six species of the swallow family represented in this territory the bank swallow is probably the commonest. Its closest rival in point of numbers is, perhaps, the more generally distributed tree swallow. With the completion of field work for a previous report (Ornithology of the Oneida Lake Region : With Reference to the Late Spring and Summer Seasons) by the present author (1932), several subjects worthy of more detailed study than could be given them at that time presented themselves for considera- tion. Among these, certain questions relating to the bank swallow seemed to merit attention. Despite the fact that this swallow has been the subject of numerous investigations, a great deal remains to be learned about it. The species is one that lends itself better than do many other birds to studies in which the banding method may be employed, for during the breeding season considerable numbers of both adults and young can be secured, and there is a reasonable prospect of recovering at least some of the banded individuals during the same season. Its nesting habits render it possible at times to secure a very considerable amount of material within limited areas — a matter of distinct advantage. The writer had previously conducted certain investigations on the bank swallow in the Lake Okoboji region, northwestern Iowa (Stoner, 1925, 1926, 1928, 1928a). A portion of the data secured in these field studies remained to be reviewed and analyzed, and as it was felt that they would supplement in a useful way the results of the investigation undertaken on the same species in the Oneida Lake region, they were incorporated in this report. The present study on the bank swallow in the Oneida Lake territory was begun on May 1 and continued uninterruptedly until July 30, 1931. At the close of the field work analyses of the stomach contents of the bank swallows were carried on in the laboratories of the Roosevelt Station, and later at the New York State Museum. Most of the larger nesting colonies of bank swallows occur along Fish Creek which meanders through the low, flat land east and northeast of Oneida Lake. Preliminary reconnaissance trips were made, beginning May 1, on which were noted the habits and activities of the bank swallows that had already arrived. In addition, efforts were made to determine the abundance, distribution and points of greatest concentration of the birds, and at the same time to locate nesting sites which would be reasonably accessible. As a result it was decided to concentrate our endeavors in certain localities in which a number of populous colonies were conveniently within reach by motor car from our base, permitting daily visits to 130 Roosevelt Wild Life Annals two or more. The more important of these localities are on Fish Creek, east of Oneida Lake, and, together with others, will he described beyond. Starting with the beginning of the reproductive period it was our purpose especially to follow the seasonal history of certain individual burrows, or groups of burrows, with reference to such questions as rate of excavation, burrow tempera- ture, time and duration of nest building, egg laying and incubation, weight and temperature of both adults and young, and the rate of weight increase and the temperature variations in the young. While these questions were our primary concern, others that might arise in the course of the study were not to be overlooked. In order that something might be learned regarding local and seasonal move- ments of individuals, or of family or other groups, considerable time was spent in banding. The bands were of the regulation aluminum type furnished by the U. S. Biological Survey. During the 1931 season about 1100 bank swallows were banded, and the subsequent recovery of several of these birds casts some inter- esting light on certain phases of the present study. However, the mere banding of a large number of bank swallows was not an end in itself in the present study. The season of 1932 (May 1 to August 15) was devoted almost wholly to the organization, tabulation and analysis of the data obtained during the previous summer and to the preparation of the manuscript of this report. However, it was deemed advisable to spend at least a few days afield, principally in an attempt to recover banded birds and to band additional adults if possible. Accordingly about five days (June 2—1, June 10 and 29 and July 19) were occupied with such efforts. The results of these field trips exceeded expectations. For the privilege of carrying out this project as well as for helpful criticism and timely advice during its prosecution T am indebted to Doctor Charles E. Johnson, Director of the Roosevelt Wild Life Forest Experiment Station. My thanks also are due Mr. Wilford A. Deuce, Ichthyologist and Assistant Director of the Roosevelt Station, for efficient and valuable assistance of various kinds and for numerous courtesies extended during the execution of the field work, the preparation of manuscript and in subsequent activities relating to the issuance of this account. Owing to the nature of the field work which serves as the basis for this report it was necessary to have frequent access to a considerable number of creek banks and sand pits located on private property. Permission to visit such proper- ties was always secured from the landowners before conducting any investigations thereon. In granting permission to enter upon their property, and in conferring other favors as well, the local residents were uniformly obliging and generous. To these persons as a group thanks are tendered, and I wish to acknowledge particularly the generosity and goodwill of Messrs. Bryant, W. A. Beamis, T. Cook, M. Coates, J. Collins, I. Daniels, Bernard Delahunt, Chas. Edell, Geo. Herder, E. H. Onderdonk, Herbert Walker and the Misses Grems. For the determination of most of the insects herein mentioned I am indebted to L. L. Buchanan, E. A. Chapin, H. E. Ewing, W. S. Fisher, C. T. Greene and W. M. Mann of the U. S. Bureau of Entomology and Plant Quarantine, to K. F. Chamberlain of the New York State Museum, and to J. G. Needham of Cornell University. Studies on the Bank Swallow 13] Other persons whose assistance is herewith acknowledged are C. R. Crosby of Cornell University, W. L. McAtee and J. E. Shillinger of the U. S. Biological Survey, H. F. A. Meier, Miss H. Ruth Merrill and L. E. Partelow, New York State College of Forestry, Harold Morrison, U. S. National Museum and W. J. Schoonmaker, New York State Museum. For furnishing certain professional equipment while conducting investigations on the bank swallow in northwestern Iowa, part of which was also employed in the present study, I am indebted to the Graduate College of the State University of Iowa through the kindness of Dean C. E. Seashore; and for the necessary banding and collecting permits, to the U. S. Biological Survey and the New York State Conservation Department. I wish to acknowledge also the courtesy extended by the New York State Museum, through its Director, Dr. Charles C. Adams, in allowing me temporary relief from museum duties in order to complete certain laboratory examinations concerned with this study, and the preparation of manuscript. For valuable assistance in the field, in the compilation of tables and other data, and in the preparation of this report I am under deep obligation to my wife, without whose help so large a series of records would not have been available. THE ONEIDA LAKE REGION For the purpose of the present report, a detailed discussion of the territory here concerned is unnecessary. For a brief consideration of the topographical and certain other physical conditions of the region the reader is referred to the earlier report on the Ornithology of the Oneida Lake Region : With Reference to the Late Spring and Summer Seasons (Stoner, 1932, pp. 284—294). Seasonal Data on Temperature, Humidity and Precipitation. — The climato- logical features of any region influence to some degree the rate of appearance, habits and local behavior of many migratory and resident species of birds occur- ring therein. In the case of the bank swallow such activities as choice of a nesting site, rate of burrow excavation and allied features of early-season conduct are more or less modified by the prevailing climatological factors which, in addition, have some bearing upon burrow temperatures and possibly on the length of incubation period and rapidity of development of young as well as temperature of both adults and young. As a basis for correlating these facts certain climato- logical items which may appertain to the points in question are presented in the following table (No. 36) which has been compiled from the U. S. Weather Bureau's "Climatological Data," New York Section, Volume 43, numbers 5, 6 and 7 for May, June and July, 1931, the season when the field work for the present study was undertaken in the Oneida Lake region. The records were made at the U. S. Weather Bureau station in Syracuse, New York, eleven miles southwest of Oneida Lake and 599 feet above sea level, or about 228 feet higher than the lake itself. In the Oneida Lake region the mean monthly maximum temperatures for May, June and July, 1931, were respectively 68.5°, 76.8° and 84.2°. Mean monthly minimum temperatures were respectively 49.9°, 56.8° and 66.5°. Roosevelt Wild Life Annals Table No. 36. — Climatological Data for Oneida Lake Region. May, June and July, 19.31. (Syracuse Station — U. S. Weather Bureau.) MONTH Temperature in Degrees Fahr. Precipitation in Inches Number of Days Prevail- ing direc- tion of wind Mean Highest Date Lowest Date Total Great- est in 24 hrs. With precipita- tion of 0.01 inch or more Clear Partly cloudy Cloudy May 59. 2 90 29 34 1 4.42 0.88 15 11 7 13 S. June 66.8 96 20 45 2 1.15 0.58 9 14 10 6 NW. July 75.4 100 2 60 12 3.81 1.11 14 6 15 10 S. A table (No. 37) showing climatological data for the Lake Okoboji region, northwestern Iowa, also is included here. It has been compiled from the U. S. Weather Bureau's "Climatological Data," Iowa Section, volumes 36, 37 and 38. All of the Iowa weather stations lie at some distance — five to fifty miles — from the localities where the actual field work was done, and it may be assumed that the temperature and other climatic conditions at these various weather stations and the corresponding field stations differed somewhat on the days that the observa- tions were recorded. In addition, it is probable that the lakes themselves (Oneida and Okoboji) exert some local influence on the temperature in their immediate territory. However, for the purpose here intended, the records presented will serve, for only correlations of broader scope and significance will be attempted. Table No. 37. — Climatological Data for Lake Okoboji Region. June and July, 1925, 1926 and 1927 (Spencer Station.) Temperature in Degrees Fahr. Precipitation in Inches Number of Days Prevail- ing direc- tion of wind YEAR Month Mean High- est Date Lowest Date Total Great- est in 24 hrs. With precipita- tion of 0.01 inch or more Clear Partly cloudy Cloudy 1925 June. . . 69.7 95 6 41 28 3.55 9.67 11 10 16 4 SE. July . . . 74.5 103 15 41 22 1.75 0.91 7 16 15 0 NE. 1926 June. . . 65.8 99 11 39 4 3.53 1 .40 7 16 10 4 NW. July . . . 74.8 104 20 45 13 6.45 2.60 11 12 10 9 NE. 1927 June. . . 65.8 98 28 37 5 1.84 0.65 9 18 7 5 sw. July.. . 71 .8 96 27 49 23 (I 47 0.34 4 17 13 1 sw. The mean monthly maximum temperatures in the Lake Okoboji, Iowa, region during the progress of the part of the investigation here reported were as follows : For June and July, 1925, respectively, 79.2° and 84.5° ; for July, 1926, 86.9° ; for June and July, 1927, respectively, 76.7° and 82.4°. Mean monthly minimum temperatures for June and July, 1925, were respectively, 54.4° and 56.5° ; for July, 1926. 62.4° ; for June and July, 1927, respectively, 52.9° and 56.9°. 133 Fig. 24. View upstream of site of bank swallow colony, south bank of Fish Creek at Fish Creek Landing. June 10, 1932. 134 "•fJJK Site of bank swallow colony, north bank of Fish Creek at Fox farm, northeast (upstream). July 23, 1931. Looking Fig. 26. Section of north bank of Fish Creek at Fox farm showing group of closely spaced bank swallow burrows, and also the way in which burrow entrances become enlarged toward the close of the season through erosion and the frequent passage of occupants. July 23, 1931. Studies on the Bank Swallow 135 Upon comparing the climatological data it will be noted that the mean monthly temperatures for June and July in the two regions (Oneida and Okoboji lakes) under consideration were very similar; the greatest departure is that for July, 1927, and July, 1931, where a difference of 3.6 degrees F. occurs in the monthly mean for the two regions. Somewhat greater differences occur in the corresponding daily maxima and minima for the two regions, but in these records, also, the similarities are more obvious than the differences. The two territories are sepa- rated by about 950 statute miles, but lie in nearly the same latitude, the approximate position of Oneida Lake being 43° 8' N., and that of Lake Okoboji, 43° 35' N. DESCRIPTIONS OF FIELD STATIONS REPRESENTING TYPICAL BANK SWALLOW HABITATS IN THE ONEIDA LAKE REGION The bank swallow may be found in many other localities throughout the present territory, but the principal nesting places and concentration points which I have found are near the east end of Oneida Lake, along the banks of Fish Creek and its tributary, Wood Creek (Map 5). In the hilly sections about McConnellsville and Humaston (Fig. 36), and near the village of Cleveland, are rather extensive sand pits the walls of which attract varying numbers of bank swallows, as do also the low but precipitous banks of a few small creeks in the flat country north and east of the lake ; but by far the greatest concentrations occur along the higher banks of Fish Creek which flows through the low, flat land northeast of Oneida Lake. In the western part of the territory concerned in this study, the principal nesting sites of the bank swallows were a large gravel pit on the south side of Oneida River, near Schroeppel's Bridge, and a sand pit on the Dutcher farm, nearby. Fish Creek follows an exceedingly tortuous course and is a fairly rapid stream. Considerable erosion occurs at or near the sharp curves, and the seeping water so softens the banks from below that they continually slump away. During the course of these studies a number of bank swallow colonies were visited at fairly regular intervals throughout the breeding season. Some of the sites were much alike in their essentials, but considered as a group they represent several sets of conditions and brief descriptions of the typical ones and their immediate surroundings seem desirable. The location of the principal bank swallow colonies visited, together with certain other information, is shown on Map 5. Each site, or field station, bears a number which corresponds with that in the legend, and following this, the name of the property owner is indicated in order that the position of the colony may be further identified. Station I. Herder Farm, South Bank of Fish Creek. — At this site, which is about two and one-half miles, air-line, northeast of the village of Sylvan Beach, occurs one of the largest bank swallow colonies in the entire territory (Figs. 23 and 24). In 1928 this was one of the most populous single communities that I had ever seen ; but here as elsewhere, the number of birds fluctuates considerably from year to year. While this colony was much reduced in 1929, it attained something like its earlier proportions in the seasons of 1931 and 1932. Roosevelt Wild Life Annals This colony is directly opposite the little village of Fish Creek Landing, where the creek is about forty yards wide, and the south hank rises to a height of 5 to 10 feet, depending upon the stage of the water. The north bank of the creek directly across from the colony presents a low, sandy, willow-clad flat. A pasture extends south and west from the sheer banks occupied by the swallows, while throughout the entire surrounding low, flat country are open cultivated fields and pastures. Several farm buildings are in the immediate vicinity. The bank here is composed of fine sand, and portions of it slump from time to time, sometimes carrying with them nests and eggs as well as adults and young of the bank swallow. The mortality from this cause assumes considerable pro- portions. Early in the season, too, after the birds have begun to excavate their burrows, their labors are often prolonged owing to constant erosion which is accelerated by the usually copious spring rains. While the sandy banks are constantly falling away, the expanse of beach immediately in front of the colony remains fairly uniform, for the sand that has fallen is soon washed away by the stream. Hundreds of burrows are excavated here each season by the swallows and, in addition, a few old burrows which may have come through the winter are repaired and reoccupied. Groups of as many as twenty to thirty or more burrows are sometimes found in one small area of the vertical bank, leaving adjoining and apparently equally acceptable sections sparsely occupied or vacant. This concen- tration of burrows with its attendant honey-combing effect is one of the principal contributory factors to the slumping of the banks, particularly during rainy weather. Although grazing sheep and cattle frequent this bank of Fish Creek, and fishermen often visit the shores, guide their motorboats or row up and down the stream past this site, these activities seem in no way to frighten or discourage the swallows. Indeed, so far as my observations go, the birds appear rather to favor a site in the vicinity of intermittent activities of such nature. On May 2, 1931, along approximately 600 linear feet of bank here, I found evidence of a total of 244 burrows. Most of these were old ones, but some of them still remained open to a depth of more than three feet. Others represented burrows just started that season, while still others were evidently the terminals only of old burrows, left after slipping of parts of the bank. By May 12, the number of burrows at this site had increased to 301 ; by May 14, to 327; and by May 19, to 378. So far as could be determined, the number of burrows in the colony did not increase after the last mentioned date. Station II. Mott Farm, North Bank of Fish Creek. — A quarter-mile north of the Fish Creek Landing colony Fish Creek makes a sharp bend. The waters of the stream rush against the north bank at this bend, resulting in continuous erosion and a nearly vertical exposure. The section of bank which attracts the greatest number of bank swallows rises from five to nine feet above the average level of the water (Fig. 34). Topping the bank is a heavy sod. A fine loose sand-loam facilitates the excavation of deep burrows, on the part of the swallows, but also results in excessive erosion of the bank. An adjoining field is ordinarily planted to corn, while in the general vicinity are other fields, pastures, groups of farm buildings, and the little village of Fish Creek Landing. Almost daily through- Studies on the Bank Swallow 137 out the summer fishermen visit this or immediately adjoining parts of the bank, or pass by in rowboats or motorboats. At no time during the course of my field work in this section did I find this colony very populous. Observations made here on May 7, 1931, on 100 linear feet of hank, showed the presence of but nineteen hank swallow burrows, and only one of these appeared to have been freshly worked. The others were old burrows which had remained more or less intact from former seasons. A visit to this bank five days later revealed the presence of 12 to 15 swallows, and considerable activity in excavating, but although some further work was done on burrows that had been started, no new excavations were made during that season, and the colony remained about stationary in size. It is possible that too frequent disturbance of the birds by fishermen, the fine loose sand that results at times in partial or total collapse of the burrows, or a too rapid wearing away of the banks are factors limiting the size of this colony. Station III. Fox Farm, North Bank of Fish Creek. — One of the largest and most typical colonies of bank swallows in the territory occurs here. It was, fortunately, also easily accessible, so that we visited it at regular and frequent intervals ; in fact, a considerable part of our results were secured at this station. At the site of the principal part of this bank swallow colony is a bend in the creek, so that the stream flows against and constantly cuts away the north bank which here attains a height of 10 to 15 feet above normal water level. The steep- ness of the bank of mixed sand and clay is maintained through the action of the current which carries away rather quickly much of the eroded material. The south bank of the creek here is low, and a broad sand beach slopes gradually to the water's edge. Here as at many other places in this territory a rather rank growth of grass occurs along the immediate borders of the bank, while a little farther away lie cultivated fields and pastured land, the last named in some places including the creek border. A much traveled paved highway parallels Fish Creek which at this point is 30 to 40 yards wide. The nearest farm buildings are about half a mile distant. The section of the bank which has the greatest number of nesting swallows is a bend about 200 yards long (Fig. 25). The bare face of the bank with its numerous burrow entrances has an east and south exposure, so that most parts of the colony have more hours of direct sunlight throughout the breeding period than they would otherwise have (Fig. 27). This results in a higher and also a more uniform burrow temperature. Although the burrows of this colony are distributed over a considerable part of the bank section concerned, there are two centers of concentration, one at the north end and the other near the south end. Between these the burrows occur more sparsely. On May 14, 1931. at the most populous part of this colony, on a section of the bank 250 feet long, I counted 226 excavations one inch or more in depth. At that time some of these shallow excavations were the beginnings of future inhabited burrows ; others, which were later abandoned, apparently represented sporadic attempts at digging prior to the full onset of the nesting urge. There were also 138 Roosevelt Wild Life Annals burrows of a former season, or seasons. By May 22, the number of swallows attracted to this colony had been considerably augmented, and on that date I counted 460 excavations one inch or more in depth on the same 250-foot extent of bank. On May 29, within the same area, I counted 498 excavations. Due principally to continual erosion and periodic slumping- of parts of the bank, burrows in all stages are completely obliterated from time to time, so that even though the digging of new and the deepening of old ones continues, the number of countable burrows may actually decrease as the season advances. For example, on June 10, 1931, the number of burrows on the 250-foot section of bank which on May 29 contained 498 excavations, now had only 468. A count on July 11. following, of the number of excavations two inches or more in depth, gave a total of 526. This was the greatest number of burrows recorded for any bank area of similar extent. Many of the burrows here were rather widely spaced in a single, more or less irregular series (Fig. 40). Sometimes two or three short rows of burrows were constructed one above the other. Again a section of bank was closely honey- combed, onlv two or three inches of loose sand wall separating adjoining burrows (Fig. 26). ' As the season advances vegetation of different kinds takes root about, and even invades, the entrances to the burrows. The plants most frequently found in such situations are the field horsetail (Equisetum arvense Linn. ) and the red sorrel [Rumex Acctosclla Linn.). Sometimes these plants nearly hide the open- ings of the burrows before the young of the first brood are ready to leave the nests. By late July, also, a considerable amount of turf from the top of the bank has slid to the base of the declivity and here forms thriving patches of grass. Station IV. J. Cook Farm, South Bank of Fish Creek. — This station lies about two miles northeast of Fish Creek Landing. In general the nesting site here is similar to that at the Fox farm and at Fish Creek Landing. There were few bank swallow burrows at this station early in the season, and no banding or other work was attempted at the colony later established here, which occupied several scores of burrows at one particularly favorable section of the bank, and scattered ones in neighboring sections. Station V. Daniels Farm, North Bank of Fish Creek. — A fair-sized colony here inhabits the precipitous sandy banks of Fish Creek, less than 100 yards from a State road. At the site of the main part of the colony the bank rises to a height of ten feet above normal water level. The bank faces to the east. Farm buildings are situated a hundred yards away, and the adjoining lands consist of cultivated fields and pasture. The greater part of this colony is confined to about a fifty-foot length of bank. In some parts the burrows are arranged in single rows, in other places series of two or more fairly distinct rows appear. In the heart of the colony this serial arrangement disappears and the burrows present a honey-comb arrangement. On May 7, 1931, I counted fifty-eight burrows here. Most of these were old ones, used in preceding seasons, and varied in depth from 18 to 24 inches. Since the average depth of bank swallow burrows is about 28 inches this would indicate Fig. 27. Site of bank swallow colony, north bank of Fish Creek at Fox farm. July 11, 1931. Fig. 28. Bank swallow burrows in sand-clay soil, north bank of Fish Creek at Fox farm. Lumps of slumped turf at foot of vertical bank. July 23, 1931. 140 Fi°- 29. Site of bank swallow colony, south bank of Fish Creek at Edell farm. Looking north- ° west (downstream). Creek at low stage. Current continually cuts away south bank of stream and deposits sand on the north shore to form a low, flat beach. June 24, 1931. Studies on the Bank Swallow 141 that erosion of the bank here does not proceed so rapidly as at some other colonies along Fish Creek. On May 14, 1931, there was evidence of 96 excavations, indicating consider- able activity during the seven days since our earlier visit. Subsequent counts showed that few if any new burrows were started in this colony after May 14, although work in some of those already under way was continued after that date. Not all the burrows found in the colony were eventually occupied, even many of the new ones being abandoned before completion. A few scattered burrows occurred on each side of the main section of the colony, and about fifty yards northward was found a group of ten. Stations VI and VII. Chas. Edell Farm, South Bank of Fish Creek.— In the summer of 1931, Station VI had one of the most populous bank swallow colonies in this section. Although two separate colonies were present here, our studies were made principally at the larger one (Station VI). Fish Creek here makes a rather sharp bend, and the two colonies thus are separated by a neck of land about 200 yards wide. About 300 yards upstream, on the north bank of the creek, is the colony on the Daniels farm described above as Station V. At the smaller colony (Station VII) the creek flows in a south-southeasterly direction, so that the force of the current is spent against the sand-clay bank, resulting in a vertical wall varying in height from about five to nine feet. Each season a section of 300 to 600 feet of this bank attracts a considerable number of swallows. The bank faces westward and thus is exposed to the warm afternoon sun. There is no marked concentration of burrows in any particular part of the bank, and the excavations occur in a more or less regular, horizontal and linear arrangement. Cultivated fields and pasture also adjoin this section of the creek. The larger swallow colony (Station VI ; Fig. 30) lies in a broad cove where the sand-clay banks rise from 10 to 15 feet or more above the average level of the stream (Fig. 31). A hay field adjoins and extends across the neck of land to Station VII. The opposite (north) bank is low and presents a broad sandy beach containing a growth of willows and other vegetation. The nearest farm buildings are a quarter of a mile distant (Fig. 29). While the major part of this colony (Station VI) occurs within a 1000-foot length of bank, the burrows in some parts are irregularly and sparsely distributed ; sometimes separated by 10 to 20 or more feet. In others a marked concentration occurs, the bank being literally honey-combed with burrows. Although some sections of the bank appear to be more resistant than others, the constant weathering and the undermining influence of the stream results in periodic slumping with accompanying destruction of bank swallows and their habitations. At the time of our first visit to this colony. May 7, 1931, very few bank- swallows were seen in the vicinity and there was no evidence of fresh diggings, although numerous burrows of the preceding season were found. A week later, a few bank swallows were at the site, and work on the burrows had started. Some of the old burrows were being reconditioned, and several new ones from 3 to 8 inches in depth had been dug. However, even at this time there was less activity here than in some other colonies. On the 1000-foot bank I counted 113 burrows, old and new, two inches or more in depth. 142 Roosevelt Wild Life Annals By May 27 this colony had increased considerably in size, a good many nests had been built and egg-laving was in progress. On this date burrows two inches or more in depth, on the 1000-foot bank, numbered 228. On June 9, on the same area, the number had increased to 267. Station VIII. E. H. Onderdonk Farm, South Bank of Fish Creek.— A small colony occurs here in the sandy south bank of Fish Creek, about three and a quarter miles, air-line, northeast of Fish Creek Landing. The creek bank rises about eighteen feet above the level of the water, and its face is nearly vertical. It is in a pastured field. The principal part of this colony is included within a length of about seventy- five feet of bank. This colony appears to vary but little in size, from season to season, and is representative of the numerous smaller colonies scattered along Fish Creek and other similar streams in the region. Station IX. W. P. Graham Farm, South Side of Fish Creek. — This colony is about 300 yards northeast of the bridge across Fish Creek, known locally as Herder's Bridge, and about four miles northeast of Fish Creek Landing. In many respects the site of this colony is similar to that at the Edell farm (Station VI). At this point in its course Fish Creek flows almost due north for about one-fourth mile. A short distance upstream from the site occupied by the swallows the current is deflected sharply by a point of land extending into it from the north. As a consequence the south bank of the creek has received the full force of the stream along this quarter-mile stretch, with results that are plainly evident (Fig. 33). The north hank of the creek is low and presents a strip of sandy beach. The south bank rises to a height of 15 to 20 feet above the water. Most of it has an east and southeast exposure. In places it has slumped badly, but in others its smooth bare exposures have resisted both weathering and stream action to a marked degree (Fig. 41). As the season advances some of the piles of slumped earth become covered with a rank growth of grass, weeds and other vegetation ; and field horsetail and red sorrel, as in places previously mentioned, have found lodgment on the exposed vertical banks and partly conceal the openings of some of the inhabited burrows. Fishermen frequent the banks here almost daily, and cattle graze to the edge, but the swallows are seldom disturbed by either of them. The nearest farm buildings are about half a mile distant. The burrows were variously distributed, some being separated by no more than two or three inches. On May 15, 1931, on a stretch of 175 feet of bank where the swallows had begun to congregate, I counted 125 burrows in various stages of construction. By May 21 the number of burrows in this sector had been increased to 137: and hut few new ones were excavated after that date. However, toward the end of June, after first-brood young had left the nests, excavating activities were renewed and continued for a short period, but for the most part were confined merely to deepening or restoring old burrows. On June 26, when the number of burrows had reached the maximum, I counted 260, both old and new, in the entire colony. Although at no time during the summer were all the burrows occupied, a conserva- tive estimate of the number of nesting bank swallows in this colony at the height of the season would be 140 to 160 pairs. Studies on the Bank Swallow 143 Station X. Grems' Sand Pit, One-fourth Mile South of McConnellsville. — A still different type of nesting site to which hank swallows are attracted in considerable numbers, locally, is represented by more or less irregularly worked sand pits. For some reason certain sand pits in the territory seem to be avoided by these swallows, as nesting sites, while others, apparently no more advantageously situated or offering greater attractions, are regularly inhabited. One of the larger sand pits in the territory lies in the hilly section about a quarter-mile south of the village of McConnellsville, and is known as Grems' sand pit (Figs. 35 and 36). A railroad spur from the main line extends to this pit, and several carloads of sand are loaded here in the course of the summer. This huge pit is in the side of a hill. It is about 450 feet in circumference and the sides vary in height from 15 to almost 100 feet above the floor of the pit. The sand deposit is covered by a thin but rather resistant surface layer of mixed clay and loam in which not only gravel and small pebbles but even larger boulders are imbedded (Fig. 37). It is in this rather shallow layer that the bank swallows dig their burrows, which are distributed in groups of 3 or 4 to 8 or 10. As the sand is removed from the bottom and sides of the pit the walls at times slump, and may then carry with them bank swallow burrows and their occupants. Few burrows remain intact to the following season. The land surrounding the pit is mostly cleared and mainly devoted to pasture. The nearest stream is the West Branch of Fish Creek which also flows through McConnellsville village (Fig. 36). A visit to this pit on May 4, 1931, revealed no bank swallow activity, but many of this species were seen coursing along the creek at McConnellsville, together with other swallows. At that time the sand pit was not being worked. On May 11 a considerable number of bank swallows were flying about the sand pit, and between 25 and 30 burrows had been started. At the height of the breeding season (early June) there were between 60 and 70 occupied burrows. Little excavating was done subsequently, but following the departure of the first brood of young some new burrows were constructed and a few old ones repaired, although, because of the breaking of the banks, the number of existing burrows at no subsequent time during the season reached that attained in early June. Stations XI and XII. Bryant's Sand Pits. — About a mile northeast of McConnellsville, at the margin of an extensive hilly and more or less wooded section, are two deep excavations known as Bryant's sand pits. The larger of these is no longer worked, but from the other, a quarter-mile away, sand is removed in considerable quantity at intervals throughout the summer. Bank swallows frequent both these pits, but the number of nesting pairs at the one that is worked is considerably greater than at the larger but unused one. As at Grems' sand pit, the burrows here are limited to the firmer sand-clay-gravel layer overlying the sand (Fig. 38). The sides of the unworked pit are composed of gravel and boulders, making it rather difficult and sometimes even impossible for bank swallows to complete their burrows. No doubt this condition accounts for the small number of swallows nesting at this pit. A short section of the sandy loam layer at the top of the west 144 Roosevelt Wild Life Annals side of the pit is the only portion of the bank that is habitually inhabited by the birds. On May 4. 1931, a lew birds were flying' about this pit; several old burrows remained from a previous season; evidences of fresh workings were present in some of these but the efforts of the swallows here lacked the seeming spontaneity of purpose which usually characterizes the birds later in the season. Apparently they were not yet firmly domiciled. By May 11 several additional burrows had been excavated, but the breeding stations appeared to be taken up less rapidly than those in the banks along Fish Creek, 3 to 5 miles away. Neither this colony nor the one at the new pit came to be represented by any large number of individuals during the season, so that these pits were not visited as frequently or studied as intensively as were some of the more populous colonies in the banks of Fish Creek. Whether the immediate presence of water in the latter situations may be the cause for their popularity as nesting sites I do not know, but some factor, either of site or surroundings, appears to render them more desirable than the sand pits. The operating Bryant pit is about 300 feet in length, and its steep sides rise from 20 to more than 60 feet. Intermittent removal of considerable amounts of sand causes almost constant slumping of the banks. On May 4, 1931, I found no birds or burrows at this pit, but by May 18 several burrows were under con- struction. Not more than thirty burrows were found here at any one time, even during the height of the season. Here as in most other colonies, the swallows concentrated their excavations largely at two places; only a few scattered burrows occurred between these groups. Station XIII. Humaston Sand Pit. — This was the most extensive sand pit noted in the region. Here the walls of a yawning cavity roughly 600 feet square were presented for bank swallow occupancy, but the isolation of the pit and its present unworked condition probably were responsible for its lack of tenants. A number of other sand and gravel pits and sections of creek banks were examined at various times during the course of this study, but with the exception of the Delahunt sand pit mentioned below, none of these was visited regularly. Delahunt Sand Pit. — This sand pit is situated a mile northeast of Cleveland and is operated intermittently throughout the summer season. The sand is of a fine grade and is used in the manufacture of glass. The pit covers a considerable area adjoining a cut-over section now supporting a low second-growth. The loca- tion is poorly drained, so that a considerable part of the pit is constantly flooded with water. This necessitates the use of a dredging outfit for the removal of sand. The banks of the pit are nowhere over ten feet high and in most parts rise only three or four feet above the water. Owing to the fineness of the sand as well as to the effects of dredging opera- tions, few if any bank swallow burrows here last more than one season. While more distant parts of the pit harbored a few swallows, the principal concentration was within a few yards of the dredging outfit, where a group of 12 burrows was found (1931 season). These burrows were excavated between May 7 and 12, but Fig. 32. Bank swallow burrows, south bank of Fish Creek at Edell farm. Sand-clay soil. Outer ends of some burrows directed slightly upward, others slightly downward. June 4, 1932. 146 Fig. 34. Site of small bank swallow colony at Mott farm, north hank of Fish Creek, one-fourth mile north of Fish Creek Landing. May 12, 1931. Studies on the Bank Swallow 147 not all were later occupied. The birds seemed rather indifferent to the dredging activities in their midst. Gravel Pit, Oak Orchard District. — Some thirty miles west of the above described stations is a large roadside sand and gravel pit, intermittently operated. It lies but a few yards from the banks of Oneida River. Its main wall is semi- circular, faces approximately northwest, and is from 25 to 50 feet high (Fig. 39). Nesting possibilities are limited to a layer of loam-clay which overlies the gravel to a depth of several feet. In the 1931 season this pit contained about 100 bank swallow burrows, distributed along this layer. Dutcher Farm Sand Pit, Oak Orchard District. — About a quarter-mile northeast of the gravel pit just mentioned is a sand pit perhaps 100 feet long and 10 to 15 feet deep, situated in an open pasture on the Dutcher farm. It had not been recently operated, and according to local information there are fewer bank swallows nesting at this pit now than when it was in operation. Other small sand and gravel pits in the vicinity of McConnellsville, Vienna, Jewell and Cleveland offered seemingly attractive possibilities for bank swallows, but relatively few of the birds were present. The greatest concentration of these swallows in the nesting season in the Oneida Lake region, are the precipitous sandy banks of Fish Creek and similar streams. (For a consideration of the conditions under which the bank swallow occurs and breeds in the Lake Okoboji, Iowa, region, the reader is referred to the writer's report of 1925 : The Auk, 42, No. l,pp. 86-94). THE LOCAL ABUNDANCE, DISTRIBUTION AND SEASONAL OCCURRENCE OF THE BANK SWALLOW IN THE ONEIDA LAKE REGION In the Oneida Lake region, bank swallows usually begin to arrive late in April, and maximum abundance is attained by late May. Occasional cold waves doubtless cause considerable hardship among the early arrivals, by reducing the activities of insects upon which these birds depend for food. In May, after weather condi- tions have become more settled, well-marked waves of bank swallow arrivals are frequently noted. On May 2, when our field observations in 1931 began, only about a dozen bank swallows were seen in the vicinity of Fish Creek Landing, and but little digging had been done in the creek banks. Similar conditions were noted at other places along Fish Creek as well as at the Grems', Bryant and Delahunt sand pits. Mr. Delahunt stated that a few birds had appeared at his place about April 25, but he had seen none since that time. The sand pits appear to be less popular with the birds early in the season, possibly because of the absence of water in the near vicinity, and a consequent scarcity of flying insects. Many of the insects taken by the birds are aquatic in their larval stages and are among the first to complete their metamorphosis in the spring ; others, perhaps, in the moister and more sheltered situations of the creek banks, recover from their winter dormancy a little in advance of their kind in drier and more exposed situations. There is some indi- cation that these swallows, early in the season, center their activities also about farm buildings, doubtless because of a relatively more abundant food supply in the form of flies and other insects about barnyards and stock enclosures. 148 Roosevelt Wild Life Annals Until about .May 10 the bank swallows arc wary, and their dicing activities random and sporadic. .Much of the time is spent in flying about, feeding, perhaps; but from the first of May to about the 25th, burrow excavation is the dominant activity, reaching its maximum intensity usually between May 15 and 22. Previous to the onset of nesting activities proper, new arrivals of bank swallows add to the local population. The birds course over the streams, ponds and lakes, often in company with barn, cliff and tree swallows, uttering their low harsh twitter. Sometimes a flock of 75 to 100 individuals may congregate in a favorable feeding place, then disperse, and after a short interval unite again. This behavior is a marked feature of the bank swallow at this period. But daily, now. local groups begin to assemble about sand pits and at nesting sites along the banks of streams. Fresh diggings begin to appear where a few days previously such signs were lacking. Breeding stations have not yet been taken up definitely by any considerable number of birds, more of which are arriving daily. By way of illustrating the sporadic occurrence and looseness of local movements of the bank swallow, a specific example will serve. On May 12, 1931, I found three new bank swallow burrows at the Delahunt sand pit, which Mr. Delahunt told me had been constructed on May 9, when "a dozen or fifteen" of the birds had visited the locality. Only an occasional swallow had been seen since that time. One of these burrows was 3l/2 inches deep, the second 26, and the third 31 inches. After the burrows have been dug, nest-building proper begins, and this is carried on. in this territory, mainly between May 16 and 25. Blades of grass, weed stalks and other material may then be seen partly obstructing the burrow entrances. As with most other birds, the wariness of bank swallows decreases with the progressive development of the reproductive period, reaching its lowest phase in the late incubation stages and during the time that young are in the nest. At the Fish Creek Landing colony an interesting trait of the bank swallow was observed on May 12, 1931. At intervals between their diggings some of the birds lit on a strip of sand beach at the water's edge below the burrows and appeared to pick up something in their bills. Whether it was food or not I could not be certain, but the presence of fragments of limestone and coarse gravel in the stomachs of several adult bank swallows collected here and at other places later in the season, probably explained these visits. Nest-building as well as burrow-digging is accompanied by much harsh, unmusical twittering. Thereafter comparative quiet reigns during the period of egg-laving and incubation ; vocalization is resumed as soon as the young are hatched. When alarmed, the bank swallow gives a shrill "ke-a-g-h" or "te-a-r-r," which, though much feebler, reminds one of the note of the common tern. Apparently this is a warning note. When uttered by a swallow in flight as, for example, a frightened individual leaving the nest, or the bander's hand, it acts as a signal to other members of the colony, and a general rush of the birds from their burrows in that immediate vicinity ensues. As group after group takes up the alarm, the prospects of the bird-bander rapidly decline. When captured and held under restraint in the hand, some adult individuals utter a musical but exceedingly high-pitched sound, barely audible to my ear. And Fig. 35. South side of Grems' sand pit, one-fourth mile south of McConnellsville. Bank swallows in the shaded layer of earth, sand and boulders at the top. May 18, 1931. 150 Fig. 37. Bank swallow burrows, east side Grems' sand pit, near McConnells- ville. May 26, 1931. Fig. 38. Bank swallow burrows at Bryant's "old" sand pit. The groove below the burrows marks the boundary of Field Experiment No. 2 ; the tags suspended on wires above two of the burrows mark those chosen for special observation. May 11, 1931. Si ud ics on I he Hank Swallow 151 it may be remarked here that the notes of young bank swallows of flying age are somewhat lower and coarser than those of adults. Between May 15 and July 22, in a few bank swallows (8 males and 13 females) which were secured for dissection, the developmental state of the reproductive organs was found to reflect in a general way the degree of activity displayed in burrow-digging and nest-building. (Table 38.) Table No. 38. Measurements of Testes and Ova of Adult Bank Swallows. Oneida Lake Region, 1931. Greatest Diameter (Mm.) U'piitlit vv cigxi i FV a TT? sex fCl m c 1 Remarks Right Lei i Largest second Third Fourth ^umS. } testis testis egg egg egg egg Mav- lS m 7.0 9.0 15.7 f 8 6 3 9 lO, 5 About 40 additional e;^gs, about 1 mm. in diameter m 6.0 7.5 18 f 13 5 3 -3 14.7 Several smaller eggs m 6.0 7.0 13.5 19 f 14.5 Eggs all very small f 16 9 7 4 17.1 Another egg 2 mm.; several smaller ones 21 f 17 8 5 2 Largest egg about ready for deposition; several smaller 26 f -1 -1 -1 - 1 13.4 Eggs very small; ovary reduced in size; egg-spent 27 m 6.0 2.0 12.7 June 2 f 18 -1 - 1 - 1 13.9 Incubating and evidently ready to deposit last egg 4 f -2 -2 -2 -2 14.5 Incubating; egg-spent 9 f 2 1 1 1 15.5 Incubating; egg-spent 10 m 7.5 8.0 14.3 Testes large for a late-breeding bird 16 f -1 -1 -1 - 1 Incubating; large number small eggs in ovary 24 m 4.0 5.5 14.2 Incubating 4 eggs m 6.0 6.5 13.6 Incubating 3 eggs; nest with heavy feather lining 29 f 2 2 2 2 15.4 12 eggs 2 mm.; several smaller July 7 f - 1 - 1 -1 -1 14. 7 Egg-spent; ovary much reduced m size 9 m 1 .0 2.0 13.5 Brooding 2 young about 3 days old 22 f -1 -1 -1 -1 Brooding 2 young 4-6 days old; ovary and ova very small By the last week of May most of the nests have been completed, and egg- laying has begun. Meantime a feather lining has been added to the nest. The appearance of the first young marks the beginning of a new activity. (The peak of abundance of young in the nests comes in the period between June 15-25.) For two or three days after hatching, one or the other parent broods the young almost constantly, but after this period most of the daylight hours are spent by the parents on the wing, feeding and carrying food to their young. From what we were able to learn on our nocturnal visits to bank swallow colonies, in both the Oneida Lake and the Lake Okoboji regions, neither parent spends the night in the burrow after the young are Avell covered with feathers. Although the question was not satisfactorily answered during our study, the young apparently attempt their first flight eighteen to twenty-two days after hatch- ing. In general, those young that were reared in short burrows so situated as to receive a maximum of direct sunlight, spent a shorter time in the nest than did those in burrows of similar length but less exposed to the sunlight. The same 152 Roosevelt Wild Life Annals was noted for occupants of burrows a few inches below tbe surface as compared with those in burrows several feet below it. (The temperature differences in these nests are shown in Table No. 45.) It is not improbable, however, that the time of departure of the young from certain nests was hastened by the disturbance we caused. If the young of a given family were of uniform size they usually left the nest on the same day; but in cases where there was some notable difference in size the larger birds left first. Even in their first flights young bank swallows show remarkable control and maneuvering ability, but they lack endurance. I have often watched the birds fly surprisingly well for a few minutes, circle about near or above the burrows which they had just left, then evidently become fatigued and alight on the ground to rest for a few seconds before resuming flight. I have also seen apparently exhausted young birds light for a few seconds on the smooth surface of Fish Creek, and then suddenly and without apparent difficulty rise and continue flight. By July 1, about four-fifths of the young of the first brood have left the nest. A large proportion of the adults leave their nesting sites, but signs of mating and preparations for rearing a second brood are evident among some of those that remain. With the young now fairly well able to shift for themselves, but still attended to some degree by the parents, the latter become much more wary and spend less time in and about the burrows. A few adults mate, a second clutch of eggs is laid and a family of young is reared. Some fresh burrows are dug, and other rather desultory and purposeless excavating is attempted, but it lacks the energy and per- sistence which characterize this activity earlier in the season. Preparation for the second brood consists mainly in repair of the previously used quarters. Between Tuly 17 and July 23, 1931, intermittent trips were made to practi- cally all of the field stations which, earlier in the season, we had visited frequently and regularly. All the earlier colonies were then practically deserted. Occasionally a few adults, and young of the year, flew about the nesting sites, but they did not stop. At the Fox farm, on July 17, more than 200 burrows were examined. Of these, one was occupied by two immature birds, while a half dozen others were inhabited by incubating or brooding adults. An egg in one of the nests was opened and found to be within 3 or 4 days of hatching. In another nest were nestlings about twenty-four hours old. Similar conditions were found at the other large colonies visited during this period, supporting the view that comparatively few bank swallows rear a second brood in one season. On August 12 a round of visits was made to our principal field stations. At neither Grems' nor Bryant's sand pits was there evidence of recent occupancy. Many of the burrows had been destroyed by slumping of the bank; others were more or less filled with sand. Most of the nests were damp, and all of the several examined contained fleas ; one nest also contained a large number of flea larvae. Another old nest contained forty-three empty dipterous puparia. Only three swallows were seen in flight at Grems' pit, and none at Bryant's. Among the scores of burrows examined at the Fox farm, only one was occupied ; it contained three young about ready to leave the nest. None of the other burrows showed evidence of recent occupancy. Considerable erosion of the Studies on the Bank Swallow 153 face of the bank had taken place since our last visit. Within the period of nearly an hour spent at this station, less than a dozen bank swallows were observed in its vicinity. At Fish Creek Landing an occasional bank swallow flew overhead, and a half dozen sat on the high tension lines that crossed the creek. Four weeks earlier I counted over 500 individuals on these wires at one time. There was no evidence that any of the burrows had been inhabited within several days. Fleas, both larvae and adults, were plentiful in the bottom debris of all the nests examined. I !arn and tree swallows were fairly common on roadside wires and about Fish Creek, but bank swallows were few. The following will summarize briefly some of the main facts observed with regard to the reproductive season : The first eggs were found on May 19, freshly laid eggs as late as July 13, and incubating adults as late as July 20. The first young were found on June 4 ; and newly hatched young as late as Julv 17 (July 19, 1932). Adult birds were observed in copula as late as July 12. The bank swallow begins its southward movement earlier than its near rela- tives. Even by mid-August a considerable diminution in its local abundance is noticeable, and by September 1 comparatively few individuals remain in the Oneida Lake region. WEIGHT STUDIES OF BANK SWALLOWS Not much seems to be known regarding the weights of our native passerine birds, and one of the objects of the present study was to secure some information on this subject with reference to the bank swallow. For this purpose a single- beam laboratory balance was used, fitted with steel knife edges and agate bearings, and sensitive to one-tenth of a gram. A wooden box was constructed in which the balance could be carried in the field and in which it could be operated without being affected by the wind. Weight of Adults. — The weights were obtained for 249 different adult bank swallows. Seventy typical examples have been assembled in Table No. 39. For most of these birds only a single record is available. However, for several indi- viduals a second record was secured, and one individual was caught and weighed three different times. Since this table also includes temperature readings on these birds, along with certain supplementary information to which reference is made later in this report, it is felt that the inclusion of these detailed individual records is warranted. Immediately upon its capture the temperature of each swallow was taken by means of a specially designed mercury thermometer (described later). The bird was then placed in a small gauze bag and weighed. Then it was examined for parasites, possible deformities or other features of interest, and banded. Since it ordinarily is not possible to distinguish the sexes in the bank swallow by external characters, several specimens were killed and dissected in order to be sure on this point. The stomach contents of these specimens were examined later and the results are tabulated beyond. 154 Fig. 39. Bank swallow burrows in soil at top of roadside sand pit south of Oneida River, Oak Orchard district. Portions of the bank have slumped as sand and gravel have been removed. Burrows 40 to 50 feet above floor of pit. The figure shows only a part of east wall of the pit. July 22, 1931. * Fig. 40. Using mirror to illuminate interior of burrow. Various items of field equipment rest against the hank. North bank of Fish Creek at Fox farm. June 18, 1931. Slit/lies on the Bank Swallow 155 Table No. 39. Weight and Temperature Records of Adult Bank Swallows, Oneida Lake Region, 1931. Ex., excavation in progress ; F., birds Hew into burrow immediately prior to capture ; N., nest in burrow; In., incubating; Br., brooding. Figures in parentheses indicate number of eggs or young in nest. In cases where two birds were taken simultane- ously from same burrow the data in date and hour columns are not repeated for the second bird. Hour June DATE May 11. 14. 15. 19. 19. 22. 22. 22. 22. 22. 22. 26. 27. 27. 28. 2. 3. 3. 4. 4. 5. 5. 5. 8. 8. 9. 9. 9. 10. 10. 10. 10. 10. 11. 11. 15. 15. 15. 16. 16. 16. 16. A. M. 10:00 9:15 1 1 :00 10:00 10:30 10:40 P. M. 3:30 5:00 3:30 12:45 2:30 1:40 2:15 2:30 3:45 3:50 5:00 2:00 2:45 1:45 3:30 4:00 3:30 5:15 2:20 2:30 2:40 2:40 5:00 12:45 2:40 5:00 1:00 1:30 1:35 1:45 2:30 1:15 1:30 5:25 2:45 3:30 1:00 Sex m f m m f f f f m m f f f f m f f m f m f f Status of bird Ex. F. F Ex. '. '. '. In. (1) F nV. '. '. '. In.' (3) F F.'. In.'.'.'. F In.'.'.'.' In. . . . In. (4) In.'.'.'.' In. . . . In. . . . In. . . . In. . . . In. . . . In. . . . In. . . . In.. . . In.. . . Br. (3) In. (5) In. . . . In. (4) In. (5) In. (4) In. . . . In.. . . In.. . . In. . . . In.. . . In.. . . In. . . . Weight (Grams) 13 o 101 8 15 7 109 6 16 5 108 1 15 4 108 0 13 5 106 8 14 7 107 8 14 5 108 2 17 1 107 6 18 6 107 o 13 8 108 6 13 1 107 1 15 9 108 6 16 4 107 o 17 8 109 6 17 5 107 4 15 2 108 6 16 7 107 4 13 4 108 o 14 3 107 2 14 5 107 5 12 7 107 0 17 3 107 5 13 9 106 4 16 6 105 8 13 8 105 8 13 5 106 5 14 5 106 6 14 6 106 3 13 9 106 8 15 7 106 0 13 2 109 6 14 1 106 4 14 6 106 8 14 6 105 9 15 5 105 6 14 2 105 4 14 7 105 6 15 3 106 4 13 8 106 0 13 7 106 0 12 9 107 9 13 8 105 8 14 0 107 0 13 4 105 8 14 3 105 5 13 6 105 4 14 8 104 5 16 1 106 9 15 8 105 2 156 Roosevelt Wild Life Annals Table No. 39 — (Continued) June July 16. 16. 16. 18. 18. 19. 23. 23. 24. 24. 24. 25. 29. 30 7. 8 9! 13. 14. 17 20. DATE Hour A. M. r. M. 2:45 3:05 4:45 z :oo 3:00 12:50 v :zo 9:35 9:35 9:50 10:30 4:00 11:10 1 1 :45 i 2:35 4:25 1:50 2 10:25 3:20 8:35 3:15 Sex m m f m f Status of bird In. . . . In. . . . In. . . . Br! '(3) In.. . . Ex.... In. . . . In.. . . In. (2) In. (4) In. (3) In. . . . In. . . . In.. . . In. (4) Br. (2) Br. (4) In. . . . Br. (4) In. . . . Weight (Grams) Temp. (°F.) 1 J. L -I O 1 07 u 1 2 1 O A t: 1 HQ lUo 13 3 105 6 1 1 1 L o LVD 13 9 107 6 13 0 108 1 13 3 105 6 13 4 103 7 12 8 105 6 13 1 106 4 14 2 13 6 106 3 15 4 104 9 13 1 107 8 14 7 105 4 13 3 106 4 13 5 12 9 109 3 15 4 108 4 13 1 106 0 14 0 107 2 1 Repeat; see record for June 24, 9:35 A. M. - Repeat; see record for June 24, 9:35 A. M. and June 30, 11:45 A. M. Eight adult bank swallows banded in 1931 were recovered as returns in June, 1932 (Table No. 40). Table No. 40. Weights and Temperatures of Bank Swallows Banded as Adults in Season of 1931 and Recovered as "Returns" in Season of 1932. DATE Hour Weight (Grams) Tempera- ture °F. Breeding status May 19, 1931 1 1 :35 A. M. 15 7 107 6 Incubating June 2, 1932 12:30 P. M. 15 0 104 8 Incubating May 22, 1931 4:15 P. M. 16 8 107 6 Incubating June 3, 1932 1 :50 P. M. 14 1 107 9 Incubating May 27, 1931 11:15 A. M. 13 3 106 4 Incubating June 2, 1932 2:45 P. M. 14 5 105 6 Incubating May 27, 1931 1 1 :45 A. M. 13 9 107 4 Incubating June 2, 1932 12:05 P. M. 15 7 105 8 Incubating June 3, 1931 1:55 P. M. 12 9 107 8 Incubating June 2, 1932 11:35 A. M. 13 5 105 0 Incubating June 4, 1931 1 :50 P. M. 15 9 104 6 Incubating June 3, 1932 12:10 P. M. 15 4 105 8 Incubating June 4, 1931 2:20 P. M. 14 8 106 2 Incubating June 3, 1932 1 :40 P. M. 15 0 108 6 Brooding 5 young June 8, 1931 3:10 P. M. 13 4 107 1 Incubating June 10, 1932 10:05 A. M. 12 4 104 6 Incubating Studies on the Bank Szi'allow 157 Summary of Weights of Adult Bank Swallows, Including both Males and Females, May, June and July, 1931. Number of individuals 249 Number birds weighed twice 19 Number birds weighed tbree times 1 Average weight, all individuals, both sexes, for the entire period. . 14.56 grams Average weight, 89 individuals, both sexes, for May 14.65 grams Average weight, 150 individuals, both sexes, for June 14.54 grams Average weight, 10 individuals, both sexes, for July 14.13 grams Average weight, known females (22 individuals) for period 15.60 grams Average weight, known males (13 individuals) for period 14.03 grams Average weight, 12 females for May 16.36 grams Average weight, 9 males for May 14.27 grams Average weight, 9 females for June 14.70 grams Average weight, 4 males for June 13.42 grams Weight heaviest adult (a female), May 22 18.6 grams Weight lightest adult (2 males?), May 26 and June 10, each 12.0 grams Number birds weighing 18 to 18.6 grams 2 Number birds weighing 17 to 17.9 grams 10 Number birds weighing 16 to 16.9 grams 17 Number birds weighing 15 to 15.9 grams 53 Number birds weighing 14 to 14.9 grams 85 Number birds weighing 13 to 13.9 grams 63 Number birds weighing 12 to 12.9 grams 19 As indicated in Table No. 39, two adult bank swallows were sometimes found together in the same burrow, especially in the month of May. Presumably these were male and female for, as indicated elsewhere in this report, the male often sits in the burrow by the female while she lays the eggs, and also during the early part of the incubation period. In most cases where two birds were found in a burrow at the same time a considerable difference in weight between the indi- viduals was noted, the greatest difference in this respect being 4.8 grams, as shown for the two birds captured on May 22, 1931, at 1 :40 P. M. In only one case, that of the two birds captured at 11 :15 A. M. on May 27, 1931, were the weights identical for two adults occupying the same burrow at the same time. This difference in weight affords at least circumstantial evidence that the two birds were male and female, which is supported by the facts in those cases where the sexes were positively determined. Among the nineteen "repeats," each of which was weighed on two different occasions, was one for which the greatest length of time elapsing between weighings was twenty-eight days (May 18-June 15), but this individual showed a difference in weight of only 0.4 gram. In only four individuals did the difference in weight on the two occasions of capture amount to as much as one gram or more (1.3-1.7 grams). The 1.7 grams difference in weight was shown in an individual after a 7-day interval. In two individuals the weights were identical on the two dates of capture, after intervals of six and thirteen days, respectively. Eight indi- viduals showed an increase in weight after intervals varying? from six to thirteen 158 Roosevelt Wild Life Annals days; and nine showed a loss in weight after intervals varying from two to twenty- eight days. In this connection it is to be recognized, of course, that individuals may suffer various fluctuations in weight within varying periods of time, and that the differences above noted do not represent the only losses or gains that may have occurred within the periods concerned. And in all cases allowance is to be made for the variable factor represented by the stomach contents. For the single individual that was caught and weighed on three different occasions in the same season, the record was: June 24, 9:35 A.M., 12.8 grams; June 30, 11:45 A.M., 13.1 grams; July 9, 1:50 P.M., 13.5 grams. This indi- vidual, which upon dissection proved to be a male, was incubating on both dates of its capture in June, while on the July date it was brooding two young about three days old. Among those individuals (returns) that were recaptured and weighed the second season, the smallest weight difference (gain) shown was only 0.2 gram; and the greatest difference (loss), 2.7 grams. For comparison it may be remarked that thirty-six adult bank swallows of both sexes, taken at Lake Okoboji, Iowa, between June 17 and July 10, 1924, gave a weight range of 11.7 to 17.1 grams. Their average weight was 14.89 grams, or 0.33 gram more than the average for the 249 Oneida Lake examples taken from May to July, 1931. Growth Rate in Young. — In an attempt to learn something about the rate of growth in young bank swallows, weights of a number of broods were taken at stated intervals. These young, after their first weighing, were also banded for subsequent individual identification. The weight results are presented along with other data. Attention may be directed to the fact that in Table No. 42, for example, the greatest difference in weight among the individuals of one family was 4.7 grams (Nos. 46486F and 46489F for June 18) ; yet within a period of 8 days this differ- ence had been reduced to .8 of a gram. Likewise, it may be pointed out that the lightest nestling bird on the date of initial weighing was one of the first two individuals to leave the nest. The greatest discrepancy in weight between the smallest and the largest indi- viduals of any brood of young for which weights were taken was 9.7 grams (June 24). From the records secured it appears that young bank swallows are heaviest between the twelfth and seventeenth days of their existence, approximately, their weight up to this time having increased at a fairly uniform rate. From about the sixteenth day in the nest to the time of first leaving it, a few days later, the young birds lose on the average from 1 to 3 grams. But even at the time of initial flight, ordinarily at the age of 19 to 21 days, the average weight of the young is somewhat greater than that of adults, so that further weight reduction occurs after the young have learned to fly. This decrease in weight is doubtless due primarily to the reabsorption of fat, the more cancellated condition of the bones and the shrinking of the oral and other membranes. The stomach of a bird 12 to 15 days old also is larger and of greater capacity than that of the mature bird. Studies on the Bank Swallow 159 BODY TEMPERATURE OF BANK SWALLOWS Body Temperature of Adults. — Comparatively little detailed information is available regarding the body temperature of our native birds. The studies of Wetmore (1921), Kendeigh and Baldwin (1928), and Baldwin and Kendeigh (1932), are the principal recent inquiries into this field. Aside from these, during the past decade, particularly, there have appeared occasional references to the temperature of wild birds, but comparatively few definite records. The writer's first attempts at studies on bird temperatures were made in the Lake Okoboji region in the summer of 1925, when readings were taken by means of specially constructed mercury thermometers on about 500 living young and adult bank swallows. This work was resumed during July, 1926, and between June 15 and July 18, 1027. Summaries of the results have been published (1926a; 1927).' In the course of our work on the bank swallow in the Oneida Lake region, temperature readings were obtained on more than 300 adult and 200 young bank swallows. For about fifty individuals, adults and young, serial records numbering well over 1000 have been made. Essential parts of these records are incorporated in the present account. Methods of Obtaining Temperatures. — The temperature records incorporated in this report were secured by means of specially designed mercury thermometers of two different types. One of these is essentially like the clinical thermometer used by physicians. It is 4% inches long, three-edged, maximum diameter 3/16-inch, graduated to tenths of a degree, with a registration range from 92° to 120° Fahr. The mercury column has to be shaken down after each reading. Although this thermometer requires only' a few seconds to register the tempera- ture, a practice was made of leaving it inserted for at least one minute. After a few days in actual practice it was found that the temperature of an individual bank swallow might vary several degrees within a few minutes, which meant that in order to get a record of these changes, repeated insertions would be necessary with the above-mentioned type of thermometer. This was time-con- suming and moreover involved an undesirable amount of handling of the bird, and possibly considerable discomfort to it. In view of these objections, the help of the Taylor Instrument Company of Rochester, N. Y., was sought with the result that another type of thermometer was designed, which I have used since 1927. Briefly, the specifications of this thermometer are as follows : Total length, 6 inches; greatest diameter of the nearly cylindrical tube, 3/16-inch, length of mercury bulb, 7/16-inch, diameter 1 /16-inch. The instrument is graduated to tenths of a degree and carries a registration range from 90° to 118° Fahr. It is extremely sensitive and registers immediately any slight rise or fall of the temperature. The mercury column, which requires no shaking down, rises and falls with the tem- perature changes of the subject. All of our temperature records apply to the living bird. During the early period of our studies the best method of securing the tem- perature record was given some attention. For a bird as small as the bank swallow, introducing the thermometer into the rectum is not satisfactory, since too much preliminary handling of the subject is required, and in the insertion of the instru- 160 Roosevelt Wild Life Annals ment apparently a considerable degree of discomfort is caused the bird. The axillary method — inserting the thermometer into the axilla and holding the wing closely pressed against the body —also has its questionable features. The most satisfactory method was to insert the thermometer by way of mouth and oeso- phagus into the proventrieulus (Figs. 43 and 44). The readings thus obtained showed close agreement with those secured by the rectal method, and besides, resulted in no apparent discomfort to the bird. This method was used in all our subsequent work on bank swallow temperatures. By tbis method the thermometer can be kept in position for minutes at a time, while the bird is held gently in the hand. Even very young bank swallows "take" the instrument readily and suffer no apparent discomfort. Indeed, young that are only a few days old sometimes "swallow" the thermometer greedily. However, there is some difference of opinion regarding the question of true temperature indications for birds, and some of the arguments involved may here properly be discussed. 1. The bird may be frightened in the process of capture and so will not show a normal degree of temperature. Possibly this is true to some slight extent, but in the present study the time elapsing between the moment when the swallow was first disturbed in the burrow and the moment when the thermometer was intro- duced was, in most cases, not more than thirty seconds ; often it was less, some- times a little more. While some change in their temperature no doubt takes place in this interval, all adult swallows were subject to this factor. Excitement, fear, discomfort, etc., may affect the temperature to a certain extent, but I see no way of eliminating these factors without killing the bird. In some cases adult and young bank swallows were purposely allowed to flutter in the collecting net, or exposed to unusual influences in order to observe possible effects on their registered tem- peratures. Most of such individuals showed an increase in temperature, but some showed a decrease. 2. The muscular exertions of the bird may cause a rise in its temperature. Doubtless this is true, but to what extent is difficult to determine; and therefore temperature records for the live bird should probably be interpreted as approxima- tions only. However, a large series of records on a given species of bird, secured under various conditions, should give at least a fair notion of the ordinary or "normal" amount of variation, so that a reasonably accurate figure can be set down as representing the average temperature. In such a species as the bank swallow, where a variation of 2 to 5 degrees seems to be common, it is hardly possible to ascribe to the bird more than an approximate normal temperature or normal temperature range. Even though a bank swallow reposing quietly in the hand may make no active effort to escape, it may be in a state of tension. It often happens, however, that adults, while in the hand and with a thermometer inserted down their throats, appear to be wholly relaxed and at ease. In such circumstances, at least, the tem- perature readings may be considered to represent what for all practical purposes may be called a normal condition. 3. Some of the bird's heat may be absorbed by the glass of the thermometer. Fig. 41. Bank swallow burrows, south bank of Fish Creek at Graham farm. Several of the burrows have been marked for identification. July 3, 1931. Fig. 42. Balances used in weighing bank swallows ; an adult bird held captive in a gauze bag on the weighing platform. Fish Creek Landing. June 29, 1931. Fig. 44. Thermometer inserted into gullet of adult bank swallow. Fish Creek at Graham farm. June 4, 1931. Studies on the Bank Swallow 163 The temperature of bank swallows has shown itself to be so variable that it is difficult to say just how much consideration should be given to this factor. Since most of our work was done when the air temperature was 60° F. or above, and since the thermometer usually was carried in a pocket of the clothing where some heat from the body reached it, the instrument was at no time actually cold. Furthermore, the instant response of the thermometer after being introduced would seem to preclude any considerable amount of heat being absorbed by the instrument before the mercury reached maximum height. It seems probable, too, that most of whatever amount of heat might be absorbed by the thermometer would be taken up by the heavier glass in the tube above the mercury bulb. And this would probably affect a very young bank swallow more than it would an adult. Numerous experiments were made in the field, in which the thermometer was inserted when at air temperature, as well as after it had been warmed by holding in the closed hand. The results, however, were nearly identical. In fact, in some cases the reading shown for an individual on which the cold thermometer was used was actually higher than for one on which the warmed instrument was used. In view of the above, it is felt that correction of the readings obtained is not necessary in order to arrive at reasonably safe conclusions regarding the temperature of these birds. It may be added that the second type of thermometer above described, with its smaller bulb and tube, appears to furnish more accurate readings than the first type. In their discussion of the temperature of house wrens, Kendeigh and Baldwin (1928, p. 253) state that the thermometer readings "are usually only .2°-.8° lower than the original thermocouple reading of the body temperature." It is probable that in a larger bird like the bank swallow this difference would be still less. 4. The accuracy of a mercury thermometer as compared with that of a thermo- couple may fairly be questioned. However, both portability and expense were important items for consideration in the present study, and moreover, the delicacy of the electrical device would preclude its extended use under the field conditions encountered in our bank swallow work. It is freely granted that the mercury thermometer has its limitations. 5. The temperature of a bird may be modified somewhat by the time elapsed since feeding. An individual might perhaps show a lower temperature reading immediately following a feeding period than later. Since, however, most of the temperature records in this paper are those of incubating or brooding individuals, the food factor is to a large extent eliminated. 6. The air temperature at the time the readings are taken may modify the body heat of the bird. While this may be true in cases of extremes, our findings show that the temperature of adult bank swallows bears no close and constant relationship to the ordinary fluctuations of air temperature. Some of our highest readings on adult birds were recorded on the coolest days of spring, and some of our lowest readings were procured on the hottest days of summer. The follow- ing records from the Oneida Lake region, 1931 season, substantiate this statement. 164 Roosevelt Wild Life Annals Date Max. Air Max. Bank Swallow Temp. °F. Temp. °F. May 14 63 109.6 May 18 78 108.1 May 28 89 105.3 May 29 90 105.1 June 3 53 108.3 June 19 92 106.9 July 2 100 104.7 While the ordinary summer temperatures may have little effect on adult hirds, greater extremes would doubtless be a different matter. In general, I am inclined to agree with Wetmore (1921, p. 13) that "Where the individual is in normal health and is sufficiently supplied with food, the agencies of temperature control will tend to maintain an even body heat. Any variation that may occur, other than that incident to the daily rhythmic rise and fall of body heat, may be attributed to some other condition that under normal conditions would disappear within a comparatively short period through a readjustment of the bodily functions." Moreover, during the incubating and brooding seasons, when most of our records were secured, the swallows came from burrows with temperatures more nearly constant than the temperature of the outside air. In the case of young birds, however, the situation is different. Lacking the feather covering and the more efficient temperature control of mature' birds, they are notably subject to the influence of the surrounding temperature. On this account, therefore, special precautions were exercised in obtaining temperature read- ings for nestlings. Average Temperature of Adult Bank Swallows. — For the seasons of 1925, 1926 and 1927 in the Lake Okoboji region, and the 1931 season in the Oneida Lake region, temperature records for a total of 603 different individuals (both sexes) were secured. Most of these bank swallows were captured under similar conditions, while they were incubating or brooding ; a few were working on burrow excavations ; several had just flown into the burrows with food for young, while a number of the May and early June individuals had merely entered the burrows during the general mating activities. A condensed summary of the readings secured on these birds together with maxima, minima and averages by months, average temperature for the period, and other data, are incorporated in Table No. 41 submitted below. It is of interest to note that for corresponding months the temperature records for the Iowa bank swallows are higher than those for the New York birds. On the other hand, the New York records present greater uniformity, neither the high nor the low extremes of the temperatures for Iowa birds being noted in those for the Oneida Lake region. The data presented elsewhere tend to show that, in general, bank swallows which have just come to rest from flight exhibit a somewhat higher body temperature than those which have been resting in the burrow, and it is possible, therefore, that the temperature differences noted are thus explained, in part, at least. Studies on the Bank Swallow 165 O . H w 55 t/i * o o >-l t/3 g > c 3 ■ - _ - •- +* c u O C c; 0. H 3 u. **• t- x <0 £ ? X ■ 00 • <-i X O -N00 6 8 .2 « >6h c o 3 O J00O iri i/l . 166 Roosevelt Wild Life Annals I 'aviation and Fluctuation in Temperature of Adult Bank Swallows. — It is obvious that distinction must be made between fluctuation in temperature, and variation. Fluctuation refers to changes in temperature that may occur within the individual, perhaps as a reaction to stimuli of various kinds. Variation, on the other hand, is here interpreted as any difference in temperature found in different individuals at the same time, at different times, under similar or dissimilar situations, or in the same individual at different times. In other words, fluctua- tion in temperature may occur within the body of an individual bank swallow, while variation in temperature may be exhibited by different individuals at the same time, or by a given individual at different times. A considerable number of temperature readings were made on individual birds at 10-second intervals, the rise and fall of the mercury column being observed continuously for periods of half an hour or more. As a rule the bird exhibited no outward symptoms of alarm or excitement. Sometimes it struggled for a brief moment in the hand, and due note was made of that fact. In the next few pages a number of outstanding examples of fluctuation and variation in the temperature of adult bank swallows are presented, records from both the Lake Okoboji and the Oneida Lake regions being included. The principal reason for cataloguing these numerous records is to emphasize the remarkable degree of fluctuation and variability in body heat of the living bird ; conclusions based on a limited number of records are likely to be misleading. Moreover, records obtained from recently killed birds would seem to be unreliable unless they be secured within a few seconds after death. No. 61147 A was taken June 16, at 8:10 A.M., from a burrow in a roadside cut four miles north of Milford, Iowa. The mean temperature for the day at the U. S. Weather Bureau Station at Lake Park, some ten miles away, was 58.5 °F. The temperature of the bird at the time of capture was 107.3°. Twenty seconds later the reading was 106.8° ; five minutes later, 99.6° ; during the next minute the temperature dropped to 98.0°, and during the succeeding 70 seconds it fell to 97°. Immediately thereupon the following readings were taken at 10-second intervals: 96.8°, 96.6°, 96.4°, 96.2°, 96.0°, 95.9°, 95.6°, 95.4°, 95.2°, 95.1°, 94.8°, 94.6°, 94.4°, 94.2°, 94.3°, 94.1°, 93.8°, 93.6°, 94.0°, 94.1°, 94.1°. 93.8°, 94.0° (bird restless), 94.2°, 94.2°, 94.4° (bird again restless), 94.2°, 94.0°, 93.6°, 93.6°, 93.5°, 93.2°, 93.0°, 92.8°, 92.8°, 93.0°, 93.0°, 93.1°, 92.9°, 92.8°, 93.2° (bird again restless); and now, after a 5-second interval, the reading was 92.8°; then three 10-second intervals gave 92.6°, 92.5° and 92.0°. Two minutes later the reading was 93.5°. and two succeeding 10-second intervals gave 92.6° and 92.4°. After a period of five minutes the mercury failed to rise to 90° and the bird was returned to the burrow. At 3 :30 P. M. on the same day, the same swallow was recovered from another burrow in the same roadside cut. Its temperature now was 105.9°. The follow- ing readings then were noted at 10-second intervals: 105.6°. 105.4°, 105.2°, 105.0°, 104.8°, 104.8°, 104.6°, 104.6°, 104.5°, 104.5°. Another individual, No. 61165A, which had just flown into a burrow in a roadside cut, was captured at 3:45 P. M. on June 16. Thermometer readings at 10-second intervals were as follows: 106.9°, 106.2°, 105.8°, 105.4°, 105.0°, 104.8°, 104.4°, 104.2°, 104.0°, 103.8°, 103.5°, 103.1°, 103.0°, 102.8°. 102.6°, 102.4°. Studies an the Bank Swallow 167 102.0°, 101.9°. 101.8°, 101.6°, 101.6°, 101.4°, 101.1°, 101.0°, 100.8°, 100.6°, 100.5°, 100.4°, 100.2°, 100.0°, 100.0°, 100.0,°, 100.0°, 99.8°, 99.7°, 99.6°, 99.7°, 99.9°, 99.7°, 99.7°, 99.5°, 99.4°, 99.3°, 99.2°, 99.2°, 99.1°, 99.0°. It will be noted that in this bird the temperature dropped 7.9 degrees during a period of seven minutes and fifty seconds. No. 61168A, an incubating bird, was taken in a roadside gravel pit four and one-half miles north of Milford, Iowa, at 4:00 P.M., June 17. The mean tem- perature for the day was 57° F. The bird's temperature when captured was 104.4°. In the following series of thirty- four readings taken at 10-second intervals, the body heat dropped to 97.0°, — a total decrease of 7.4 degrees — where it was main- tained for seventy seconds, when the thermometer was withdrawn and the bird banded. This required two and one-half minutes. The thermometer was inserted again, and it was found that the temperature of the swallow had risen to 98.4° ; but at the end of one minute it had dropped once more to 97.0°, where it was maintained for ninety seconds, when the observa- tion ended. No. 61180A, incubating, in a gravel pit two miles east of Spirit Lake, Iowa, June 18. The mean air temperature on this date was 69°F. At the time of its capture, 10:10 A.M., the temperature of the bird was 106.8°. Readings taken immediately thereafter at 10-second intervals showed that this temperature was maintained for thirty seconds, when it began to rise gradually until, in four minutes and twenty seconds it reached 109.0°, where it remained for one minute and forty seconds. The thermometer was now withdrawn, for the usual banding operation, which required two minutes, and thereupon a few readings at 10-second intervals gave the following results: 109.4°, 109.2°, 109.2°, 109.2°, 109.2°. No. 61182A, an incubating bird, was taken at 10:30 A. M., June 18, in a gravel pit two miles east of Spirit Lake, Iowa. The mean air temperature was 69°F. ; maximum for the day, 81 °F. Temperature of the bird immediately follow- ing capture, 107.4°. The swallow was then held in the hand in the hot sun, when its recorded temperature dropped to 107.2°, and where it remained for one minute; in the following one minute and fifty seconds it rose to 108.0°, where it remained for several minutes. In this case the warmth from the sun may have been responsible for checking the earlier drop in temperature as well as causing the later rise. However, other individuals subjected to essentially identical conditions responded somewhat differently. For example, incubating bird No. 61183A, taken at 10:40 A. M., on the same date in the same gravel pit as No. 61182A gave, when captured, a temperature reading of 105.8°. This swallow, too, was held in the hand in the hot sun. Within twenty seconds its temperature dropped 0.2 degree ; immediately thereafter it began steadily to rise until, at the end of four minutes it had reached 107.4°, where it was maintained. No. 97018A, an incubating bird, was taken at 10:35 A.M., June 22, in a gravel pit near Arnolds Park, Iowa. The mean air temperature on this date was 60°F., minimum 49°, and maximum 71°. At time of capture the bird's tempera- ture was 105.0°. Within two minutes it had dropped to 100.1°. In the immediately 168 Roosevelt Wild Life Annals following forty seconds its temperature rose 1.9 degrees; but during the next minute dropped again, this time to 100.0°; then, in the next one and one-half minutes, rose to 101.2°. No. 97037 'A was taken from a burrow in a roadside cut six miles west of Spirit Lake, Iowa, at 9:30 A.M., June 23. The mean air temperature on this date was 64.5°F., with a minimum of 51° and a maximum of 78°. At the time of capture the bird's temperature was 104.8°. Readings at 10-second intervals were: 104.6°. 104.2°, 104.1°, 104.1°, 104.0°, 103.9°, 103.8°, 103.8°, 103.8°, 103.7°, 103.8°, 103.6°, 103.7°, 103.6°. Following an interval of two minutes, during which the thermometer was withdrawn, readings at 10-second intervals were: 104.3°, 104.0°, 104.0°, 104.0°. The bird was then left to flutter about in the collecting net for five minutes, when ;i third series of readings at 10-second intervals gave the following: 110.8°, 110.6°, 110.4°, 110.4°, 110.2°, 110.2°, 110.2°, 110.0°, 109.8°, 109.7°. It is evident here that muscular exertion had a marked influence on the body temperature of the bird, and hence it is to be expected that bank swallows captured shortly after they have entered their burrows, following flight, usually will show a higher temperature than that of incubating or brooding individuals. No. 97175A was secured from a burrow in a roadside cut one mile south of the Iowa Lakeside Laboratory, at 7:30 P.M., June 29. Mean air temperature 79°F.. maximum 92°, and minimum 66°. Fifty-six temperature readings at 10-second intervals were: 105.9°, 105.9°, 105.9°, 105.9°, 105.8°, 105.7°. 105.9°, 105.8°, 105.9°, 105.9°, 105.9°, 106.0°, 105.9°, 106.0°, 106.1°, 106.2°, 106.3°, 106.3°, 106.5°, 106.4°, 106.5°, 106.5°, 107.1°, 107.4°, 107.6°, 108.0°, 108.2°, 108.4°, 108.5°, 108.6°, 108.7°, 108.7°, 108.9°, 109.0°, 109.1°, 109.1°, 109.2°, 109.2°, 109.3°, 109.3°, 109.2°, 109.0°, 109.2°, 109.4°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°, 109.6°. In this swallow the temperature reading at the time of capture was lower than for many others taken in similar circumstances, but the readings immediately following capture indicated a rise in the body temperature. Within the period of four and one-half minutes this swallow's temperature had risen from 105.9° to 109.6°, — an increase of 3.7 degrees — where it remained to the close of the observation about two minutes later. No. 5650315, an incubating female, was captured at 11:45 A.M., May 27, from a burrow on the south bank of Fish Creek at Fish Creek Landing, New York. The mean air temperature was 62°F., minimum 47°, and maximum 77°. At the time of capture the bird's temperature was 107.4°. Readings at 10-second intervals as the bird lay quietly in the hand: 107.2°, 107.2°, 107.4°. 107.6°, 107.9°, 108.0°, 108.2°, 108.4°, 108.2°, 108.2°, 108.0°, 108.1°, 108.5°, 108.4°, 108.4°, 109.0°, 109.0°, 109.2°, 109.3°, 109.1°, 108.9°, 108.9°, 108.8°, 108.8°, 108.9°,' 109.0°, 109.0°, 109.0°, 109.0°, 109.0°, 109.0°. The swallow was weighed and banded, thereupon left in the collecting net for ten minutes. Its temperature was then found to have increased to 110.4°; but at the end of the following two minutes it had dropped to 108.7°. Tin's bird was then left to flutter about in the collecting net for only one minute, after which its temperature was 108.6°. Then it was carried to the near-by creek and its feet and tarsi held in the water for one minute; temperature Studies on the Bank' Swallow ]<>'> of water, 56.0° F. Thereupon temperature readings made at 10-second intervals were as follows: 105.6°, 104.6°, 103.8°, 103.6°, 103.0°, 102.0°, 102.2°, 101.8°, 101.0°, 100.6°, 100.6°, 100.6°. The bird was again left to flutter in the collecting net for one minute, after which six serial readings gave the following: 100.4°, 99.8°, 99.2°, 99.0°, 99.0°, 99.0°. Then it was left to flutter for a 10-minute period, at the end of which its temperature was 108.8°, a rise of 9.8 degrees in the ten minutes. Eight days later, at 2:40 P.M., this individual was recovered from the same burrow, when its temperature reading was 106.0°. This swallow was recovered at the same site on June 2, the following year, at 12:05 P.M. On this occasion, also, the bird was incubating; its temperature was 105.8°. Then, after a period of ten minutes, during which the bird was examined and weighed, its temperature had risen to 105.9°. No. 56512B was incubating when captured from a burrow in the south bank of Fish Creek, at 1:45 P. M.; its temperature was 108.0°. Temperature of the burrow, 66°, and air temperature on the face of the bank, 81°. Following the initial reading the swallow was placed in the gauze collecting net and exposed to the warm sun, while for a period of two minutes it was left to flutter and struggle in its efforts to escape. At the end of that time its temperature readings at 10-second intervals were: 112.4°, 112.2°. 112.1°, 112.0°, 111.9°, 111.9°. The bird was thereupon held with its feet and legs in the water of the creek, the temperature of which was 64°, with the following results at 10-second intervals: 111.2°, 110.8°, 110.4°, 110.2°, 109.9°, 109.4°, 108.4°. Following this the swallow was held in the hand in the hot sun with these results at the same intervals: 108.8°, 108.6°, 109.4°, 108.8°, 108.8°, 109.0°, 108.8°, 108.6°. Comparing some Iowa records : No. 97822A was an incubating bird taken in a burrow in a roadside gravel pit one-half mile south of the Iowa Lakeside Laboratory at 8:10 A.M., July 13 (mean air temperature, 66°F., maximum 77°, and minimum 55°). Swallow's temperature at capture, 107.3°. Temperature readings at 10-second intervals: 107.2°, 106.8°, 106.4°, 106.2°, 106.0°, 105.8°. The bird was then held so that its feet rested on a cake of ice while the readings continued: 105.8°. 105.6°, 105.2°. 105.0°, 104.6°, 104.2°, 104.0°. 103.7°, 102.8°, 102.4°, 102.1°, 101.9°, 101.7°, 101.4°. 100.5°, 100.4°, 100.0°, 100.0°, 99.6°, 98.8°, 99.0°, 99.1°, 99.1°, 98.8°, 99.2°, 99.0°, 98.7°, 98.9°, 99.0°, 98.7°. With ice removed, the bird was held in the hand for two minutes, during which its temperature rose to 99.3°. No. 97823A, incubating, was captured in a burrow in the gravel pit just mentioned at 8:50 A.M., July 13. Temperature of bird 108.2°. Readings at 10-second intervals as follows: 107.7°, 107.5°, 107.2°, 107.1°, 106.8°, 106.6°. The feet of this individual were then held on a cake of ice, as with No. 97822A, and 10-second readings were resumed: 106.2°, 106.1°, 105.9°, 105.7°, 105.1°, 104.9°, 104.9°, 104.7°, 104.4°, 104.0°, 103.6°, 103.0°, 102.7°,' 102.3°', 101.9°', 101.5°, 101.2°, 100.9°, 100.7°, 100.3°, 99.7°. 99.4°, 99.1°, 98.9°, 98.8°, 98.7°,' 98.7°, 98.6°, 98.5°. As with No. 97822A, this swallow was now held in the hand with its feet off the ice for a period of two minutes when its temperature registered 98.2°. 170 Roosevelt Wild Life Annals The bird then likewise was left to flutter about in the collecting net for twelve minutes, and its temperature rose to 112.2°, an average rise of almost 1.2 degrees per minute. Subsequent readings at 10-second intervals were as follows: 112.0°, 111.8°. 111.7°, 111.4°, 111.1°, 110.9°. Regarding the two last-mentioned swallows (Nos. 97822A and 97823. \) the following points are of interest: 1. Both birds were incubating at the time and were taken from the same gravel pit, within fifty minutes of each other. 2. There was a difference of 0.9 degree in their temperatures at capture. One minute later the temperature of the first had dropped 1.4 degrees, that of the second, 1.1 degrees. 3. When the feet of these birds were held on ice the rate of temperature decrease accelerated in both birds for a period of four minutes: in one, the average rate of decrease was 1.55 degrees per minute, in the other, 1.7 degrees per minute. In both, the rate of decrease diminished after their temperature had reached 99.6°. 4. After equal rest periods, followed by equal periods of muscular activity of similar nature, they showed a difference of 1.2 degrees in temperature. How- ever, the individual with the higher initial temperature also showed the higher temperature at this stage ; and at the end of one minute of rest showed a drop of 0.7 degree as against 1.1 degrees for the other individual. 5. These facts show that both the rate and the range of temperature fluctua- tion differ only slightly in different individuals under similar conditions. Other results have shown that they may vary in the same individual at different times. The following records are for a male bank swallow captured from a burrow in the north bank of Fish Creek at 3:30 P. M., May 14. Air temperature at mouth of burrow, 64°. The burrow, which contained no nest material, was 16 inches below the top of the bank, 31 inches long, and its temperature 58.5°. The bird had just flown into the burrow and when captured showed a temperature of 109.6°. Readings at 10-second intervals were as follows: 109.4°, 109.4°, 109.4°, 109.4°, 109.4°, 109.4°, 109.4°, 109.2°, 109.0°, 109.1°, 109.2°, 109.1°, 109.2°, 109.1°, 109.0°, 109.0°. 109.0°, 108.9°, 109.0°, 109.0°. 108.8°, 108.7°, 108.6°, 108.4°. 108.2°, 108.0°, 108.0°, 107.6°, 108.0°, 108.0°, 107.8°, 108.0°, 108.0°, 108.0°. 108.0°, 108.0°, 107.8°, 107.6°, 107.6°, 107.6°, 107.6°, 107.6°, 107.6°, 107.6°. 107.6°, 107.6°. For one minute the bird was now held by its feet while it fluttered and struggled to free itself, following which three temperature readings at 10-second intervals were: 107.6°, 107.4°, 107.4°. Then the swallow was placed in the collecting net for one hour, during which time it fluttered and climbed about in its efforts to escape. At the end of the hour its temperature was 106.4°. The swallow was then killed by puncturing its brain with a sharp instrument, whereupon temperature readings at 10-second intervals were as follows: 105.8°, 104.0°, 103.6°, 102.4°, 101.8°. 101.2°, 101.0°, 100.4°, 99.4°. 98.6°. 97.8°. Points of special interest in connection with this experiment are: 1. The bird had flown into the burrow just before capture and although the air temperature was comparatively low the subject's temperature was higher than Studies on the Bank Swallow 171 that of the average that had been found for quiescent incubating or brooding individuals. 2. Its temperature fell — not continuously, but with occasional brief rises — very slowly until a low of 107.6° was reached. 3. After this temperature had been maintained for ninety seconds, a period of muscular activity, consisting of flapping of the wings in an attempt to escape, followed, but this did not affect the registered temperature. Two adult bank swallows, Nos. 46429B and 46430B, probably male and female, were captured from the same burrow at the Bryant sand pit at 1 :00 P. M., May 18. Temperature of burrow which contained a nest but no eggs. 55°. Air temperature at mouth of burrow, 72°. Temperature of No. 46429B at capture, 107.4°; in two minutes its temperature had dropped to 105.4°. Temperature of No. 46430B at capture, 107.6° ; in two minutes its temperature had dropped to 106.8°. While the initial temperature of these two birds differed by only 0.2 degree, the temperature of the first bird dropped 2 degrees in 2 minutes, that of the second only 0.8 degree. No. 46438B and No. 46439B, probably male and female, were taken from a burrow in the south bank of Fish Creek, at 10:00 A.M., May 19. After they had been banded and weighed the following readings were secured at 10-second intervals on No. 46438B : 105.8°, 105.8°, 105.8°, 105.7°, 105.6°, 105.5°. 105.6°, 105.7°, 105.8°, 105.8°, 105.9°, 105.9°, 106.0°, 106.0°, 106.0°, 106.0°, 106.0°, 106.0°. The swallow was then held by its feet and left to struggle and flutter for one minute, with the following readings as a result: 106.6°, 106.4°, 106.2°, 106.2°, 106.2°, 106.2°, 106.4°, 106.4°, 106.5°, 106.5°, 106.5°, 106.5°, 106.6°, 106.4°, 106.4°. Eighteen readings were taken on No. 46439B, as follows: 106.4°, 106.2°, 106.1°, 105.8°, 105.6°. 104.8°, 104.5°. 104.0°, 103.8°, 103.7°, 103.4°, 103.7°. 103.8°, 104.6°, 104.4°, 104.1°, 103.8°, 103.6°. This second swallow, too, was held and allowed to flutter and struggle for one minute, the result as follows: 104.8°, 104.4°. 104.0°, 103.8°, 103.6°, 103.4°, 103.0°, 103.2°, 102.9°, 102.7°, 102.7°, 102.8°, 102.8°. 102.6°, 102.6°. These two birds showed a difference of 0.6 degree in body temperature at the initial reading. The same number of serial readings was taken for both, and these showed a general upward trend in one case (No. 46438B), but downward in the other. However, the upward trend was accompanied by brief temporary drops of the mercury, and the downward movement (in No. 46439B ) by momentary rises. At the close of this observation the temperature of No. 46438B was 0.6 degree higher than at the beginning, while in the other it was 3.8 degrees lower. Evidences of Diurnal Rhythm in the Temperature of Adult Bank Szmllows. — In order to determine whether any difference existed between the forenoon and afternoon temperatures of the bank swallow, a study was made of the records for a total of 213 adults captured between 8:00 A. M. and 11 :00 A. M., and for a total of 243 adults captured between 2:00 P.M. and 5:00 P.M. The average temperature for the first mentioned group was found to be 107.00°F. ; for the second, 107.34°F. The difference of .34 degree appears too small to have any significance. 172 Roosevelt Wild Life Annals In a further attempt to discover any diurnal rhythm that might exist, tem- perature records for 662 individuals were grouped according to the hourly periods in which they were ohtained. The result was that no clear evidence of rhythm on this basis was demonstrated, although there was a suggestion of a gradual rise in temperature in this swallow until late afternoon, with possibly a corresponding decrease during the night. Following is a summary of the results of the grouping of the 662 records by hourly periods. Total number readings 662 Number different birds represented in readings 603 Number temperature readings taken between 8:00 and 11 :59 A. M 306 Number temperature readings taken between 2:00 and 6:00 P. M 301 Average temperature for 306 birds taken between 8:00 and 11 :59 A. M.. . 107.20° Average temperature for 301 birds taken between 2:00 and 6:00 P. M.. . . 107.24° Average temperature for 356 birds taken between 1 :00 and 9:00 P. M.. . . 107.19° Average temperature for 50 birds taken between 8:00 and 8:59 A. M 106.96° Average temperature for 97 birds taken between 9:00 and 9:59 A. M 107.20° Average temperature for 99 birds taken between 10:00 and 10:59 A. M.. . 107.00° Average temperature for 60 birds taken between 11 :00 and 11 :59 A. M.. . 107.72° Average temperature for 41 birds taken between 1 :00' and 1 :59 P. M 106.89° Average temperature for 83 birds taken between 2 :00 and 2 :59 P. M 107.09° Average temperature for 101 birds taken between 3:00 and 3:59 P. M.. . . 107.71° Average temperature for 79 birds taken between 4:00 and 4:59 P. M 106.99° Average temperature for 38 birds taken between 5 :00 and 5 :59 P. M 106.85° Average temperature for 11 birds taken between 7:00 and 7:59 P. M 106.83° Average temperature for 3 birds taken between 8:00 and 8:59 P. M 107.30° 8:00-11:59 A.M. 1:00- 9:00 P.M. Fig. 45. Graphic representation of diurnal fluctuation in body tempera- ture of adult bank swallows (Riharia r. riparia) as obtained from 662 readings made in June and July, 1925, 1926 and 1927, in the Lake Okoboji, Iowa, region and in May, June and July, 1931, in the Oneida Lake, New York, region. Body Temperature and Sex. — In some species of birds, at least, variation in the body temperature seems to be correlated with sex. However, in such work as has been done in this field the results appear to vary. This may be due, I believe, to three causes: (1) insufficient data; (2) the failure to give proper con- Studies on ///c Bank Swallow 173 sideration to the fluctuations in temperature that may occur in a bird within a short time; (3) variations in temperature caused by surrounding conditions. My own records for hank swallow temperatures in relation to sex are wholly inadequate, but they are presented for whatever value they may possess. As is well known, it is usually not possible to distinguish between the sexes in the bank swallow by external characters, and in this study we did not feel justified in killing more than a few. The temperature records of the bank swallows in which the sex was definitely ascertained, appear in Table No. 39 (page ). It will be noted that in a few instances during May a male and a female were taken from the same burrow at the same time. For the eight males taken in May the average temperature was 107.8°, while for the three males taken in June, the average temperature was 106.5°. Thirteen May females averaged 107.2°. and ten June females, 105.9°. These records would seem to indicate that during the breeding period the tempera- ture of male bank swallows averages a little higher than that of females. Body Temperature of Adult Bank Swallows with Reference to Muscular Activity. — As mentioned earlier in this report, muscular exertion incident to ordinary flight affects the bank swallow's temperature as indicated by the ther- mometer readings, and the evidence bearing on this question is herewith briefly presented. For 258 adult bank swallows, incubating, brooding or quiescent individuals, captured from burrows at various hours of the day between 8:00 A. M. and 8:00 P. M., the average temperature was 106.46°F. ; maximum 110.0°, minimum 103.1°. For 56 birds that flew into the burrozc, as we watched between 8:00 A.M. and 8:00 P.M., and were captured immediately thereafter, the average temperature was 108.06° F. or 1.60 degrees higher than for quiescent individuals. Among these records the maximum was 110.0°, the minimum 104.6°. A further examination of our data shows that for 83 adult bank swallows captured in burrows as quiescent individuals, between 8:00 and 11:00 A.M., the average temperature was 106.27°F., with the maximum reading 109.5° and the minimum 103.1°. Similarly, for 31 adult birds that were observed to fly into the burrow immediately preceding capture, between 8:00 and 11 :00 A. M., the average temperature was 108.01 °F., with a maximum of 110.0° and a minimum of 104.6°. Individuals just returned from flight, therefore, show an average temperature 1.74 degrees higher than for quiescent individuals during the same hours of the day. If now we compare these average morning temperatures with average after- noon temperatures of quiescent and recently active (in flight) birds we have the following interesting results. For 134 adult bank swallows captured from the burrow as quiescent individuals between 1:00 and 5:00 P.M. the average temperature was 106.57°F., with the maximum reading 110.0°, and the minimum 104.4°. This average is 0.30 degree higher than for forenoon adults of the same status. And for 22 adults that were observed to fly into the burrow immediately preceding capture between 1 :00 and 5:00 P. M. the average temperature was 107.79°F., with a maximum of 110.0°, and a minimum of 105.8°. This average is 0.22 degree lower than for forenoon adults of the same status. As a matter of fact we should expect an average tern- 174 Roosevelt Wild Life Annals perature for this latter group of somewhat more than 108.01°F., and possibly this would be the case were it possible to have a larger series of records. An average difference of 0.22 degree is scarcely of significance in any species with a tempera- ture as variable as that of the bank swallow. Nevertheless our figures do suggest again that muscular exertion is responsible for the higher temperature of recently active birds as compared with quiescent individuals. General Conclusions on Temperature of Adult Bank Swallows. — The average temperature for 603 adult hank swallows (including both sexes) was 107.1 °F. Eleven adults known with certainty to have been males, had an average temperature of 107.50° F. (maximum 109.6°, minimum, 105.6°). Twenty-three adults known with certainty to be females had an average temperature of 106.99°F. (maximum 109.6°, minimum, 104.5°). For adult bank swallows, irrespective of sex, what may be called the normal temperature range falls between an average minimum of approximately 103.5° and an average maximum of approximately 110.0°F. Although not invariably true, the temperature of adults usually drops after rbe thermometer has been introduced into the proventriculus, and continues to drop more or less regularly until a low is reached, this low varying with the individual. The temperature of adult bank swallows shows considerable variation, not only in different individuals at different times and in different conditions, but also in the same individual at different times in similar conditions. Individual variation in the temperature of adult birds may have a range of 20 degrees or more (from maximum of 112.4°F. to minimum of 90.0°F.). The rate of fluctuation of temperature in adult hank swallows sometimes may be as high as one degree in one minute. Under practically identical external circumstances the body heat of certain individual adult bank swallows has registered a rise of 9.8 degrees in 10 minutes ; in other individuals there has been a decrease of 7.9 degrees in 8 minutes. The variations and the fluctuations in the temperature of adult bank swallows do not appear to be correlated in any definite way with the surrounding air tem- perature. However, an appreciable amount of radiation of bodily heat occurs through the unfeathered legs and feet of the bird when these parts are subjected to temperatures considerably below that of average summer air temperatures. The marked fluctuation and variation of temperature in mature or nearly mature bank swallows may be dependent in some measure on vasomotor control of certain parts of the blood-circulatory system. Our limited experimental data provide no good evidence of a marked diurnal rhythm in the temperature of adult hank swallows. During the breeding season, at least, the temperature of adult male bank swallows averages slightly higher than that of adult females. Muscular activity such as that engaged in by the bank swallow during flight has a tendency to increase the body heat and may be an important factor in the observed variations and fluctuations of the registered body temperature. During incubation and in other quiescent periods the registered body temperature is lower than in the periods of activity. 176 Roosevelt Wild Lije Annals Body Temperature of Young. — In the course of this study several thousand temperature readings have been taken on young hank swallows of various ages, hut because of limitations of space these can here he presented only in much condensed form. In all cases in which very young individuals were concerned, due care was exercised to select such as had been brooded only a short time before, or from which the brooding bird was flushed at the time. The same instruments and methods were employed as already described for adult birds. One of our first objects in the temperature studies on young bank swallows was to ascertain any changes that might occur in their body temperature from the time of hatching to the time of flight. It was frequently necessary, of course, to estimate the age of the bird at the time the readings were made. Temperature readings taken on several broods of young bank swallows at regular intervals over a period of time varying from 3 to 8 days, showed that the body heat of young birds shortly after they are hatched, even when brooded by the parent, is considerably below the temperature of the adults. As the bird increases in size and acquires its feather covering its average temperature rises until at the time it first leaves the nest it has almost attained the average tempera- ture of adults. During the early nest life of the young birds their temperatures are quickly modified by any change in temperature of the surroundings. However, it is apparent that even at a very early age, different individuals subject to essentially the same temperature conditions created through the brooding parent, also may exhibit varying degrees of bodily heat. In general, this variation is no greater than that exhibited by the adults. As in the adults, too, marked and rapid fluctua- tions of body temperature are frequently to be noted. A condensed summary of the temperature data obtained on 501 young bank swallows appears in Table No. 43 (p. 179), including records for both the Lake Okoboji and Oneida Lake regions. Since in most cases it was not possible to know the exact age of the individual bird, in making our groupings we arbitrarily arranged all of them into four categories. After having handled some thousands of young bank swallows the worker becomes more or less proficient in judging their approximate ages, and therefore the estimated age assigned probably in no case would be in error by more than 1 or 2 days. While the age groupings were made without any such thought in mind, it will be noted that the increasing temperature averages for these groups correspond rather closelv with the temperature increases for individuals as represented in Table 42. Studies ill! the Hank Sv wallow •a j. „• ■ oo s £o § o" o g S on a C „ C S^H 2 T3 CO rf « 00 C 3 £ f*) 3 +j O «rf V-. 3 o o P. rt as' 5P c (a - u c3 < O QJ O •- C O 'O 3 3 i/l tN ► r. §2 ^2 00 On I I I ■ r*} C*J f*5 fO o o o o o \0 'O ~o o *o U"; IT) IT) IT) Roosevelt Wild Life Annals 3 O £0 -C 5 S- 3 C u 0* v 0< r*5 c C C n i** ^ ^, o o »t X O ^, «-j tJ. rv] ^ 5- * lO o io LO i*0 O *■* ^ H l^, ft 1^00^0*00 X t- — 0^ X :> c o c o © X ~- x »o iO c-i (0 p*3 QiCU §8 E « 8 ■- be etc rt X X A X C o «c o o o Studies on the Bank Swallow Table No. 43. Condensed Summary of Temperature Data Obtained ox 501 Young Bank Swallows. Records from Seasons of 1925, 1926 and 1927, Lake Okoboji Region, Iowa; and Season of 1931, Oneida Lake, New York. APPROXIMATE AGE OF BIRD Locality and date Number individual readings Maximum reading Minimum reading Average Grand average, all birds this age Lake Okoboji, 1925. . Lake Okoboji. 1926. . Lake Okoboji, 1927. . Oneida Lake, 1931... 43 13 24 36 °F. 105.2 104.2 103.6 102.5 "P. 94.4 96.0 94.2 92.0 °F. 100.98 101 .77 100.34 97.89 99.89 Fledgling (5-9 days) Lake Okoboji, 1925. . Lake Okoboji, 1926. . Lake Okoboji, 192 7. . Oneida Lake, 1931. . . 98 6 28 42 104.6 104.5 105.6 104.3 94.8 102.2 101 .5 98.9 102.27 103.76 104.17 101.83 102.51 Juvenal (10-15 days) Lake Okoboji. 1925. . Lake Okoboji. 1926. . Lake Okoboji, 1927. . Oneida Lake, 1931. . . 32 11 24 22 109.2 107.8 108.4 106.4 102.0 101.9 102.8 101.4 105. 10 105.30 104.85 103.05 104.56 Lake Okoboji, 1925. . Lake Okoboji, 1926. . Lake Okoboji, 1927.. Oneida Lake, 1931... 37 63 8 14 109.1 110.7 107.4 107.5 102.8 103.4 100.8 103.1 108.80 105.33 105.55 104.30 106.24 Variation and Fluctuation in the Temperature of Young Bank Swallows. — In the course of our temperature studies on young bank swallows it was noted that fluctuations in the thermometer readings in some instances were even more marked than in the adults. A number of examples are here listed: Nestling 97359A, about two days old ; one of a family of four; roadside gravel pit, Lake Okoboji, July 4, 1927, 8:35' A. M. Maximum air temperature for day, 86° ; minimum, 56° ; temperature of bird, 95.8°. Although the air at this hour was warm, a cool wind was blowing and after the small naked bird had been exposed to it for ten seconds its body temperature showed a decrease of 2.4 degrees. Fledglings 97522A. 97523A and 97524A, roadside cut near Lake Okoboji, July 6, 1927, 11 :40 A. M. Maximum air temperature for the day, 87° ; minimum, 55°; temperature of fledglings 104.0°, 103.8° and 105.6°, respectively. Imme- diately after banding they were exposed to the hot sun for two minutes, where- upon the temperature readings were as follows: 108.4°, 108.4° and 107.2°; or increases of 4.4, 4.6 and 1.6 degrees respectively. Immature bank swallow No. 251247; gravel pit near Spirit Lake, Iowa; 11 :00 A.M., July 9, 1927. Maximum air temperature for day, 81°; minimum. 56°. Banded, but temperature not recorded. Immediately after banding, this individual, which could fly well, was placed in the collecting net and left to flutter, exposed to the hot sun, for five minutes, whereupon it gave a temperature reading of 115.0°F. The bird was then held in the hand and shaded from the sun for two minutes, during which its temperature fell to 112.8°. Upon release it flew away, apparently none the worse. Two other immature bank swallows captured at the same time were left to flutter in the collecting net, in the hot sun, for about four minutes, whereupon one of them showed a temperature of 112.4°; the other, 112.6°. The above record of 115.0° is the highest obtained for a bank swallow during the course of these studies. It will be recalled that previously the highest temperature reading that had been obtained for an adult swallow when exposed 180 Roosevelt Wild Life Annals to the hot sun, and struggling to escape from the collecting net, was 112.4°. Perhaps this may indicate a more efficient temperature regulatory mechanism in the adult than in the immature at this stage; or that fluctuations of temperature in the immature bird may be greater. The prompt response of young birds to changes in the surrounding air temperature is well illustrated by a family of four fledglings, Nos. 46471F to 46474F. handed in the Oneida Lake region at 11:15 A.M., June 18, 1931; air temperature, 90° F. Xo. 4C471F; temperature, 102.6°; bird held in operator's hand, in shade. In 20 seconds its temperature had fallen to 101.8°; in another 20 seconds, to 100.4°. Bird then held in the hand, in the hot sun, with the following readings at 20-second intervals: 100.5°, 101.8°, 102.1°. No. 46472F ; bird held in shade ; readings at 20-second intervals as follows : 102.4°, 101.6°, 100.8°. Bird then exposed to sun; readings at 20-second inter- vals: 101.4°, 101.4°, 102.3°. No. 46473F ; bird held in shade ; readings at 20-second intervals as follows : 102.4°, 101.6°, 101.6°. Bird then exposed to sun; readings at 20-second inter- vals: 102.3°, 102.0°, 102.3°. No. 46474F ; bird held in shade ; temperature readings at 20-second inter- vals as follows: 100.8°, 100.0°, 100.0°. Bird then exposed to sun; readings at 20-second intervals: 100.5°, 101.2°, 101.4°. For examples of variation in the temperature of young bank swallows of the same brood, when exposed to apparently identical conditions, the reader is referred to Table Xo. 42 (p. 178), 'where it is shown that variations of 1 to 3 degrees frequently occur, and may probably be termed normal. However, varia- tions of 4 to 6 degrees and even more are not uncommon. Conclusions Regarding the Temperature of Young Bank Swallows. — Varia- tion in the temperature of immature bank swallows is more marked than in adult individuals, and may have a range of 25 degrees and possibly more: Maximum, 115°F. ; minimum, — 90°F. for surviving young. Sudden and well-marked fluctuations of temperature in young bank swallows may be more pronounced than in adults. During the nest life of the bank swallow its temperature control mechanism apparently continues to increase in efficiency, so that by the time the bird is able to fly it presents a temperature range almost within the limits of that of the adult. The average temperature of young birds at the time they first leave the nest (106.24°) is slightly below that for adults (107.1°). The average rate of increase is about 0.5 degree per day until flight ability is attained. These temperature studies indicate that during nest-life the young bank swallow changes from a more or less poikilothermous to a more or less homoiothermous condition. From a phylogenetic standpoint this feature is of interest in that it appears to offer additional ground for the belief that the precursors of birds were poikilothermous. BREEDING HABITS OF BANK SWALLOWS Burrows. Method of digging. — With the onset of the breeding season the first excavating efforts of the bank swallow are intermittent and seemingly aimless. A certain amount of lengthening and renovating of old burrows was noted during Studies on the Bank' Shallow 181 the first week of May, but not until about May 10 did the excavation of burrows begin in earnest. Frequently several individuals congregate in a particular part of a sand pit or on a given section of creek bank, with much chattering and fluttering and occa- sional mating. Others cling to the face of the bank to rest or perhaps to explore it for a spot in which to begin excavating. Sections of bank may become thickly scarified as a result (Fig. -18). When starting a burrow it appears that the birds first cling to some slight projection on the face of the bank from which they can reach the point of attack, either with the claws or bill or both. After a time a slight concavity is formed, its sides creased with the marks of claws and bill. As the work proceeds, the ceiling takes on a distinctly arched form, while the floor is practically flat. Observations indicate that the more deeply scarred appearance of the inner or apical part of the cavity is due to the use of the bill. The bird clings to the walls and dislodges the particles of sand or gravel by pecking with a rapid side-to-side movement of the head. As soon as a shelf has resulted the feet with their long sharp claws are brought into action. Both bill and claws of captured individuals were sometimes caked with moist earth. In digging, the tail is frequently used as a support, as in the woodpeckers. Dissection of a few specimens taken showed that both male and female take active part in the task of digging. Often one individual of the (apparently mated) pair remains clinging to the face of the bank immediately beside the one that is working. When the latter hesitates or flies away its place is immediately taken by the other. Until about May 15, the bank swallow seems to have a strong impulse to dig, but many more burrows are begun than are later occupied. However, as the season advances the impulse evidently becomes stronger, and excavating is con- fined to single burrows which increase in depth rapidly and steadily. At this time, too, the birds exhibit marked sensitiveness to disturbance. With the deepening of the excavation the dislodged materials fall to the floor of the shallow burrow whence they are ejected by frequent vigorous kicks accompanied by a kind of wriggling movement of the body. In this action the wings also are "shuffled" rapidly in a backward and forward direction as well as from side to side, thus aiding in whipping the sand out behind the bird. Ordi- narily, excavating is interrupted by frequent flights, probably at times to feed, but at other times for no very evident reason. When a swallow returns to a burrow under construction it often enters imme- diately, and at once sends forth a shower of sand from the burrow by the rapid backward kicking action of its feet. After a few seconds, digging evidently is resumed and a new pile of loose material results ; this is ejected by the same individual or its mate. In colonies where the burrows are close together the considerable amount of sand which is thus dislodged from them often accumulates in distinct windrows on the bank below. When old burrows are renovated, the labor involved, in many if not most instances, is probably less arduous, the task then being mainly that of enlarging or extending the tunnel, and clearing out the remains of the old nest as well as other debris. 182 Roosevelt Wild Life Annals Since, in the course of our studies, it became necessary to identify individual burrows, various methods of marking them were tried. Because of the continual crumbling of the face of the banks, marks scratched on its surface (Fig. 50) were not always to be depended upon. The most satisfactory results were obtained by using wires which were inserted into the banks. To these, tough brown paper tags were attached, on which notations could be made in pencil, if desired. If the face of the bank crumbled somewhat, the wires remained in place, or required merely to be pushed in a little deeper. Ordinarily, when a burrow in process of excavation was enlarged by us to permit examination, it was abandoned by the swallows. Usually, too, a burrow thus opened, even though it might contain partly incubated eggs, would be aban- doned, particularly if any of the eggs had been broken by our procedure. But if the eggs were about ready to hatch, or if the nest contained young, any ordinary interference with the burrow did not cause the birds to abandon it. In some cases where burrows were dug into and enlarged the birds lengthened the burrow and built another nest. So far as the swallows themselves were concerned, after having been netted once they usually became exceedingly warv and difficult to capture a second time. Rate of Excavation. — In an attempt to obtain information as to rate of burrow excavation, certain sections of gravel pit or creek bank which promised to be concentration points for bank swallows, were marked off, and conditions noted at more or less regular and frequent intervals (Fig. 37). The following table shows the progress of digging in certain groups of burrows at four different field stations. These examples will serve to illustrate the generally prevailing conditions. Table No. 44. Depths and Rate of Excavation of Bank Swallow Burrows. DATE Locality Burrows: Depth in Inches Remarks A B C D E F G H May 7 14 21 29 Field Station No. 5: Daniels farm 2 12 20 20 4 18 18 22 Nests were built and eggs laid in both burrows Mav 4 11 18 26 Field Station No. 10: Grems sand pit No burrows on this date Burrow E with freshly constructed nest; no eggs Burrow A contains nest with 3 eggs; bur- row B, nest with 4 eggs; burrow E, 5 eggs. Shorter burrow due to slumping of bank 5 31 31 13 36 36 3 30 30 1 25 30 20 23 22 May 12 14 15 18 21 26 June 2 Delahunt sand pit 3i 6 8$ 16 30 35i 39 31 32 36 39 45 45 45 26 28 30 37 42 42 44 No nests in any burrows Burrows C and D end in a single terminus Nests in burrows B and C Birds in burrows A, B, C and F today li 65 15 28 28 28 i 1} 1} 22 25J 28 3 1 3 I i 27 32 \ 2 3 29 37 40 1 15 21 21 May 14 Field Station No. 3: Fox farm 3 21 l,X Soil fine, loose, sandy 183 Fig. 48. Bill and claw marks of bank swallows in sand-clay bank where the birds have begun excavating burrows. South bank of Fish Creek at Graham farm. June 3, 1932. Fig. 49. Field horsetail (Equisetum arvense Linn.) growing in mouth of bank swallow burrow. The burrow in the center has been enlarged by the observer. The shallow concavities appearing at both right and left represent beginnings of bank swallow burrows. July 8, 1931. 184 Fig. 50. Marked bank swallow burrows, and various items of field equipment includ- ing trowel, thermometers, collecting net, exploring wire, bird bands, etc. The burrow set off by the circle is the one from which repeat adult No. 56549B was recovered after it had been transported by automobile to a distance of 11 miles (p. 22(>). North bank Fish Creek at Fox farm. June 18, 1931. Fig. 51. Bird bands (on bent wire) and pliers used in closing the bands; two thermometers of clinical type (not used in the present study) ; and one (longer) thermometer of design used in the present study for recording fluctuating temperature of bank swallow. Studies on the Bank Swallow 185 Points of interest in the above table are : 1. Excavating activities were most marked between May 7 and 21. 2. Little excavating was accomplished after May 26. 3. In general, burrow excavation progressed at the rate of 3 to 4 inches a day. 4. The rate of excavation was greatest in a burrow at the Fox farm which was increased 65 inches between May 14 and 21. This also was the deepest burrow that we discovered. 5. Ordinarily the rate of excavation is greatest in soft sandy soil. Depth. — On or about May 12 most of the new burrows ranged from 3 to 24 inches deep, while a few were as much as 30 inches ; the last were mostly old burrows. Occasionally the birds constructed a nest in a burrow not more than 16 or 18 inches in depth, but usually the completed burrow was from 22 to 36 inches deep. For 89 occupied burrows in which temperature readings were taken, the average depth was 28 inches, with a minimum of 15 inches and a maximum of 47 inches. Burrows 40 to 50 inches in depth were not uncommon. Our observations indicate that burrows which are begun early in the season, in soft sandy soil, are deeper as a rule than those begun later in the season after ovulation has started. In general, too, excavations carried on in fine sandy soil are likely to be deeper than those in harder or gravelly soil. Further History. — In general the burrows are excavated directly into the bank, that is, with the long axis approximately at right angles to the face of the bank. However, an occasional one may form an acute angle with the face of the bank and thus comes to intersect other burrows in its course. Sometimes also the sandy partitions between adjacent burrows are so thin that they give way, two or more burrows thus ending in a common chamber with the nests lying side by side. While there is no hard and fast rule as to the course which a burrow may take after excavating has begun, more often than not the terminus is higher than the mouth. Such a construction prevents the entrance of water into the burrow during heavy rains. However, it has the disadvantage of rendering more difficult the return to the nest of impatient young which may have scrambled to the burrow entrance to await the parents' return with food. Sometimes such young lose their footing and fall from the burrow; again they may lack sufficient strength to push their way up-hill to the nest and so succumb to hunger or exposure near the burrow entrance. But not infrequently burrows are directed more or less downward from the entrance while others are continued on the same level throughout their extent. Some curve more or less to the right, others to the left. It is possible that a marked deflection to the right or left may be due to having encountered some obstruction during the digging. Occupied bank swallow burrows are not always clear of obstructions which more or less hamper ingress and egress of the occupants. Stones often jut into the tunnel, and occasionally rootlets form a partial barricade, particularly in burrows lying a short distance below the surface of the ground. Freouently the entrances of burrows become partially concealed and choked by the foliage and rootlets of field horsetail (Equisetum arvense), red sorrel (Rumex Acctosella) and other vegetation that has gained a foothold on the face of the bank. Roosevelt Wild Life Annals Often the terminus of the burrow is somewhat enlarged so that the nest rests in a dome-shaped chamber. After the nest has been occupied by the incubating bird for some days it may settle somewhat into the sand, so that its margin rises only a little above the floor of the burrow. As a result the chamber is more roomy. Often also the nest is so placed that a few inches of space remain between its margin and the inner extremity of the burrow. We found that an incubating or brooding adult when disturbed by us frequently would leave the nest and crouch in this narrow space. Sometimes a swallow would persistently refuse to flush, and in order to effect its capture it was necessary to enlarge the burrow sufficiently to enable us to reach the bird with the hand. Late in June and early in July, after the young have left the nests, many of the burrows become filthy with excrement. This results from the fact that the young, after flying about for a time, return to rest in some convenient burrow — not neces- sarily the one in which they were reared. And young representing several families may congregate in the same burrow, for a time. In addition, the decaying nest materials together with an occasional broken egg or a dead nestling combine to render some of the burrows rather foul. After the first-brood young have left, some of the burrows are rehabilitated for occupancy during July when a considerably smaller second brood is reared by a limited number of the swallows of the colony. Few new burrows are excavated for second-brood rearing. As is the case with older burrows that are rehabilitated, the restoration of the current season's burrows consists principally in deepening them and in bringing in fresh nesting materials. However, the nesting birds may evidently occupy, without "repairs," burrows and nests in which a brood of young has been reared earlier in the season. Temperature of burrows. — Subterranean nest locations such as those of the bank swallow are rather unique and, so far as I have been able to learn, observa- tions on such matters as the temperature in the burrows of this bird during the period of occupancy are not available. Therefore one of the aims in the course of the present study was to inquire into the thermometric conditions under which adult and nestling birds exist while in these domiciles. For this purpose mercury thermometers were used. W hile the results may not have been so exact as those that might have been obtained with the use of an electric thermocouple outfit, nevertheless it is felt that they are sufficiently accurate to permit drawing reasonably safe conclusions. Either before or immediately after a temperature reading was taken, the burrow was examined as to its bird occupants, their number and approximate age. On sunny days this often could be done simply by directing the light from a hand mirror into the burrow, when the occupants could be seen and counted without further disturbance (Fig. 40). If this information could not thus be obtained, the temperature was first taken and the burrow then enlarged to the necessary extent. In addition to the number and size of the birds within, the depth or length of the burrow and its distance below the turf were noted. In most instances, after Studies on the Bank Swallow 187 the temperature of an occupied burrow had been taken, the temperature was imme- diately taken, also, of an adjacent but unoccupied burrow of approximately the same length and distance below the turf. The thermometer used in taking the burrow temperature was an unmounted glass laboratory instrument 6 mm. in diameter and 348 mm. (13^4 inches) long, graduated in the Fahrenheit scale from 0 to -(-300 degrees. This thermometer, attached to a slender bamboo rod or a yardstick (Fig. 53) was introduced into the burrow, to its extremity, care being taken that the bulb of the instrument did not actually crime in contact with the wall of the burrow or with any part of the nest or occupants within. After two minutes or more the thermometer was withdrawn quickly and the indicated temperature recorded. Usually, before any burrow temperatures were taken another thermometer was suspended on a short wire from the face of the bank at the mouth of the burrow (Fig. 52). The instrument used for this purpose was a Fahrenheit maximum thermometer (1-degree divisions) mounted on a 12-inch aluminum plate. It was of the type designed and used for meteorological observations. The simplicity and portability of the equipment above described are features in its favor so far as field investigations are concerned. During the months of May, June and July. 1931, a total of 90 temperature records was made in bank swallow burrows occupied at the time by adult birds only, by young birds, or by both adults and young. In the same period 89 such records were made for unoccupied burrows immediately adjacent to the occupied ones and of approximately the same length and distance below the surface of the ground as the corresponding occupied burrows. The burrows concerned in the present study ranged from 15 to 47 inches in length and lay from 7 to 40' inches below the surface. The average length of the 90 occupied burrows for which measurements were recorded was 28 inches, while their average level below the top of the bank was 17 inches. A detailed discussion of the subject of the burrow temperatures need not be presented here, but for the purpose of illustration, twelve typical readings have been assembled in the following table. 188 Roosevelt Wild Life Annals Table Xo. 45. Temperature Records of Bank Swallow Burrows (Occupied and Unoccupied) . Oneida Lake District. 1931. DATE Hour Locality on Fish Creek Burrow tempera- ture. °F. Air tempera- ture. °F. Length of burrow. Inches Distance below surface. Inches Remarks Mav 27 3:00 P. M.. . 58.0 56.0 70.0 34 18 Occupied by 2 adults; nest, but no eggs Unoccupied June 4 1:15 P. M. . Graham farm . . . 64.0 62.0 79.5 28 18 Occupied by incubating bird Unoccupied 5 3:30 P. M... Daniels farm. . . . 61.5 61 .0 68.0 36 Occupied by incubating bird Unoccupied 9 3:55 P. M... Edell farm 61 .0 59.0 58.0 28 21 Occupied by one adult Unoccupied 10 1:30 P. M... Fox farm 64.0 63.0 68.0 24 13 Occupied by 1 adult, and 3 nestlings 24 hours old Unoccupied 16 3:45 P. M... Edell farm 72.5 69.0 71.0 28 21 Occupied by incubating adult; same burrow as reported on June 9, 3:55 P. M. Unoccupied 22 2:00 P. M... Fox farm 75.0 72.0 73.0 21 10 Occupied by 5 fledglings Unoccupied 2:20 P. M... Fox farm 78.0 72.0 73.0 25 14 Occupied by 5 young about ready to leave nest Unoccupied 25 3:20 P. M... Fox farm 76.0 70.5 73.0 21 10 Occupied by 6 young all able to fly Unoccupied 26 11:55 A. M. Graham farm . . . 71.0 68.0 79.0 28 18 Occupied by 3 young able to fly; same burrow as reported on June 4, 1:15 P. M. Unoccupied Julv 2 2:30 P. M... Fox farm 87.0 80.0 92.0 21 10 Occupied by 5 birds of the year able to fly; same burrow as reported June 22. 2:00 P. M. Unoccupied 14 3:10 P. M... Fox farm 76.0 75.0 80.0 34 18 Occupied by incubating bird Unoccupied Some averages derived from a consideration of all the data are as follows : Average length of the 90 occupied burrows 28.0 inches Average distance below surface of the 90 occupied burrows 17.0 inches Average temperature for all (90) occupied burrows for period 69.6° Average temperature for all (89) unoccupied burrows for period. . . . 67.0° Average temperature for 5 occupied burrows, May 57.1° Average temperature for 6 unoccupied burrows. May 56.5° Average temperature for 63 occupied burrows, June 67.6° Average temperature for 61 unoccupied burrows, June 66.0° Average temperature for 22 occupied burrows, July 78.0° Average temperature for 22 unoccupied burrows, July 75.3° 189 Fig. 53. Bank swallow burrows enlarged so that the nestlings may be removed for examination. Each burrow is marked with one or more tags. South bank of Fish Creek at Graham farm. July 14. 1931. L90 Roosevelt Wild Lijc Annals The highest temperature recorded for an occupied bank swallow burrow was 87°F.. at 2:30 P.M. on July 2, 1931. The air temperature at the mouth of the burrow was 92°F., the highest for any day on which field observations were made during the summer. This burrow was located in a high sandy bank along Fish Creek and was 21 inches in length and 10 inches below the surface. At the time the temperature readings were made, five young of the year, able to fly, occupied the burrow. Readings on an adjacent unoccupied burrow of approximately the same length and at the same depth below the surface showed a temperature of 80°F. The lowest temperature recorded for an occupied hank swallow burrow was 51°, at 8:00 A.M. on May 26, 1931. The air temperature at the mouth of the burrow was 56.5°, which was not as low as on certain other days on which tempera- tures of burrows were taken. This burrow was in a small sand pit four feet from water; it was 42 inches long and 16 inches below the surface. A single adult occupied the burrow. In an adjacent unoccupied burrow of the same depth and location the temperature was 50°. The highest temperature recorded for an unoccupied burrow was 80°. Four readings of this maximum were recorded on July 2, 3, 8 and 9, at 2:30, 2:45, 4:00 and 3 :45 P. M., respectively. Air temperatures at the mouths of these burrows were, in order, 92°, 87°, 84° and 82°. All these burrows were comparatively short (19 to 30 inches) and were not far below the surface (7 to 16 inches), so that their temperatures were quickly affected by the prevailing air temperature. The temperature of short burrows is more nearly the same as that of the outside air than is that of long ones, irrespective of the depth at which they may lie below the surface. The temperature of burrows is fairly uniform when they are occupied by young, so that these are exposed to a more uniform degree of warmth than are the young of species reared in open nests and perhaps subject more or less to the temperature fluctuations in the outside air. Burrows occupied by single adults were, in general, from five-tenths of a degree to two degrees warmer than were unoccupied burrows of the same length and lying at corresponding levels below the surface of the ground. Evidently this is due to the heat generated by the body of the bird. W hile in most cases the difference in temperature between an occupied burrow and an unoccupied one of the same length and depth varies from 1 to 3 degrees, a few readings showed as much as 4 to 6 degrees difference. The greatest differ- ence was found at 2:10 P. M. on June 26, 1931, where in a burrow 15 inches long and 14 inches below the surface, and containing three young able to fly, the tem- perature was 78°, while in an adjacent unoccupied burrow it was only 70°. At the mouth of the burrow the temperature was 81°. It is at once apparent that the temperature of those bank swallow burrows which lie at a deeper level remains more uniform than that of burrows nearer the surface ; and those burrows are also cooler on hot days and warmer on cool days than are the burrows near the surface. When warm weather arrives the burrows near the surface become warmer during the day than those at lower levels. But there is no evidence that the incubation period is influenced by these warmer burrows. 1V1 Fig. 54. Two bank swallow nests from same burrow. The nest on the right is from a previous season. The other was built in the old one between May 26 and June 2, 1931. Grems' sand pit. Fig. 55. Nest of bank swallow with one egg. Note absence of feather lining in nest. Fish Creek Landing. May 19, 1931. Fig. 56. Left: Incomplete clutch of eggs. Nest composed of dried grass and weed stems; no feather lining. May 22, 1931. Right: Nest with the usual complement of eggs in the Oneida Lake region. Nest composed of weed stems and dried grass. May 28, 1931. Fig. 57. Left: A single white feather from a domestic fowl has been added to this nest. May 28, 1931. Right: Another nest in which six white feathers of a domestic fowl have been added. June 2, 1931. Studies on the Hank Swallow 193 Our data show that where the hot sun heats against the face of a hank the average temperature of the burrows is raised, even though the burrows may be fairly long and lie at a deeper level. Other things being equal, the temperature of a burrow is apparently directly proportional to the number and size of birds inhabiting it. (See table No. 45 and summary immediately following it.) The amount of heat generated by the bodies of three or four small nestlings (3-5 clays old) appears to have about the same effect on the burrow temperature as does that emanating from the body of a single adult bird; but the heat from the bodies of four or five young at the age when they are about ready to leave the nest produces a burrow temperature considerably higher than does that of a single adult. Nests and Nest Building. — The nest of the bank swallow is a flimsy affair placed near the inner end of the burrow. Usually a little space intervenes between the margin of the nest and the walls of the burrow. The nest is prepared by the combined efforts of male and female. The form of the newly built structure is rather flat, but after a period of occupancy a relatively deep nest cavity results. It appears that occasionally a nest of the previous season is renovated for occupancy, but in most cases the old nest materials are ejected. On several occa- sions, however, old nests were found to have been used as foundations for the new ones (Fig. 54). The type of material used depends more or less upon what is available in the near vicinity. If a freshly mown hayfield is at hand, grass stalks will be gathered from this source, both green and dried blades being acceptable. Long, thin blades or stems seem to be preferred. If a farmyard is near, a considerable amount of straw is likely to he found in the nest. Weed stalks also are used and some of these are of surprising length and size for so small a bird to manage within the limited confines of the burrow. For example, a dried grass stem 25l/2 inches long was found in a burrow on the north bank of Fish Creek. Rootlets, horse hair and pledgets of sheep's wool often form a part of the nest. Some nests, however, are composed almost entirely of blades of grass. In July, particularly, the dried, black rootlets of the field horsetail, which dangle from the sandy banks in so many places along Fish Creek, frequently comprise a large part of the nest material. The birds doubtless secure much of their nest material while they are on the wing, but I have often observed them on the ground in front of their burrows, picking up dried weed or grass stalks. Sometimes they even zvalk a few steps over the sand in their search for suitable materials. When the nest is finished the female begins to sit in it preparatory to egg- laying. During the egg-laying period two swallows were frequently captured from a single burrow, and this seems to indicate that the male remains with the female, part of the time, at least, during this period. The first eggs — and sometimes all in the clutch — are usually laid in the unlined nest (Fig. 55). However, as egg-laying and incubation proceed, feathers are added as a lining, so that by hatching time a warm and cozy bed awaits the young (Figs. 58 and 59). As an example may be mentioned a certain nest which on 194 Roosevelt Wild Life Annals June 10 contained five eggs, but no feather lining. Five days later there were five recentl) hatched young, and the nest had a generous lining of feathers. Although pure white feathers from the domestic fowl are most frequently found in the nest, there also occur at times gray, brown or black ones. The feathers may vary in size from filmy down to large quill feathers. Sometimes, indeed, a nest appears to consist largely of feathers. The poultry yards from which most of the feathers evidently are gathered, in some instances are more than a mile ili slant from the nest site. In this respect the bank swallow, as I have observed it, differs from the tree swallow as reported by Robinson (1927), which did not lay eggs until after the feather lining of the nest had been provided. While definite evidence on this point was not obtained, there is a possibility that the collecting of the feathers is done principally by the male. As previously remarked, early in the egg-laying and incubating periods both male and female often are found in the burrow together, but after these activities are well under way it is seldom that more than one of the pair is found in the burrow at any one time. This was determined by inspection, a beam of light being directed into the burrow by means of a hand mirror. When both birds were present, one of them, presumably the male, usually sat on the edge of the nest; the other was in the nest. Then, as the period of egg-laying advances the male appears to spend less time in the burrow, but his visits thereto may be indicated by the gradually increasing number of feathers which come to line the nest. Eggs and Egg-laying. — The earliest date on which a nest containing eggs was found was May 19, 1931. On this date at the Fish Creek Landing colony I opened a burrow which contained a single egg reposing in an unlined nest. Regarding color, in only one instance have I found any departure from the usual white of the eggs. A nest at the Edell farm colony contained five eggs on May 27, 1931, and two of these had numerous minute, rounded, elongate or irregular pale brownish marks. Four or five seems to be the usual complement of eggs in a first brood clutch ; three or four in the second. An unhatched egg usually was present in each of the more than 200 nests bearing young that I examined before June 30, 1931. Regarding the number of young in a nest, less than half a dozen nests contained six young, about a dozen contained five, and the rest had only three or four. Quite commonly, in northwestern Iowa, clutches of five or six eggs and broods of four or five young were found. fudging from the comparative size of the hatched young in the broods, the eggs of a first clutch are ordinarily laid within a very short time of one another; apparently, on the average, an egg a day until the full complement is laid. However, it sometimes happens that a considerable interval elapses between the first and the last laid egg of a clutch. This is illustrated by the extreme example that came to our attention on June 22, 1931, in which the smallest young one in a family of five had been out of the shell but a few hours, while the largest was at least 6 or 7 days old. The laying of the second clutch is likely to be more irregular than that of the first. It should be stated here that some of the sets of eggs and small nestlings that arc found in early July, may possibly represent only delayed breeding activities Studies on the Bank Swallow 195 of certain pairs. However, where a second set of eggs is found later in the same nest (as for example, in an instance at Fish Creek Landing, June 9 and July 8, 1931), the inference is justified that it is a second clutch. The hatching period for these late egg-clutches, whatever their explanation, falls principally hetvveen July 15 and 25. IV eight of Eggs. — Regarding the weight of hank swallow eggs, a total of 65 from 16 different clutches gave an average weight of 1.42 grams. This included one probably infertile egg which weighed only 0.8 gram. The maximal weight for any one egg was 1.7 grams; the minimal weight (for eggs assumed to be fertile), I. 2 grams. There were three of this minimal weight, one of which was found to be about ready to hatch. A marked uniformity was found in the weight among the eggs of a given clutch, the difference in no case exceeding 0.3 gram, and in most cases being not more than 0.1 or 0.2 gram. The eggs that were weighed were also measured and found to vary between II. 5 mm. and 14.0 mm., in the short axis; and between 16.0 mm. and 21.0 mm. in the long axis. A number of newly hatched young weighed 1.2 grams, while others that evidently had been out of the shell for some hours — less than 24 — weighed 1.3 to 1.6 grams. Incubation. — While there had previously been little reason to doubt that the male bank swallow takes part in incubation and brooding, this fact was established in three instances in which the bird found upon the nest at the time was killed and dissected. Two of these individuals were sitting on eggs, the third was brooding young. In the bank swallow the ventral unfeathered area or apterium is of relatively large size. Its size is particularly accentuated in the female by the distension of the abdomen, so that the heat from the body of the incubating bird is consequently applied directly to the eggs rather than through any considerable layer of inter- vening feathers, and thereby perhaps the temperature of the eggs can better be maintained. The period of incubation in the bank swallow doubtless varies within certain limits, — determined by a number of factors — but the more or less unsatisfactory evidence on this point that was secured during the present study, indicated it to be about 14 to 16 days, counting from the date on which the last egg was laid to that of the first hatching. June 4 (1931) was the earliest date on which young were found in the nest. On this date a family of four birds which apparently had been hatched the day before was found in a burrow at the Graham farm colony. However, the next day. at the colony on the Daniels farm, three young were found that appeared to be 48 to 60 hours old. These birds had, therefore, been hatched about June 2. Still earlier hatchings may have taken place, but we did not happen to find them. It was later found that the peak of the hatching period in the colonies in the Oneida Lake region came between June 10 and 20. Young. — During its first few days of life the young bank swallow is brooded almost constantly by one or the other parent. General observations indicated that 1<>.) Roosevelt Wild Life Annals the parents probably share about equally in the task of carrying food and in removing pellets of excrement from the nest. Reactions and behavior of nestlings. — By means of a flashlight or a pocket mirror it was possible, in many instances, to see and even count the young in the nest. Any slight disturbance at the mouth of the burrow was sufficient to cause them to stretch their necks and call noisily. A beam of light directed into the burrow frequently had the same effect. Usually the nestling expels a pellet of excrement immediately after being fed, and this pellet is then carried away by the parent upon its next visit. The pellet may be dropped just outside the mouth of the hurrow, or carried farther — often dropped into the water as the bird flies above the creek. Possibly, also, as in some other passerines, the parent may swallow some of the pellets, especially those appearing during the first few days of the nestling's life. The young usually discharge the pellets upon the edge of the nest, and seldom in it. With young approximately 7 to 12 days old the pellets often are deposited in a little group a few inches from the nest. By aid of a beam of light, young were observed to crawl out of the nest to defecate, and then to shuffle back again. The returning parents begin at once to remove the pellets, and keep at the task until it is finished. If the excreta in young birds accumulate for a time in the cloaca they are contained in a gelatinous sac when voided. However, if through fright or other causes the young void the excrement at frequent intervals, as sometimes happens when they are held in the hand, it is then expelled in a semi-liquid state. Young bank swallows which are fully fledged, but not ready to fly, often rush to the mouth of the burrow, in full cry and with wide-open mouths, as soon as they are aware of the approach of a parent. A beam of light cast into a burrow sometimes brings a similar reaction. Occasionally such eager young birds fall out of the burrow and perish. Fear in young bank swallows seems to appear only after they are well fledged. Young from 8 to 15 days of age when removed from the nest display a marked tendency to crawl under ledges of earth or other objects. When removed a little distance from the home burrow they almost invariably shuffle toward it or in the direction of its location, not away from it. Young birds at this stage of growth, too, have a tendency to shuffle backwards when placed on the ground outside the burrow; but this manifests itself also before they have been removed from the burrow. So far as could be determined in the present study, the average time spent by young in the nest varies from about 18 to 22 days. Behavior and Habits of Young after First Leaving" Nest. — The earliest date for young of the year able to fly was June 18 (1931), when three such individuals were flushed from a nest at the Fox farm colony. These young swallows probably had been hatched about June 1. The following day (June 19) at the Graham farm, also, a few young swallows flew out from their nests, apparently in their first aerial excursion. Possibly these flights were prompted by hunger, for because of our presence the parents had not ventured to visit their young for more than two hours. 197 Fig. 59. Left: Family of five fledglings, Nos. 56656B-56660B. June 11, 1931. Right: Fledgling bank swallow and a nestling about 24 bours old. June 16. 1931. 198 Roosevelt Wild Life Annals On what may be presumed to be their first attempt at flight, these birds (Fig. 62 i launch themselves boldly into the air, with a sharp twitter and vigorous flapping of wings. At first their progress is wavering, but steadies as they gain momentum. They soon tire, however, and usually circle back to the home bank, where they enter any convenient burrow. W hen the young birds first leave the nest, some of the adults that may be flying about in the vicinity dart toward them and circle about, twittering excitedly. A number of the immature swallows which left their burrows at the Fox farm on June 19, 1931, were unable to continue flight and fell into Fish Creek. They made strenuous efforts to reach shore by swimming, the wings serving as the principal means of support and propulsion. Since young bank swallows are fairly good swimmers (Stoner, 1928a, p. 311) they probably would have reached shore safely had they fallen in still water; but the current carried them steadily along, and the numerous eddies proved too much for them. However, one such young bird, which I had just banded at Fish Creek Landing, and which fell into the creek- after a short flight, was more successful. It flapped its way over the surface for a distance of about forty yards and landed safely on the bank. And on July 9, at the Delahunt sand pit, another young bank swallow gave a rather unexpected exhibition of wing power. It had flown only a few yards when it dropped into a pool of water. After flapping along for a few feet it suddenly rose into the air and, evidently gaining strength and steadiness with every second, continued its flight until it was lost to sight in the distance. Almost immediately after hatching, young bank swallows may be heard to utter a fine, high-pitched "cheep" ; but the nestlings are, as a rule, inclined to be silent, although hunger calls are uttered when the adults have not visited them for some time. When a family of four or five nestlings utter this call together, the sound reminds one of the buzzing of a swarm of bees. Occasionally, when well- fledged young are forcibly removed from the burrow, they utter a few loud, shrill cries of distress, — a signal for all the adults in the immediate vicinity to congregate and add their own excited notes to the clamor. But unlike many other passerine birds, the distress calls of the young usually are not long continued, even when the bird is roughly handled. Only a sharp pain then seems to provoke an outcry. Ordinarily, the young individuals that occasionally drop into the water likewise remain silent. For a few davs immediately preceding the abandonment of the nest, the young often appear at the mouth of the burrow, twittering and warbling much after the manner of the adults. At this age their voice is lower and harsher than that of the adult, but within a few days no pronounced difference in timbre was noted. After leaving the home burrow for the first time, the young bank swallows often return to it — or perhaps to one nearby — to rest or to receive food, at more or less frequent intervals. Later they return to the burrows only for the night. Then, as the season advances, less and less time is spent in and about the burrows until, by Tnly 25, approximately, practically all the first brood young have perma- nently left the locality in which they were reared. As has just been intimated, for a few days after their first flights the birds of the year often congregate about their home bank, twittering and flying in and out of the burrows, dusting themselves in the sand and disporting themselves in 199 Fig. 61. Family of four juvenal bank swallows two of which (center and extreme left) show a Protocalliphora larva in the left nostril. Grems' sand pit. July 13, 1931. Fig. 62. Top: Young bank swallows ready to leave nest. The temperature of birds of this age averages only slightly lower than in adults. June 29, 1932. Bottom: Swallow burrows 3 to 5 inches deep, excavated by young bank swallows that have left the parental burrows. July 19, 1932. Fish Creek at Fish Creek Landing. Fig. 63. Adult bank swallows Nos. 56575B-56578B. South bank of Fish Creek at Graham farm June 4, 1931. No. 56576B was recovered as a "return" in the colony where banded, June 3, 1932. Studies on the Bank Swallow 201 apparently care-free fashion. It is while perched on the sandy ledges below or near the burrows that these young of the season first engage in excavating activi- ties. A row of shallow cavities along a ledge apparently represents their first instinctive efforts at burrow digging (Fig. 62). But the cavities are seldom more than a few inches deep, and are used by the birds as resting places. FEEDING HABITS OF ADULT AND YOUNG BANK SWALLOWS During the 1931 season some attention was given to the food habits of the hank swallow in the Oneida Lake region. A number of adults and young were collected, mostly from burrows along Fish Creek, but a few from sand pits in the vicinity. Each stomach was removed immediately after the bird had been killed, and preserved in 70 percent alcohol. Most of the stomachs were collected during the incubating period late in June when the parents were busy caring for their young. It appears that during the incubating period the parents leave the nest at frequent intervals to feed, tor the stomachs of the killed individuals examined usually were found well filled. However, when young are in the nest they do not feed so frequently, evidently because much of their time is occupied with securing food for their offspring. This was borne out by the fact that the stomachs collected in May and early June were, for the most part, well filled, while those collected from mid- June on into July usually were only about one- third full. The method of analysis of the stomach contents was essentially the same as outlined by McAtee (1912, 1920), based on percentage by bulk. The following table gives the age groups, etc., of the individuals represented. Table No. 46. Summarized Record of Bank Swallow Stomachs Collected in the Oneida Lake Region, 1931. AGE OF BIRD Month Totals May June July 7 8 2 6 8 4 3 4 9 3 11 11 6 15 21 Adult 10 Totals 10 31 23 64 In making the stomach examinations the sand, gravel, nacre, bits of wood and other extraneous material were first decanted, strained or picked from the mass ; then the larger particles of food that could be readily identified were sepa- rated in other dishes. Finally the smaller fragments were examined under the binocular microscope and, as identified, transferred to proper receptacles. In some of the stomachs a considerable amount of miscellaneous, finely triturated insect 202 Roosevelt Wild Life Annals matter remained, which it was thought host, in most instances, to consider separately. After separation of the various lots they were placed on blotting paper, a careful comparison made, and percentages estimated on the basis of bulk. Percent- ages of the major food items for both adults and young, for the months concerned, were tabulated, and from these the averages for the season were obtained. For the most part the insect contents were referred only to the order, but in Mime cases family, generic and even specific determination was possible. As is generally known, bank swallows take practically all their food while on the wing. From early dawn until dusk they course over meadows and streams and about farmyards and stock enclosures, but at times they fly high above woodlands. Our observations tend to show that during the nesting season the parent birds confine their foraging flights to the general vicinity of the colony. The food of bank swallows is limited largely to flying and jumping insects, the more strictly terrestrial insects or larvae being largely lacking in their dietary. Occasionally spiders are taken, probably while suspended by their webs. Vegetable matter is largely lacking from their food; yet, as is well known, they sometimes light upon the ground and there pick up bits of vegetable material of various sorts, as well as more or less inorganic matter. Table Xo. 47. Showing Food Items of 64 (21 adult and 43 young) Bank Swallows, Collected in the Oneida Lake Region between May 14 and July 22, 1931. KIND OF FOOD Number stomachs in which present Plectoptera (Ephemeridae) Odonata (Zygoptera) Plecoptera Neuroptera Orthoptcra (Gryllidae) f Homoptera (mostly Cercopidae, Fulgoridae and Cicadel- Hemiptera \ lidae ( Heteroptera Diptera Coleoptera (principally Scarabaeidae, Chrysomelidae and Rhynchophora) Hymenoptera Miscellaneous finely ground insect fragments Araneida Vegetable material 1 1.48 4 0.11 1 0.02 2 0.05 1 Trace 47 17.81 49 6.13 59 31.59 56 36.13 51 5.66 19 0.81 5 0.14 1 0.07 Analysis of Table No. 47 shows that in the period May 14 to July 22, Cole- optera, consisting principally of small dung beetles (Scarabaeidae), leaf beetles (Chrysomelidae) and snout beetles (Rhynchophora), comprised a little more than 36 percent of the food of the sixty-four individuals. Diptera formed the next important item, representing about 32 percent of the food for the period. The third largest item consisted of Hemiptera, including both Heteroptera (true bugs) and Homoptera (aphids, leaf hoppers, spittle insects, etc.), which together amounted to almost 24 percent of the total food eaten. Hymenoptera comprised less than 6 Studies on the Bank Swallow 203 percent of the food, while the remaining animal matter, about 2 percent, embraced representatives of a number of insect orders, and some spiders (Araneida). It is of interest to note further that remains of Coleoptera were found in 56 of the 64 stomachs, Diptera in 59, Hemiptera in 62 (Heteroptera, 49; Homoptera, 47), Hymenoptera in 51, Odonata (dragon-flies and damsel-flies) in 4, and Araneida in 5. Remains of insects of the orders Plectoptera (May-flies), Plecoptera (stone- flies) and Orthoptera were found in one stomach, while representatives of the order Neuroptera (lace-wings, dobson-flies, etc.) occurred in two. Fig. 64. Diagram showing, in terms of percentage by bulk, the food of 21 adult and 43 young bank swallows, Riparia r. riparia, collected in the Oneida Lake region between the inclusive dates May 14 and July 22, 1931. Of the insect groups represented, Diptera and Hymenoptera together, forming a little more than 37 percent of the food matter, can be classed in the main as excellent flyers. Although the Coleoptera, comprising 36 percent of the diet, are not so highly aerial, the small dung beetles, leaf beetles and weevils which were most frequently taken, are good flyers. Some of the Heteroptera (6.13 percent of the food) taken are good flyers, also, while many of the Homoptera (17.81 percent of the food) can fly well and others are highly saltatorial. Although in late May and early June there was a great abundance of May-flies in the localities under consideration, remains of these insects were found in only 204 Roosevelt Wild Life Annals one bank swallow stomach. Likewise other aquatic insects were conspicuously absent from the stomachs examined. Of interest was the presence in most of the stomachs of more or less extraneous material of no food value. Tbe principal item was fine sand, some of which no doubt had been clinging to insects ingested, and possibly some had been inadvertently swallowed while the birds were digging their burrows. However, in some of the stomachs of adults the amount of sand formed so great a proportion of the contents that it could hardly be termed incidental. On many occasions both adult and immature bank swallows were observed resting on the banks near their burrows and apparently picking something out of the sand. Possibly this explains also the presence of small pebbles, fragments of nacre, wood, etc., among the stomach contents. Of the 28 stomachs of young unable to fly, 1 contained fragments of wood; 7, bits of limestone; 9, quartz pebbles; and 2, fragments of nacre. Of 15 stomachs of immatures able to fly, 2 contained fragments of wood ; 4, bits of lime- stone; 1, sizable quartz pebbles; but none of these contained nacre fragments. Of 21 stomachs of adult birds examined, 2 contained wood fragments; 3, limestone pebbles; 3, quartz pebbles; and 1, nacre fragments. The size of some of these pieces of extraneous material was considerable. In the stomach of a fledgling collected July 20, were two pieces of nacre measuring respectively 5 mm. x 8 mm. and 3.5 mm. x 3.5 mm. The stomach of one collected June 24 contained, in addition to four small quartz crystals, a fragment of lime- stone 4 mm. x 5 mm., and that of another contained a fragment 3 mm. x 5 mm., and a quartz pebble of similar size. In addition to the many insects present in the well-filled stomach of a fledgling collected June 19, there was present a small amount of fine sand, several small pebbles and fragments of rock, including one very thin, flat fragment 6 mm. x 0 mm. An adult collected June 9, had swallowed a flat fragment of hard wood 2.5 mm. x 4.0 mm. ; another, collected June 19, contained a fragment 1 mm. x 5 mm. Doubtless the gravelly and other hard particles have a grinding function. It is worthy of note in this connection that in the very young nestlings which are fed principally on soft-bodied insects, such hard parts are usually lacking or few in number, while in the fledglings and juvenals which are furnished with a greater proportion of hard-bodied insects, they are more generally present. Fragments of feathers that were found in the stomachs of four of the young bank swallows may be considered accidental. Although plant material was frequently detected in the stomachs, in only one case did it comprise as much as 5 percent of the contents. Where present it usually consisted of fine rootlets, probably pulled from the nest, and bits of grass blades and plant stems which were no doubt swallowed along with insects snapped up from the vegetation. Si ikI ics on ///<■ Bank Swallow 205 Table No. 48. Summary of the Food of 21 Adult Bank Swallows Collected in the Oneida Lake Region, May 14 to July 22, and 43 Young taken June 10 to July 22, 1931. KIND OF FOOD Plectoptcra (Ephemeridae) Odonata (Zygoptera) Plecoptera Neuroptera Orthoptera (Gryllidae) Hemiptera { Homoptera H \ Heteroptera Diptera Coleoptera Hymenoptera Miscellaneous finely ground insect fragments . Araneida Vegetable material Number Stomachs in Which Present Adults 1 2 1 9 18 19 20 16 4 4 Young 38 31 40 36 35 15 1 1 Food Percentage for Period Adults 4.52 0.05 0.14 Trace 8.24 5.33 27.55 49.76 3.26 0.72 0.43 Fig. 65. Diagram showing, in terms of percentage by bulk, the food of 21 adult bank swallows, Riparia r. riparia, collected in the Oneida Lake region between the inclusive dates May 14 and July 22, 1931. 206 Roosevelt Wild Life Annals Table No. 49. Summary of the Food of 23 Young Bank Swallows Collected in the Oneida Lake Region, June 10 to 30, and 20 Young taken July 2 to 22, 1931. Number Stomachs Food Percentage in Which Present for Period 1\ L .\ 1 J \JV I \ )\ I \ ' June Julv June July 10-30 2-22 Odonata (Zvgoptera) 1 3 0 13 0 20 IT . . J Homoptera Hermptera | Heter(^ptera 18 16 20 15 20 5 48 56 24 7 80 63 22 18 45 65 19 65 17 19 21 35 38 80 Hymenoptera 17 18 5 79 8 02 Miscellaneous finely ground insect fragments. . . . 9 6 1 04 0 65 1 Trace Vegetable material 1 0 25 Fig. 66. Diagram showing, in terms of percentage by bulk, the food of 43 young bank swallows, Riparia r. riparia, collected in the Oneida Lake region between the inclusive dates June 10 and July 22, 1931. Studies on tlw Bank Swallow 207 Remarks on the Arthropod Groups Represented in the Food of Bank Swal- lows.— Plcctoptcra (Ephemeridae). — As noted in the table, remains of these insects (more than fifty individuals) were found in hut a single stomach, that of an incuhating female, collected May 26 (1931), at the Bryant sand pit. The nearest water where May-flies might have been had in abundance was Fish Creek, a half- mile to the south. Perhaps this distance may explain, in part at least, why the bank swallows in this locality had not fed more generally upon these soft-bodied insects, but it does not explain their absence from the stomachs of the birds taken elsewhere. These insects remain abundant in this general territory for about three weeks. Odonata. — In each of two stomachs of young birds the remains of damsel-flies (Zygoptera) comprised 3 percent of the food contents; in a third stomach, 1 per- cent ; and in a fourth only the merest trace. Specific determination could not be made from these remains, but a damsel-fly found on July 9 in a bank swallow burrow containing nestlings was identified by Dr. J. G. Needham of Cornell University as Lestes unguiculatus Hagen. This form I noted to be common in the Oneida Lake region and found remains of it also in other bank swallow burrows. Plccoptera. — Stone-flies occur abundantly in this territory and it was rather unexpected to find that the bank swallows had not fed more largely on these forms than was indicated by the results of the stomach examinations. Stone-flies are seen frequently about the streams where bank swallows feed, and many times were noted clinging to the banks close to their burrows. The only stomach which contained remains of these insects was that of an adult male collected along Fish Creek, May 15 (1931), which contained two adult stone-flies, determined by Dr. P. W. Claassen, Department of Entomology, New York State College of Agriculture, as Alloperla mcdiana Banks. Neuroptera. — Remains of these insects were found in but two adult bank swallow stomachs, amounting to but a trace in one and to 3 percent of the food contents in the other. Orthoptera. — Traces of a cricket (Gryllidae) were found in the stomach of an adult taken on May 15 (1931). Ucmiptera. — For both young and adult bank swallows the bugs ranked third in importance as an item of food, remains being found in all but two stomachs examined. In this territory homopterous insects evidently do not form any appreciable amount of the bank swallow's food before mid-June ; but after that time these destruc- tive forms would seem to constitute one of the important items of diet, if one may judge from a few stomachs examined. For the 13 adults collected between May 14 and June 10, Homoptera made up only 0.46 percent of the food, while for the 8 adults collected between June 16 and July 22 the group comprised 20.75 percent of the diet. Leaf-hoppers (Cicadellidae), spittle-insects (Cercopidae) and lantern- flies (Fulgoridae) were best represented, their relative abundance as in the order given. The stomach of a brooding female collected on July 22 (1931) contained fragments of more than 32 spittle-insects, amounting to 91 percent of the food 208 Roosevelt Wild Life Annals contents. All these insects are small, but they are prolific and where plentiful may cause considerable injury to grain, grasses and other vegetation. All of the 20 birds taken in July had fed on these insects, which amounted to 10 percent or more ot the food contents in each of 13 stomachs, and comprised 24. (SO percent of the loud of the 20 individuals taken in that month. Eighteen of the 23 young birds collected in June bad been fed on Homoptera, which amounted to 10 percent or more of the food contents in each of 8 stomachs, and comprised 20.48 percent of the food of the 23 individuals taken during the month. Among the families of Homoptera represented, remains of tree-hoppers (Mem- bracidae) were found in 1 stomach, plant-lice (Aphididae) in 4 stomachs, lantern- flies (Fulgoridae) in 1 stomach, leaf-hoppers (Cicadellidae) in 19 and spittle-insects (Cercopidae) in 21 stomachs. The Cercopidae do not become an important item of diet of the bank swallow before late June, while in July they constitute a conspicuous part. In this territory it is not until early summer that any considerable number of cercopids leave the masses of white froth or "spittle" which serves as a protective covering during their inactive stages preceding maturity. Progression in the adults is mainly by jumping, in the course of which a good many fall prey to bank swallows. As many as 8 or 10 adults frequently were found in the stomachs of young birds, and in each of three stomachs 32 or more of the insects were present, comprising, respec- tively, 67, 91 and 98 percent of the food contents. The species Lcpyronia quad- rangularis Say was the commonest cercopid in the stomachs examined. Of the families of Heteroptera, representatives of the leaf bugs (Miridae) and of the chinch bug family (Lygaeidae) were of most frequent occurrence. Among the other families represented were the Corizidae, Cydnidae, Pentatomidae and Tingididae. In one stomach, remains of the corizid bug Harmostes sp. comprised 33 percent of the food contents. Members of this genus also were found in other stomachs. In the stomach of a nestling about five clays old was an example of the cydnid Galgupha (aterrima Malloch?). It is noteworthy also that bank swallows add to the diet of their nestlings such small and delicate forms as the lace bugs (Tingididae). True bugs (Heteroptera) constituted a considerably smaller proportion of the bank swallow diet than did either Diptera or Coleoptera. And although heteropter- ous remains were found in the stomachs of 49 of the 64 young, — as against 47 for Homoptera — they made up considerably less of the food than did the latter, aver- aging only 6.52 percent of the diet of all the young taken for the period. In only one stomach did heteropterous insects comprise as much as 41 percent of the food contents, while in each of twenty-three stomachs they amounted to less than 10 percent. All these bugs are plant feeders, and it can hardly be doubted that the bank swallow is an important local check against their undue increase. Diptera. — Remains of Diptera appeared in all but two of the stomachs of the adults examined and in all but three of the stomachs of the young birds. In each of twenty-five stomachs of the latter group, two-winged flies formed 18 percent or more of the food contents, and in each of thirteen stomachs they comprised 50 percent or more. Studies on llic Bank SicalUnv 209 For the twenty-three young taken in June. Diptera made up 45.05 percent of all the food contents, while fur the 20 young collected in July, the percentage dropped to 19.65. As previously suggested, this evidently is to he explained hy the greater proportion of young nestlings making up the June collection, the July birds, averaging several days older, having been fed many soft-bodied insects. Further evidence on this point is furnished hy seven nestlings ranging in age from approxi- mately 24 to 96 hours, collected between June 10 and June 23 (1931). The lowest percentage of Diptera in the food contents of any of these stomachs was 59, the highest 99, and the average for the period 81.85. For six fledglings collected during the same period, Diptera varied in amount from a mere trace in one stomach to as much as 66 percent of the food contents in another, the average being 40.50 percent for the period. For six adults, also collected in the same period, the dipterous contents of the stomachs ranged from 1 to 99 percent, the average for the period being 45.16 percent. These percentages more nearly approach those for young birds of the older age groups than those for nestlings and fledglings. As to the kinds of Diptera found in the stomachs of the young and the adult bank swallows there was no significant difference. Several families and many species were represented. Members of the family Muscidae, which includes the house fly and its allies, were commonest, and among other families were the crane- flies (Tipulidae) and the flower-flies (Syrphidae), the larvae of the last-named being considered economically beneficial in that they often feed upon aphids. As pointed out elsewhere in this report, young bank swallows are often heavily infested with Protocalliphora splcndida (Muscidae). The abundance of these parasitic larvae on some fledglings, and the absence of their remains in the stom- achs examined, suggested that the parent birds made no attempt to remove these maggots from their offspring. Coleoptera. — Insects of this order evidently comprise a very important part of the food of both adult and young bank swallows. The dung-beetles and allies (Scarabaeidae) make up by far the largest share. Remains of these insects were found in sixteen stomachs of the adults examined, and in each of nine made up 25 percent or more of the contents, while in each of four stomachs they comprised more than 90 percent of the contents. Among these beetles, Apliodius haemorr- hoidalis Linn, was the most numerous, hut A. crraticus Linn, was also included, and perhaps other species. These insects, which are mainly neutral in their eco- nomic relations, were taken in greatest numbers in May and June. Leaf beetles (Chrysomelidae) also form a fairly regular part of the bank swal- low's diet — the small, saltatorial flea-beetles (Halticini) being best represented. In the stomach of one young bird, fifteen of these forms were counted. A specimen of Chactocncma denticulata 111. was found in one stomach. Remains of snout-beetles (Rhynchophora) were found in about half the stom- achs examined, members of the genus Sitones of the family Curculionidae being most common. Other snout-beetles found were Rhinoncus pyrrhopus Boh., Ccit- tlwrliynclms marginatus Payk. and Phytononius nigrirostris Fabr. Among the more or less beneficial forms, remains of ground-beetles (Car- abidae) and of rove-beetles (Staphylinidae) were noted. Other families repre- sented were Dytiscidae, Histeridae and Scolytidae. 210 Roosevelt Wild Life Annals Hymenoptera. — Among the insects appearing consistently though in relatively small amounts in the dietary of the hank swallow are the four-winged flies, ants, bees, wasps and allies. Remains of these insects occurred in sixteen stomachs of adults, amounting to 3.26 percent of the food of all the adults taken in the period. In only one stomach did hymenopterous insects amount to as much as 14 percent of the food contents, while in each of eleven stomachs they comprised only 2 percent or less. Ants (Formicidae) occurred in two stomachs, while most of the other forms present were small ichneumon-flies (Braconidae, Ichneumonidae and possibly others). Hymenopterous remains were found in thirty-five stomachs of young birds and amounted to 6.82 percent of the food of the forty-three collected in the period. In three of these stomachs the food percentages represented by insects of this group were, respectively, 41, 43 and 65, while in each of twenty-four stomachs hymenopterous remains made up 4 percent or less of the food contents ; and in each of seventeen stomachs they amounted to 2 percent. While this ratio is about 18 percent greater than that for adults it should be borne in mind that, whereas ants (Formicidae) occurred in but two stomachs of adults and formed only a negligible percentage of the food, remains of these insects were present in nine stomachs of young birds and amounted to approximately 14 per- cent of the ant remains, — an average of approximately 3 percent for the 43 young birds. This means that about one-half the bulk of hymenopterous insects eaten by young bank swallows is composed of ants. Most of those taken were winged individuals. It may be of further interest to mention that not until early July did ants appear in the dietary of the young birds. Doubtless these insects were taken principally during their nuptial flight. Hymenopterous insects other than ants are a standard article of diet, also, for young bank swallows. As with the adults, these consist mainly of small ichneumon- oid forms, of which remains of a dozen or more frequently were found in a single stomach. Most of these small hymenopterans are predacious or parasitic, or both, and so are considered beneficial. The destruction of these insects by the bank swallow affords the basis for the principal criticism that might be raised regarding the economic relations of this bird to man. However, the numbers of annoying and injurious insects that fall prey to it far exceed those of useful forms. Miscellaneous Insect Fragments. — The stomachs of four adult and fifteen young bank swallows contained varying amounts of finely ground insect remains that could not be more definitely allocated, amounting to an average of 0.72 percent of the stomach contents for the total of 21 adults examined, and 0.86 percent for the 43 stomachs of young. Arachnida ( Araneida) . — Spider remains were found in the stomachs of five bank swallows — four adults and one young. The largest amount, in the stomach of an adult, made up 5 percent of the food contents and represented fragments of at least three individuals. The stomach of another adult contained the remains of at Studies on the Bank Swallow 211 least seven small spiders, representing 3 percent of the food contents. Only a trace of spider remains was found in the young hird. From the foregoing it appears that in the region and for the period in question the principal items of diet of hoth young and adult hank swallows are limited to a relatively few arthropod groups, mainly insects, and this apparently was true for the hank swallow colonies as a whole as well as for individuals of a given family. This is indicated, for example, in Table No. 50 (below) which shows in terms oi percentage by bulk the results of the analysis of the stomach contents of indi- viduals representing two families. Table No. 50. Results of Stomach Analyses of the Members of Two Families of Bank Swallows taken in the Oak Orchard District, Showing the Kinds of Food Present and the Amount of each in Terms of Percentage by Bulk. FAMILY GROUP Age Date collected Kind and Percentage of Food Homop- tera Heterop- tera Diptera Coleop- tera Hymenop- tera Miscel- laneous insect fragments Family No. 1 . . . Parent, f. Nestlg. A. Nestlg. B. Nestlg. C. Nestlg. D. June 19. 1931 4 28 38 11 8 49 3 Tr. 25 68 59 89 88 17 2 3 1 3 3 1 Family No. 2 . . . Nestlg. A. Nestlg. B. July 22. 1931 47 43 1 Tr. 46 55 1 1 3 Tr. 2 1 It will be noted that the four nestlings of Family No. 1 received, in general, sim- ilar amounts of the principal groups represented ; and that the food of the female parent of one of these families consisted, for the most part, of the same kinds of insects as those fed to the young, with Coleoptera as an addition. In the two nestlings of Family No. 2 the greater proportion of Homoptera was probably due to the greater abundance of these insects at that season of the year. Summary. — Summarizing briefly, it appears then, so far as these limited obser- vations extend, that there is a relative uniformity in the diet of the bank swallow in the territory in question, which probably is due largely to the fact that the bird feeds mainly on insects that are highly aerial. While the food of adults and young was much the same, in the case of small nestlings there was an excess of soft-bodied over hard-bodied forms. In its economic relations the bank swallow must be considered highly beneficial so far as man is concerned. It consumes a large number of two-winged flies, aphids, spittle-insects and their allies, together with plant-eating bugs and beetles as well as ants and other insects, and this is a service which doubtless outweighs in economic importance the bird's less desirable habit of feeding to some extent on certain parasitic and predacious hymenopterans, useful ground-beetles, dung-beetles and spiders. 212 Roosevelt Wild Life Annals DESTRUCTIVE AGENTS Man. — Since bank swallows generally fly higher than such forms as the barn and the cliff swallows, and also are less given to the habit of alighting on highways, mortality clue to speeding automobiles is apparently insignificant. In fact, I found but one such victim — an immature bird — in the Oneida Lake territory, during the course of the present field studies. From other human agency, also, the bank swallow as a rule suffers no serious harm in this region except, apparently in rare instances, from mischievous boys. For example, in late June, season of 1931, I found that some such individuals had destroyed nests and young in certain burrows, plugged up others with mud, thus sealing in young in the nests — and otherwise wrought havoc in a section of a populous colony at Fish Creek Landing. The Skunk. — In the Oneida Lake region the skunk (Mephitis nigra) is appar- ently the chief mammalian enemy of the bank swallow. Early in these studies I had heard from fishermen and farmers living in the vicinity of Fish Creek that skunks often dug into bank swallow burrows, particularly when young were in the nest, and devoured these. However, it was not until some time later, June 23 (1931), at Fish Creek Landing, that I personally found concrete evidence of the fact. Here, about three feet from the edge of the bank of Fish Creek, a skunk had dug through the thick turf and into the soil to a depth of a foot to reach a bank swal- low's nest. Mute evidence of its successful quest was found in the feathers torn from the young birds. On the same day, at the Graham farm, evidence was noted where a skunk had attempted, unsuccessfully, to gain entrance to a bank swallow burrow from the face of the bank, near its base. A few days later (June 29), at the Fox farm colony, were found the fresh excavations of a skunk that had dug out a bank swallow's nest situated twenty-two inches below the surface, in a clover field (Fig. 70). On July 2, at the Fox farm colony, within a linear distance of fifteen feet, five nests containing young birds were found to have been destroyed by the skunk within a period of two days. In this instance the animal had followed a plowed furrow at the margin of a corn field bordering the creek bank, and with the advantage thus afforded had been able to reach the nests with comparatively little digging. Fragments of egg shells, feathers of young birds and nest linings were scattered about. By July 11 several additional nests in this colony had been robbed in the same manner. Depredations of this sort were also noted in a number of other places. So far as observations extended it appeared that most of the depredations of the skunk were limited to burrows above the 20-inch level, although a few pillaged burrows were noted which were twenty- four inches below the surface. In some places along the creek where the skunk had been more energetic in its quests, its excavations had had their effect on the bank. During heavy rains the water, washing into the excavations and thereupon through the tunnels and to the face of the bank below, has a tendency to soften the earth and thus accelerate erosion and even cause a slipping of an otherwise more resistant part of the bank. 213 Fig. 67. Young bank swallows perched on high tension wires stretched above Fish Creek, near the colony at Fish Creek Landing. July 2, 1931, 12:20 P.M. Fig. 68. Banded juvenal bank swallow with a nearly mature Protocalliphora larva on the right side of the head in front of eye, and another (barely visible) on the left side of head above eye. The bird also carried many smaller maggots of the fly on its body. Note the unkempt and depressed appearance of the bird. Fish Creek at Edell farm. June 16, 1931. 214 Fig. 69. Bank swallow nest destroyed by skunk (Mephitis mephitis nigra) in the sheep pasture shown in Fig. 24. The animal dug into the burrow from above and pulled the nest to the surface ; remains of the young birds are intermingled with the nest materials. Fish Creek Landing. June 29, 1932. Fig. 70. Excavation in hay field bordering bank swallow colony along Fish Creek at Fox farm, made by a skunk to reach nest of young birds. Tune 29, 1931. Studies on the Bank Swallow 215 A slipping evidently due to this cause was noted at the Fox farm colony after the heavy rainfall of July 13 and 20, 1931. These predatory activities of the skunk can probably be attributed only to a comparatively small number of individuals living in the immediate vicinity of the bank swallow colonies. For this reason and because of the inaccessibility of most of the nests of these swallows the skunk can hardly be looked upon as a serious enemy of these birds. Slipping Banks. — The slipping of large sections of creek and sand banks is probably the chief source of destruction of bank swallows in the Oneida Lake region. While eggs and young are the principal victims, adult birds also are some- times included. Most burrows are usually sufficiently long to prevent destruction of the nest when only a moderate amount of slumping takes place. But a prolonged rainy period, accompanied by a rise in the stream, will, in certain places, so soften the earth that considerable sections of the banks give way. Accidents. — Young bank swallows when eight to ten days old, in their eager- ness to receive food from the parents, frequently crawl to the mouths of the burrows, and then individuals occasionally fall out. Although the parents will feed young of the season out of the nest when they are of flying age, I have never seen them bring food to a young one on the ground outside the burrow when it was unable to fly. Such helpless young, therefore, probably soon perish, — from hunger, exposure or predators. PARASITES AND DISEASE The principal external parasites found to infest bank swallows in the Oneida Lake region were mites and insects, and the several forms met with may here be discussed. Probably these parasites are seldom if ever directly responsible for the death of the bank swallow host, but their presence in large numbers might be the indirect cause of death in a weakened individual. Mites. — Minute grayish mites of the genus Atricholaelaps ( Acarina : Parasiti- dae) were not uncommon. Migratory nymphs of another but undetermined species of mite of the family Tyroglyphidae, were taken from young birds. While many of the young bank swallows in the nest in late June and in July harbored these forms (the numbers seemed to increase somewhat as the season advanced), yet, consider- ing the hundreds of swallows handled, the proportion of infested birds was not large. I have, in fact, seen young phoebes, robins and catbirds much more heavily infested with these parasites than were any young or adult bank swallows examined. Apparently the host suffers little discomfort. Fleas. — The flea, Ceratophyllus riparius R. and J. (Siphonaptera : Dolichopsyl- lidae), is a common occupant of burrows and nests which are or have been inhabited by bank swallows. In addition, it often is found on the bodies of both adults and voting. These insects appear to become progressively more abundant as the season 216 Roosevelt Wild Life Annals advances, and in some cases, at least, their presence must cause the host no little discomfort. Another flea, Ceratophyllus celsus Jord., also was taken from adult bank swal- lows captured in the Oneida Lake region. Early in May, before any of the old burrows which may have remained more i r less intact from the preceding season have been occupied by bank swallows, many ot the partially decayed nests are heavily infested with fleas, but numbers of them are found also in old burrows which contain no nests. And even before the season's domiciles are vacated by the swallows in midsummer, I have found hundreds of flea larvae of various sizes in the moist and decaying materials of a single nest. Evidently the fleas thrive throughout the year in these protected and well "pro- visioned" situations, transportation to new quarters being available during the seasonal occupancy of the host. Biting Lice. — Most of the adult bank swallows that I have handled harbored biting lice of the species Myrsidea dissimllis Kellogg (Mallophaga: Menoponidae). While these lice occur generally, they are, apparently, in my experience, never present in sufficient numbers to cause the host any serious annoyance. Both young and adult forms of the louse have been taken on the birds. Well-fledged young, but not yet able to fly, are likely to harbor numbers of these active, flat, brownish insects about the bases of the tail- and wing-feathers. Adult birds and young which have left the nest are not so heavily infested, probably because they are not con- tinuously exposed to infestation in the nest and are better able also to dislodge the pests. These parasites appear to attain their maximum abundance on nestling bank swallows after mid-July. Protocalliphora Larvae. — Without question the most injurious external para- site of the bank swallow in the Oneida Lake region is the larva of the two-winged muscoid fly Protocalliphora splcndida Macq. Although this blood-sucking larva appears to confine its activities largely or entirely to young birds in the nest, because of its abundance and wide occurrence, particularly in late June and throughout July, this parasite is one of considerable importance in the domestic economy of the bank swallow. The occurrence of the larva appears to vary a great deal and to be dependent on local, seasonal and possibly other conditions. Of the hundreds of young bank swallows examined in the Lake Okoboji, Iowa, region, during five seasons of work, less than a half-dozen families were found to be infested. How- ever, among the several hundred families of nestling and fledgling birds examined in the Oneida Lake region, in the season of 1931, scores of individuals carried the larvae in various stages of development. Some families in certain colonies are evidently likely to be more heavily infested than others. Counts were made of the larval Protocalliphora splcndida infesting a considerable number of the young birds, and their number and distribution on representative individuals are presented in Table Xo. 51. Studies on the Bank Swallow 217 Table No. 51. Numbers and Distribution of Protocalliphora Larvae on Nine Families of Young Bank Swallows. Oneida Lake Region. 1931. DATE Designation of bird Age of bird Number larvae on bird Distribution of larvae Remarks June 1 6 56616 B Fledgling. Fledgling. Fledgling Fledgling. 0 5 9 5 This bird recovered as a " return " on June 10, 1932 All four birds successfully reared; left nest in good condition 56617 B 56619 B 2 on head, 2 on nape and 1 in right nostril Wing, foot, head and bellv Various parts of body 22 46566 F 46567 F Fledgling. Fledgling . Fledgling. 13 0 4 All parts of body Vary in size from very small to almost mature. Feathers matted with blood. One larva had its mouth parts inserted into the body of an- other larva on the same bird. Found 34 addi- tional and nearly mature larvae and 6 puparia of Protocalliphora in nest with these young 46568 F 30 55600 F Fledgling. Fledgling. Fledgling. 1 8 2 Members of this family exhibit no evident ill effects from presence of larvae 55601 F 55602 F 5 on rump, 2 on wing, 1 on back July 2 55633 F Juvenal. . Juvenal. . Juvenal . . Juvenal . . 3 4 4 2 Birds in this family exhibit no apparent ill effects from the presence of larvae 55634 F 55635 F 55636 " 2 of the larvae in same nostril 1 larva in nostril 55685 F O.T08 / r Juvenal . . Juvenal . . T Juvenal . . 14 26 28 General; mostly head, rump and wings General; mostly head, rump ;md oeneral; mostly head, rumpand wings Larvae from 2 mm. to nearly mature. All birds in this family exhibited evidence of parasite's work in their under-nourished and emaciated appearance, their sub-normal activity and alertness and in the bloody and bedraggled appearance of the feathers 3 55688 F Juvenal . . Juvenal. . Juvenal. . Juvenal . . 34 Infested but not counted Infested but not counted 74 General Effects of presence of larvae plainly visible on all the birds. Larvae ranging in size from 1 mm. in length to two-thirds grown. Aggregated on rump and wings 55689 F General 55690 F General Stomach No. 52 General 9 55806 F Fledgling. Fledgling. 14 10 General Both birds also heavily infested with lice and mites 55807 F General 9 55839 F 55840 F 55841 F Stomach No. 59 Juvenal. . Juvenal . . Juvenal . . Juvenal . . Infested but not counted Infested but not counted Infested but not counted 77 General, espe- cially wings and rump General, espe- cially wings and rump General, espe- cially wings and rump General, espe- cially wings and rump All birds in this family heavily infested. Larvae well distributed on body; particularly prev- alent in auricular openings and on rump and wings. Bases of tail feathers matted with excrement of larvae and with blood oozing from wounds made by their mouth parts 22 55907 F 55908 F Stomach No. 69 Fledgling. Fledgling. Fledgling. 4 2 3 3 small larvae in right auricular opening, 1 larva in left 1 on each foot. . It seems likely that the sensitivity of young hank swallows to any irritation as, for example, that set up by the presence of some numbers of Protocalliphora larvae, is not so great as in older birds. 218 Roosevelt Wild Life Annals Those larvae may become attached to almost any part of the bank swallow host's body. The soles of the feet and the skin between the bases of the toes seem to be particularly vulnerable, but the head, neck, abdomen, rump and wings also are attacked. Their hold is maintained principally through the action of a suctorial disc. They commonly enter the nasal or ear openings where the skin is delicate and the blood supply generous. I have, for example, taken three larvae from one of the external ear openings of one bird ; and two individuals gorged with blood, from the external nasal passage of another fledgling. Larvae 3 to 5 or more millimeters in length, which are almost completely hidden within the nostril, are common, and a larva in each nostril of a nestling is not rare. Occasionally a nearly mature larva projects from one of the nasal openings, giving the appearance of a wart-like or horn-like protuberance (Fig. 68). Birds with the nostrils completely or even partially plugged by these maggots must be com- pelled to breathe wholly or partly through the mouth ; apparently, however, they experience no serious handicap. On one occasion I found a Protocalliphora larva with its mouth parts inserted into the body of another larva of the same kind — doubtless an accidental occurrence. Unlike the larvae of Protocalliphora azurea Fallen which infest the green-backed goldfinch (Astragalinus psaltria hesperipholis) and which, according to Plath (1919, p. 33), "do their deadly work at night and rest in the lower parts of the nest during the day," the representative of this genus which infests bank swallows is both a nocturnal and diurnal feeder. Even when the birds were handled and examined in our investigations few of the maggots became detached. Usually con- siderable force was necessary to disengage the mouth parts and suctorial disc of the larva from the body of the host. — sufficient to cause the bird some pain. So far as I have been able to determine, the adult birds make no effort to remove the Protocalliphora larvae from their offspring ; nor is there any evidence to indicate that the young birds attempt to remove the parasites from their own bodies. Protocalliphora maggots also are attracted by the rich blood supply found about the bases of the rapidly developing tail feathers and the primaries and secondaries of the wings of young birds. The areas about vent and oil glands, too, are frequent points of attachment. As a result of massing of the larvae the plumage of these parts often becomes matted with the excrement of the maggots and with blood of the host. The larvae work down into the feather follicle. Small larvae are often so well concealed within the follicle that their detection may be difficult. The late Charles W. Johnson of the Boston Society of Natural History, who has made extensive studies on these larvae, particularly in relation to bluebirds and tree swallows, in regard to the seasonal history of this insect, says (1927, p. 81): "The females apparently hibernate, probably in hollow trees and other sheltered places, and in the spring deposit their eggs in the nests of birds. As many of the birds are double-brooded, there are also two generations of flies. That the larvae of the flies are more abundant in the nests of the second broods than in the first is shown by comparing the bluebird nests collected in May with those taken in July * * *". Studies on the Bank Swallow 219 My own findings, that second brood and late first brood nestlings of the bank swallow are more heavily infested than early-season young, parallel Mr. Johnson's observations in the case of the bluebird. By early July, adults of the first genera- tion of the fly have issued, and egg-laying takes place almost immediately. At this time of the year the adult fly usually is common about the mouths of bank swallow burrows, for its population has had an opportunity to build up through the emergence of a new generation. Therefore, a heavier infestation of bank swallow nestlings in July by second generation larvae appears probable. This was apparently substantiated by our finding more generally and heavily infested nest- lings in late summer. On July 8 (1931) note was made of the fact that adults of Protocalliphora were then among the commonest insects in and about the bank swallow burrows, while the larvae were not so abundant. Apparently one generation of the fly had developed while a second generation was now well begun. A week later every nestling examined was found to be infested with Protocalliphora larvae, from 2 or 3 on some individuals to as many as 40 or 50 on others. The period of the larval life of the fly apparently is nicely adjusted to the nestling period of the bank swallow. Protocalliphora larvae appear to be well distributed but not excessively abundant on any individual bank swallow up to late June. From early July on to the close of the season the number of available young bank swallow hosts in the nest gradually decreases (second broods are relatively few), and therefore the proportion of infested birds is likely to be higher at this season. Empty puparia are frequently found in numbers in nests and debris of aban- doned bank swallow burrows. It is possible that small hymenopterous or dipterous parasites of the Proto- calliphora larvae themselves help to hold their numbers in check, but no evidence on this point was noted in the present study. Johnson (1927), however, found that certain Chalcididae and Tachinidae aided in suppressing the numbers of Protocalliphora infesting bluebirds and tree swallows. The same author (loc. tit., p. 80) queried whether nests in such sheltered and dry places as bird houses are more heavily infested by Protocalliphora larvae than those nests constructed in hollow trees and limbs. The query carries with it the implication that possibly the damper surroundings of the latter type of nesting place might offer greater possibilities for predatory insects and parasites which would aid in suppressing Protocalliphora. While I can not speak regarding the nests in "hollow trees and limbs," it is my observation that the usually more or less damp conditions existing in the nesting tunnels of bank swallows do not favor the development of any large population of possible parasitic enemies or of insects that prey on Protocalliphora. Possibly, also, the darkness of the burrows and their relative inaccessibility may be responsible, in part. Foot Disease. — In the more than 1200 young and adult bank swallows that were handled and examined in the Oneida Lake region, in the present study, not more than ten examples in all have exhibited diseased external parts, and in only four of these — all young birds — had the disease attained serious proportions. 220 Roosevelt Wild Life Annals Four juvenal hank swallows taken in June and July exhibited abnormal and dis- eased feet or tarsi or both. In two eases both legs were affected. The disease manifests itself in a swollen or turgid condition of the affected parts, with a more or less twisted, knotted or otherwise deformed appearance in advanced stages. Until the disease has attained some headway the skin covering the affected parts remains unbroken. Usually the color of the parts loses the normal healthy pinkish or purplish tinge and becomes pale grayish or yellowish, the unhealthy appearance sometimes extending to parts of the leg at some distance from the tumid or bleeding area. In those cases where the disease obviously had existed for some time and extended to the toes of the bird, all or portions of these appendages had sloughed away. The ungual phalanges seem to be the parts most likely to be affected. Since the observed cases of this disease have all been in young birds it seems likely that, directly or indirectly, it results in the death of the subject within a compara- tively short time. Its rapid development was indicated by the fact that in two individuals, not more than 15 to 17 days old, the swollen, deformed condition of the toes was already well advanced. In view of the occurrence of the larvae of Protocalliphora on some of the birds afflicted with this disease, it is perhaps possible that there is a connection between the two. Data on the four specimens exhibiting foot disease are as follows : No. 37; fledgling; one of a family of four; June 24, 1931; Edell farm colony. Right leg slightly swollen ; inner toe sloughed off to basal phalanx. Tarsus of left leg shortened ; basal segments of all toes enlarged, brownish in color, the three anterior toes much reduced in size but claws still present ; hind toe normal. No. 48 ; immature male able to fly ; July 2, 1931 ; Fish Creek Landing. Right tarsus and foot malformed, gangrenous and much swollen ; claws and digits still intact. No. 57; immature male able to fly; July 9, 1931; Edell farm colony. Right foot normal ; left foot swollen at base of toes ; the hind toe normal ; the lateral front toes reduced in size ; the middle toe discolored and attached only by shreds of skin, its terminal segment swollen and bloody. In other respects the bird appeared in good condition. No. 67; juvenal male not able to fly; the only young bird in this nest; July 16, 1931 ; Edell farm colony. Both tarsi swollen; legs and feet yellowish in color; toes all present, the middle toe of right foot swollen, but without external lesions. The bird heavily infested with Protocalliphora larvae. The last individual also exhibited a dry, scurfy, condition at the bases of the occipital feathers which were loosely attached in their follicles. Whether this condition had any connection with the foot disease or was simply an indication of the general unhealthy condition of the subject is uncertain. Regarding this foot disease all that can here be said is that, while it occurs occasionally among bank swallows it appears not to be of serious consequence, and its true nature and cause are unknown. Studies on the Bank Stealing' 221 MISCELLANEOUS In addition to the various forms noted above, a miscellaneous group of insects and other animals may be mentioned, which, although found about the abodes of the bank swallows, probably have no special significance. Adult two-winged flies (Diptera) of several species are often observed flying in and about the entrances of the bank swallow burrow s. Their numbers in these situations increase as the season advances. They are likely to congregate especially in and about burrows in which birds may have died, or masses of excrement accumulated. Among these insects are the house fly (Musca domestica), the blue-bottle fly (Calliphora vomitoria) , various members of the family Sarcophagidae, and probably others. Larvae of Sarcophaga securifera Vill. were collected on June 15, 1931, from the decaying bodies of young bank swallows found in a burrow. It seems probable also that old bank swallow nests and burrows serve as hibernacula for various flies and other insects throughout the winter. Among the Hymenoptera, ants (Formicidae) are often found in old abandoned nests, as well as sometimes in occupied nests and burrows. Apparently two or three species are concerned here, but the form Lasiurus niger americanus Em. is by far the commonest. Ants' nests containing larvae or pupae were frequently unearthed when we enlarged bank swallow tunnels. Ants are probably beneficial in their relation to the bank swallow in so far as they act as scavengers in the nests and burrows. Other hymenopterans which have been observed flying or crawling within bank swallow burrows are bumblebees (Bombidae), mining bees of the family Halictidae, and ichneumon-flies (Ichneumonidae). A small greenish halictid. Halictus (Chloralictiis) sephyrus Smith, is very common and sometimes digs its tunnels in the floor of bank swallow burrows. In places the face of the bank in the vicinity of the burrows may show great numbers of the insect tunnels, appearing as though a charge of shot had been driven into it. Of Coleoptera, several species of ground beetles (Carabidae) and representa- tives of the hister beetles (Histeridae), the short-winged scavenger beetles (Staphylinidae ) and the carrion beetles (Silphidae) have been taken from occupied or recently vacated bank swallow burrows. One of the carabids taken from a decayed nest in an abandoned burrow at the Daniels farm, May 7, 1931, was determined by L. L. Buchanan as Acupalpus earns Lec. Various kinds of beetles (Coleoptera), representatives of the leaf-hoppers and allies (Homoptera), plant bugs (Heteroptera), caterpillars (Lepidoptera) and sometimes other insects, not infrequently are found in and about a bank swallow burrow. These insects evidently come from the overhanging vegetation and their occurrence in the burrow is purely fortuitous. The predacious ground beetles as well as the carrion beetles probably are attracted by the decaying contents of the nests. On June 15, 1931, at the Fox farm a certain burrow contained the decaying bodies of three young bank swallows on which the following insects were feeding: One carrion beetle, Necrophorus sayi Lap. (Silphidae) ; one short-winged scaven- 222 Roosevelt Wild Life Annals ger beetle, Aleochara mediates Csy. (Staphylinidae) ; several ants (Formicidae) and several dipterous larvae. At the Edell farm on July 9, 1931, a similar aggregation of carrion-feeding insects was noted in a burrow, feeding on the decayed remains of a young bank swallow. Among these was an example of Necrophorns tomentosus Web. While handling bank swallows. I have, on several occasions, found a short- winged scavenger beetle. Tacky porus chrysomelinus Linn. (Staphylinidae). On one such occasion one of these beetles leaped out from among the body feathers of the bird. Other examples have been noted among the feathers of the head and various other parts of the body, both of young and adults. These insects are very active, and difficult to capture. Doubtless they are to be considered as serving in a useful capacity as scavengers and for this reason should not be included in the same category with the preceding insects. A small spider determined by C. R. Crosby as Walckcnaera vigilax Blackwall, was taken, July 16, 1931, from an abandoned bank swallow nest at the Edell farm colony. This is the only example of spider that was found in such a situation. On July 14, 1931, at a large unworked sand pit near Humaston, I found a deer mouse. Peromyscus leucopus noveboracensis (Fischer), in a bank swallow burrow containing a swallow's nest with a deep feather lining. There was no visible evidence that the mouse had eaten young birds or eggs. For the purpose of comparison it may be of interest to mention the following forms recorded (Stoner, 1926 and 1928) from the burrows of bank swallows in Iowa: A camel cricket (Ceuthophilus testaceus Scud.) ; a ground beetle (Harpalns pennsylvanicus Dej.) ; a sarcophagid fly (Sarcophaga bidlata Park) ; an anthomyid fly (Helina sp.) ; the parasitic hymenopteron, Xyalosoma sp., also was reared from sarcophagid puparia collected from bank swallow burrows. Among higher animals, a garter snake (Thamnophis sp.), a northern flicker (Colaptes auratus Intend Bangs) and a prairie harvest mouse (Reithrodontomys megalotis dychei Allen) have been taken in abandoned bank swallow burrows. BANDING STUDIES Since one of the objectives in this study depended upon later identification of the same individuals, a considerable number of bank swallows were banded, although the mere banding of large numbers of the swallows was not in itself a major objective. Table No. 52 presents a summary of these banding efforts. St ml ics on the Bank Swallow 223 Table No. 52. Summarized Record of Bank Swallows Banded in the Oneida Lake Region. 1928-32. FIELD STATION' Age of bird 1928 1 (l )() 1931 1932 Totals j une July July May June July June July Ad. * ng. 9 2 11 Ad. Von i ng. 1 24 1 2 Daniels farm Ad. i ng. 5 6 14 n 1 4 Ad. i ng. 3 g 3 2 Edell farm Ad. x ng. 7 26 236 23 15 56 318 64 3 Ad. x ng. 25 41 49 4 48 49 9 119 109 3 Ad. i ng. 18 43 175 S 105 20 86 282 2 Ad. Yng. 33 20 117 1 8 25 2 79 131 4 Ad. Yng. 6 14 29 1 21 38 9 Ad. Yng. 2 2 Onderdonk farm Ad. Yng. 14 14 Others Ad. Yng. 2 21 2 26 S 23 5 2 103 781 285 147 12 1.358 Method of Capture. — Capturing adult bank swallows presents certain difficul- ties. Obviously the birds are not to be taken successfully by the use of ordinary traps. Capturing them in their burrows during the breeding season seemed to be the only practicable method in this study. Although it might seem a comparatively easy matter to secure the birds when once within their burrows, we found this often to be far from the case. A bank- swallow will not always emerge promptly into the waiting net, and sometimes no amount of beating on the top or the face of the bank, or exploration of the burrow with a long wire, will dislodge the occupant. Nor does any reasonable degree of watchful waiting always avail. Individuals that have been captured previously, those with eggs near hatching time, or with small young, are most reluctant to leave the burrow. One method that proved to be fairly successful in dislodging these reluctant swallows was that of directing, by means of a small hand mirror, a beam of light into the burrow (Fig. 40). The occupant then at once shows uneasiness and usually faces the source of light, shaking its head vigorously. Often the bird then, seemingly unable to endure the beam, crawls or shuffles to the mouth of the burrow and launches itself into the net. 224 Roosevelt Wild Life Annals Occasionally the bird attempts to avoid the light by crouching behind the nest, between it and the burrow wall, and refuses to move. Then it becomes necessary to dig the bird out — if it is to be captured. Digging, however, has the disadvantage not only of mutilating the burrow but of causing the swallows to desert it, unless the eggs are near hatching time, or the nest contains young. In addition to its other advantages, the beam of light, as previously remarked, permits a general examination of the nest and its occupants. It has also disadvantages in that it can not be used on cloudy days and in certain situations. The use of an electric flashlight, even of the focusing type, was found to be much less satisfactory'. But for removal of nestling and fledgling birds from the nest it is necessary to enlarge the burrow so that the arm can be introduced (Fig. 53). In case its end is beyond arm's length, the burrow may be shortened by cutting away the face of the bank immediately around its mouth. After having finished with a burrow containing young, it was our practice always to restore the entrance to its original size by means of mud near at hand. Behavior of Captured Birds. — Nestling bank swallows and older young, before the flying age, offer little or no resistance to handling. Most adult individuals, as well as the immature of flying age, void their excre- ment immediately upon emerging from the burrow into the net. Adults also often struggle a little at first, but after a few seconds usually become quiet when held in the hand. From all outward appearances most of these individuals are perfectly submissive and unalarmed ; nevertheless, they are alert and ready to fly on the instant if an opportunity offers. Usually neither adults nor young offer any vocal remonstrances upon capture. An occasional adult individual will utter a few harsh cries of terror, apparently, when first taken ; and so likewise do occasional immature birds of flying age, and these individuals, characteristically, will also struggle violently in efforts to escape. These outcries usually result in an immediate evacuation of the adjacent burrows by all tenants able to fly, thus ruining one's hopes for a good catch. As before remarked, some individuals while held in the hand utter a low, barely audible "song," which may be continued for some time. In my experience neither the adults nor young attempt to peck or pinch with the bill. Recoveries of Banded Birds. — While more than sixty repeat records are available, only a few of them furnish anything of interest, and these may now be mentioned. 1. Young Repeats. — No. 56678B was one of a family of four fledglings in a burrow on the south side of Fish Creek, Graham farm, banded on June 11. 1931. On Tune 19 it was not yet capable of extended flight, and the only one of the four remaining in the burrow. If the other three young of this family had left the burrow of their own accord the adolescent period could not have been more than 16 or 17 days. On June 26 No. 56678B was recovered from a burrow situated six inches below its original home. It was now able to fly and was alone in this burrow. Studies on I he Bank Swallow 225 No. 46438F was one of a family of four nestlings in a burrow on the south side of Fish Creek, Edell farm, handed on June 16. Recovered June 30, in the same colony, hut from a burrow twenty-eight inches removed from its original one. This bird was also alone, and now well able to fly. No. 46443F was one of a family of five nestlings banded at the Edell farm, June 16. Recovered on June 30, in the same colony, but from a burrow eighteen inches removed from its home burrow. Nos. 46649F to 46652F banded as nestlings from a burrow at the Edell farm. June 24. All four were found together in the same colony, but in a different burrow, on July 9. In this burrow at the same time were also two other immature repeats, and two unhanded immature swallows of the year. There were, therefore, representatives of at least three different families of young in this burrow. Nos. 46699F and 46700F were two of a family of four nestlings banded at the Edell farm, June 24. They were recovered, July 9, from a burrow four feet removed from their home burrow. In this instance, therefore, a part of the same family was found togther after having left the home burrow. Other such instances will be noted below. No. 46747F was one of a family of three nestlings banded from a burrow at the Fox farm. June 25. Recovered on July 7 from a different burrow in the same colony. No other birds in this or in the parental burrow on date of recovery. Nos. 46749F and 4675 IF were of a family of four juvenals banded at the Fox farm, June 25. Recovered together in a different burrow in the same colony, July 7. Nos. 55555F and 55558F were of a family of five juvenals banded at the Edell farm, June 30. The first one was recovered in the same colony but from a different burrow at 2:05 P. M., July '9. In the burrow with it were five banded birds from two different families, and two other unhanded birds of the season. In this burrow, which was not the original habitation of any of its banded occupants, were, therefore, immature representatives of at least four different families of bank swallows. No. 55555F was again recovered 50 minutes later from a burrow seventy-five yards distant from where it was found earlier, and twenty feet from its home burrow. Another immature individual was in the burrow with it, but evaded capture. No. 55558F was recovered at 3 :00 on the same afternoon, from its home burrow where it was found alone. Nos. 55566F and 55569F were of a family of five juvenals banded at the Edell farm, June 30. Recovered together from a burrow three feet distant from the original one, July 9. No other members of the family present. No. 55729F, one of two immatures able to fly, banded from a burrow at the Fox farm, July 7. Recovered from a burrow three-fourths of a mile distant, on the Edell farm, July 9, in company with three unhanded immature birds. No. 55882F, banded as a fledgling from a burrow at the Fox farm, July 14, was found July 21, on the creek bank in front of its burrow. It was still unable to fly. This helpless individual was carried to a colony 100 yards distant and placed in a nest containing three nestlings considerably smaller than itself. This was the only nest in the entire colony now found to contain young birds. A visit to this 226 Roosevelt Wild Life Annals burrow two days later showed all the young, including the "foster child," to be present and evidently well cared for. 2. Adult Repeats— No. 4643311 was banded at Bryant's sand pit, May 18. The bird was in a burrow containing a nest but as yet no eggs. On June 15 this ^wallow was found incubating at Grems' sand pit, in a burrow from which, on June 8, adult No. 56596B was captured and banded. This sand pit is a little more than a mile from Bryant's sand pit. Possibly as a result of our intrusion No. 46433B had abandoned its original nest and colony ; and while the evidence is only circum- stantial, it seems that this individual may have changed mates during the twenty- seven days between its first capture and the subsequent recovery. No. 46465B was banded at the Graham farm colony, May 21. Another individual which had flown into the burrow at the same time became No. 46466B. There was a nest with three eggs. On May 26, No. 46465B was recovered from a burrow at Grems' sand pit, together with an unhanded adult which now received band No. 46483B. This burrow was twenty-one inches deep and had been dug within the last six days. Immediately prior to capture both birds had flown into the burrow and were engaged in digging. The place of recovery is four miles from where the first of these two birds was banded. This recovery indicates that within a period of five days No. 46465B had changed not only its nesting site, but also apparently its mate. Polygamy or polyandry may be suggested. The record is of further interest in that it illustrates movement of this swallow after the nesting season was well under way, for the individuals concerned on May 21 evidently were paired, a clutch or a partial clutch of eggs had been laid, and incubation had at least begun. No. 56548B. Banded as an incubating bird from a burrow at the Fox farm colony, June 2, at 4:30 P. M. This bird, together with No. 56549B which was captured as an incubating bird from a different burrow in the same colony, at 5 :00 P. M., was placed in a small gauze sack within a canvas carrying bag and taken to the outskirts of the village of Cleveland, eleven miles, air-line, northwest of the point of capture, where both were released at 5 :35 P. M. At once both birds set out in a southeasterly direction. On our next visit to the colony, June 5, No. 56548B was not in the same burrow from which it was taken three days before ; but an unhanded individual was incubating the eggs. Assuming that these two individuals were male and female, it now seemed evident that both sexes took part in incubation. Our next visit was on June 10, when we found No. 56548B back on the nest. With regard to No. 56549B, it may be said that this individual, too, was recovered from the original burrow at 1 :00 P. M. on June 15, when it was brooding very small young. No. 56566B was banded as an incubating bird from a burrow at Fish Creek Landing, June 3, 4:20 P. M. This bird and three other incubating individuals captured and banded at this colony on the same afternoon, were placed in a gauze net and taken by automobile to the outskirts of the village of Cleveland, where all were released at 5 :30 P. M. The point of release is about nine miles northwest of Fish Creek Landing. Immediatelv on being freed all four birds rose high in the air and without the Studies on the Bank Swallow 227 slightest hesitation set out in a southeasterly direction, that is, in the direction of their nesting locality. At 2:30 P. M., June 5, No. 56566B was recovered from a hurrow six inches removed from the one in which it was originally captured, and in which another individual, No. 56567B, was taken and handed on June 3. None of the remaining three individuals released at Cleveland was suhsequently recovered. No. 56620B was banded as an incubating bird from a burrow at the Edell farm colony, June 9, at 4:50 P. M. Recovered from the same burrow, June 16, at 4 :45 P. M. This individual and two other incubating birds captured from nearby burrows in the same colony were likewise placed in a gauze sack and taken by automobile to the village of Cleveland, eleven miles northwest of the home colony, and there freed at 6:10 P. M. After liberation the birds kept more or less together, circling about in an uncertain manner. In a few moments they were joined by a fourth bank swallow, and all then set out in a northeasterly direction away from the home colony. None of these birds was subsequently recovered. On June 24, examination of the burrow in which No. 56620B had been captured indicated that the nest and its two eggs had been abandoned. 3. Returns. — From the beginning of our work in the Oneida Lake region in the season of 1928 to the close of the 1932 season, a total of 1358 bank swallows was banded. Of these, 14 were found dead in the 1931 season. In the 1932 season it was possible to spend only five and a half days in the same territory, and during that time ten live returns were secured from three of the seven 1931 banding stations visited. Since some interest is associated with the history of any bird that has made a round-trip flight from its nesting grounds to its winter home in South America, data on the bank swallow returns obtained in the 1932 season may be briefly mentioned (Temperature records for these individuals appear in Table No. 40). No. 46442B. Banded as an incubating bird May 19, 1931, at 11:35 A. M., from the colony on the south bank of Fish Creek at Fish Creek Landing. Recovered as an incubating bird on June 2, 1932, at 12 :32 P. M. ; same colony site where origi- nally captured, but in a burrow about 600 feet distant from the previous one. No. 46479B. Banded as an incubating bird May 22, 1931, at 4:15 P. M., from the Fox farm colony. Recovered as an incubating bird at the Graham farm colony on June 3, 1932, at 1 :50 P. M. The point of recovery is two miles north- east of the place of banding. No. 56502B. Banded as an incubating bird May 27, 1931, 11:15 A. M., at Fish Creek Landing. Recovered as an incubating bird at the Fox farm, about two miles northeast of the point of original banding 2:45 P. M., June 2, 1932. No. 56503B. Banded as an incubating bird May 27, 1931, at 11:45 A. M., from the colony on the south side of Fish Creek at Fish Creek Landing. The bird was still incubating when it was recovered from the same burrow as a repeat on June 5, following, at 2:40 P. M. Recovered as a return June 2, 1932. at 12:05 P. M.; incubating, in the same colony, but a few yards removed from its previous season's burrow. A female ; the outline of a large egg about ready for deposition could be seen through the abdominal wall. 228 Roosevelt Wild Life Annals No. 56556B. Banded as an incubating bird June 3, 1931, at 1:55 P.M., from a burrow on the south side of Fish Creek at Fish Creek Landing. Recovered as an incubating bird (4 eggs) a few yards from the point of original capture, June 2, 1932. at 11 :3S A. M. No. 56572B. Banded as an incubating bird at south end of main colony at the Graham farm, June 4. 1931, at 1 :50 P. M. Recovered as an incubating bird June 3, 1932. at 12:10 P. M., in same colony where banded, but in a burrow about 100 yards north of the original one. No. 56576B. Banded as an incubating bird at south end of main colony at the Graham farm, June 4, 1931, at 2:20 P. M. Recovered from a burrow near north end of same colony, about 100 yards distant. June 3, 1932, at 1 :40 P. M. No. 56594B. Banded as an incubating bird June 8, 1931, at 3:10 P. M., from a burrow on the south side of Grems' sand pit. Recovered as an incubating bird from a burrow on the south bank of Fish Creek at Fish Creek Landing, June 10, 1932, at 10:05 A. M. The point of recovery about three and one-half miles south of the place of banding. This individual was nesting in a sand pit, on the first occasion, and in a creek bank on the second. A second individual from the same burrow, and evidently its mate, was banded on June 2, 1931, as No. 56533, but was not later recovered. No. 56617B. Banded as a nestling 5 to 7 days old, in a family of four; from a burrow at the Edell farm colony, June 9, 1931, at 4:40 P. M. Recovered as an incubating bird from a burrow on the south side of Fish Creek at Fish Creek Landing, June 10, 1932, at 10:50 A. M., two and a half miles from where it was hatched. No. 55784F. Banded as an immature, able to fly, of a family of three, from a burrow on south side of Fish Creek at Fish Creek Landing, July 8, 1931, 4:00 P. M. Recovered as a brooding individual from a burrow in same colony, June 4, 1932, 1 :4S P. M. Remarks. — Of the total number of bank swallows banded by us in the Oneida Lake region, 1199 individuals were presumed to be available for returns in 1932. Ten of these were recovered in the five and one-half days spent afield that season. Although more extended field work is needed before any safe conclusions can be drawn in regard to this question, it would seem that the number of recoveries obtained as compared with the number of birds available as returns, indicates that few individuals return to breed in the exact locality in which they were reared. A somewhat larger number apparently returns to the general territory. The avail- able evidence, though inadequate, further indicates that adult birds are more likely to return, season after season, to the same locality to breed, than are first-year individuals to the place of their birth. This is in accordance with our findings in the Lake Okoboji region (Stoner, 1928 and 1928a). In apparent substantiation of this statement is the fact that, so far as known, only 266 adults were available for returns in 1932, and eight of these were recovered as such. At the same time 919 birds banded as young of the season in 1928, 1929 and 1931, were available for returns in 1932, but only two of these were recovered as such. Studies on the Bank Swallow 229 In no case were recovered birds found to be nesting in the burrows they had been known to occupy in a previous season, but it is entirely possible that some individuals may do SO. Within certain ill-defined limitations, selection of a nesting place appears to be largely a matter of chance and is not governed by any dominant attraction to the birthplace. The data obtained here and in northwestern Iowa indicate that the bank swallow breeds in its first year. In no case has a return bank swallow been found to be mated with the same individual as in the preceding season. All evidence points to the conclusion, also, that few if any of these birds retain the same mate for the second brood — when there is one — of a given season. And through some disturbance or interference a change of both burrow and mate may take place before the laying of the first clutch of eggs has been completed. 2M) Roosevelt Wild Life Annals BIBLIOGRAPHY American Ornithologists' Union 1931. Check-list of North American Birds. Fourth Edition. Pp. 1-526. Lancaster, Pa. Audubon, John James 1840. The Birds of America. 1840-44. Austin, O. L. Jr., and Low, S. H. 1932. Notes on the Breeding of the Tree Swallow. Bird Banding. Vol. 3, No. 2, pp. 39-44. Baird, S. F., Brewer, I. M., and Ridgway, R. 1905. A History of North American Birds. Vol. 1, pp. 1-596. Boston. Baldwin, S. Prentiss, and Kendeigh, S. Charles 1932. Physiology of the Temperature of Birds. Scientific Publications of the Cleveland Museum Nat. Hist., Vol. 3, pp. 1-196. Beal, F. E. L. 1897. Some Common Birds in their Relation to Agriculture. U. S. Dept. Agric, Farmers' Bull. 54. pp. 1-40. 1918. Food Habits of the Swallows. U. S. Dept. Agric, Biol. Surv. Professional Paper 619, pp. 1-28. Bergtold, W. H. 1917. A Study of the Incubation Periods of Birds. Pp. 1-109. Denver. 1926. Avian Gonads and Migration. The Condor, Vol. 28 (May), pp. 114-120. Birge, E. A., and Juday, Chancey 1920. A Limnological Reconnaissance of West Okoboji. Univ. Ia. Studies in Nat. Hist., Vol. 9, No. 1, pp. 1-56. Blatchley, W. S. 1910. An Illustrated Descriptive Catalogue of the Coleoptera or Beetles (exclusive of the Rhynchophora) Known to Occur in Indiana. Bull. 7, Ind. Dept. Geol. and N. R. Pp. 1-1386. Indianapolis. Brewster, William 1898. Revival of the Sexual Passion in Birds in Autumn. The Auk, Vol. 15, No. 2, pp. 194-195. Burns, Frank L. 1915. Comparative Periods of Deposition and Incubation of Some North American Birds. Wilson Bull.. Vol. 27, No. 1, pp. 275-286. 1924. The Philosophy of Birds' Nests and Comparative Caliologv in Consideration of Some Local Nidicolous Birds. Wilson Bull., Vol. 36, No. 2, pp. 78-88; No. 3, pp. 121-127; and No. 4, pp. 188-200. Cahn, Alvin R. 1920. Bird Notes from Itasca County, Minnesota. Wilson Bull., Vol. 32, No. 4, pp. 103-122. Chamberlain, K. F. 1928. The Insect Food of the Dusky Skunk. New York State Museum Handbook 4, pp. 93-100. Chapman, Frank M. 1932. Handbook of Birds of Eastern North America. Pp. 1-581, Second Revised Edition, D. Appleton and Co. Coutant, Alrert F. 1915. The Habits, Life History and Structure of a Blood-sucking Muscid Larva (Proto- calliphora acurca) . Jour. Parasitology, Vol. 1, No. 3, pp. 135-150. DeKay, J. E. 1843. Natural History of New York. Zoology, Pt. 2, Birds. Pp. 1-380. Albany. Eaton, Elon H. 1914. Birds of New York. New York State Museum Memoir 12, Part 2, General Chapters: Land Birds. Pp. 1-719. Albany. Studies on the Bank Swallow 2M Ewing, Henry Ellsworth 1929. A Manual of External Parasites. Pp. 1-225. Charles C. Thomas, Baltimore. Forbush, Edward Howe 1907. Useful Birds and Their Protection. Mass. State Board Agric. Pp. 1-437. 1916. The Natural Enemies of Birds. Mass. State Board Agric, Economic Biology. Bull. 3, pp. 1-58. 1927. Birds of Massachusetts and Other New England States. Part 2. Land Birds from Bob-whites to Crackles. Mass. State Board Agric, pp. 1—461. Gardner, Lkon L. 1930. On the Bodv Temperatures of Nestling Altricial Birds. The Auk, Vol. 47, No. 3, pp. 367-379. Henshaw, H. W. 1908. A Parasitic Fly Injurious to Our Native Birds. The Auk, Vol. 25, No. 1, pp. 87-88. Herrick, Francis H. 1900. Care of the Nest and Young. The Auk, Vol. 17, No. 2, pp. 100-103. Honywill, Albert W., Jr. 1914. Additions to "Notes on Some Summer and Fall Birds of the Crooked Lake Region, Cass and Crow Wing Counties, Minn." The Auk, Vol. 31, pp. 82-86. Hopkins, Thomas C. 1914. The Geology of the Syracuse Quadrangle. N. Y. State Museum Bull. 171. The University of the State of N. Y. Pp. 1-80. Albany. Hough, Garry DeN. 1899. Some North American Genera of the Dipterous Group, Calliphorinae Girschner. Ent. News, Vol. 10, No. 3, pp. 62-66. House, Homer D. 1918. The Vegetation of the Eastern End of Oneida Lake. N. Y. State Museum Bull. 197, pp. 61-110. Albany. Howard, H. Eliot 1920. Territory in Bird Life. Pp. 1-308. E. P. Dutton & Co., New York. Imms, A. D. 1929. A General Textbook of Entomology. Pp. 1-703. E. P. Dutton & Co., Inc., New York. Johnson, Charles W. 1927. Further Studies of Protocalliphora Infesting Nestling Birds. Bull. Northeastern Bird Banding Association, Vol. 3, No. 4, pp. 77-82. 1929. The Injury to Nestling Birds by the Larvae of Protocalliphora. Ann. Ent. Soc. Am., Vol. 22, No. 1, pp. 131-135. 1931. Nestling Birds Destroyed by the Larvae of Protocalliphora. Bull. Boston Soc. Nat. Hist., No. 59 (April), pp. 21-24. Kellogg, Vernon L. 1908. American Insects. Pp. 1-693. Henry Holt & Co., New York. Kelso, Leon 1931. Some Notes on Young Desert Horned Larks. The Condor, Vol. 33, No. 2, pp. 60-65. Kendeigh, S. Charles, and Baldwin, S. Prentiss 1928. Development of Temperature Control in Nestling House Wrens. Am. Nat., Vol. 62, pp. 249-278 (May-June). Contrib. No. 14, Baldwin Research Laboratory, Gates Mills, Ohio. Lantz, David E. 1914. Economic Value of North American Skunks. U. S. Dept. Agric, Farmers' Bull, 587, pp. 1-22. Leonard, Mortimer D., et al. 1928. A List of the Insects of New York, etc. Cornell University Agr. Exp. Sta., Memoir 101, pp. 1-1121. Ithaca. 232 Roosevelt Wild Life Annals Lincoln, Frederick C. 1927. Returns from Banded Birds 1923 to 1926. U. S. Dept. Agric, Tech. Bull. 32, pp. 1-95. Lincoln, Frederick C, and Baldwin, S. Prentiss 1929. Manual for Bird Banders. U. S. Dept. Agric, Misc. Pub. 58, pp. 1-112. Lvrz, Frank E. 1918. Field Book of Insects. Pp. 1-509. G. P. Putnam's Sons, New York. Martin, Ernest G., and Weymouth, Frank W. 1928. Elements of Physiology. Pp. 1-7S4. Philadelphia. McAtee, W. L. 1912. Methods of Estimating the Contents of Bird Stomachs. The Auk, Vol. 29, No. 4, pp. 449-464. 1920. Farm Help from the Birds. U. S. Dept. Agric, Yearbook Separate 843, pp. 253-270. 1926. The Relation of Birds to Woodlots in New York State. Roosevelt Wild Life Bull, Vol. 4, No. 1. pp. 1-152. 1929. Further Notes on Insect Inhabitants of Bird Houses. Proc. Ent. Soc. Wash, Vol. 31, No. 6, pp. 105-111. Metcalf, C. L, and Flint, W. P. 1928. Destructive and Useful Insects, their Habits and Control. Pp. 1-918. McGraw-Hill Book Co, Inc., New York. Meyer, G. Ralph 1916. Weight and Contents of Birds' Eggs. The Auk, Vol. 33, No. 1, pp. 81-82. Nuttall, Thomas 1832. Manual of the Ornithology of the United States and of Canada. The Land Birds. Pp. 1-683. Cambridge, Mass. Pearse, A. S, and Hall, F. G. 1928. Homoiothermism. Pp. 1-119. John Wiley & Sons, Inc., New York. Plath, O. E. 1910. Parasitism of Nestling Birds by Fly Larvae. The Condor, Vol. 21, No. 1, pp. 30-38. 1919a. A Muscid Larva of the San Francisco Bay Region which Sucks the Blood of Nestling Birds. Univ. Calif. Pub. in Zool, Vol. 19, No. 5, pp. 191-200. Ridgway, Robert 1904. The Birds of North and Middle America. Bull. U. S. Nat. Mus, No. 50, Part 3, pp. 1-801. Robinson, Helen J. 1927. Tree Swallow Habits and Behavior at Brewer, Maine. Bull. Northeastern Bird Banding Assoc, Vol. 3, No. 4, pp. 89-93. Rogers, Charles Gardner 1927. Textbook of Comparative Physiology. Pp. 1-635. McGraw-Hill Book Co, Inc., New York. Sadler, Nettie M. 1926. Birds of Syracuse and Vicinity. Pp. 1-20. William Nottingham Junior-Senior High School Print, Syracuse, N. Y. Saunders, Aretas A. 1926. The Summer Birds of Central New York Marshes. Roosevelt Wild Life Bull, Vol. 3, No. 3, pp. 335-476. 1929. The Summer Birds of the Northern Adirondack Mountains. Roosevelt Wild Life Bull, Vol. 5, No. 3, pp. 320-499. Shannon, Raymond C, and Dobroscky, Irene D. 1924. The North American Bird Parasites of the Genus Protocalliphora (Calliphoridae, Diptera). Jour. Washington Acad. Sci, Vol. 14, No. 11, pp. 247-253. Shaw, William T. 1928. The Spring and Summer Activities of the Dusky Skunk in Captivity. New York State Museum Handbook 4, pp. 1-92. Studies on the Bank Swallow 233 Simpson, S., and Galbraith, J. J. 1905. An Investigation into the Diurnal Variation of the Body Temperature of Nocturnal and Other Birds and a few Mammals. Journ. Phys., Vol. 33, pp. 225-238. Starling, Ernest H. 1926. Principles of Human Physiology. Pp. 1-1074. Lea and Febinger, Philadelphia. Stoddard, Herbert L. 1931. The Bob-white Quail, Its Habits, Preservation and Increase. Pp. 1-559. Charles Scribner's Sons, New York. Stoner, Dayton 1917. The Pentatomoidea of the Lake Okoboji Region. Bull. Lai). Nat. Hist., State Univ. Ia., Vol. 7, No. 4, pp. 39^7. 1922. Bird Banding and Incidental Studies. Proc. la. Acad. Sci., Vol. 28, pp. 151-159 (1921). 192.1. Observations and Banding Notes on the Bank Swallow. The Auk, Vol. 42, No. 1, pp. 86-94. 1925a. The Toll of the Automobile. Science, N. S. Vol. 61, No. 1568. pp. 56-57. 1926. Observations and Banding Notes on the Bank Swallow, II. The Auk, Vol. 43, No. 2, pp. 198-213. 1926a. Temperature Studies on the Bank Swallow. Anat. Record, Vol. 34, No. 3, p. 32 (Abstract). 1927. On the Temperature of the Bank Swallow. Proc. Ia. Acad. Sci.. Vol. 34, p. 338. 1928. Observations and Banding Notes on the Bank Swallow, III. The Auk, Vol. 45, No. 1, pp. 41^15. 1928a. Observations and Banding Notes on the Bank Swallow, IV. The Auk, Vol. 45, No. 3, pp. 310-320. 1928b. The Increase in Temperature and Weight of Young House Wrens. Proc. Ia. Acad. Sci., Vol. 35, pp." 337-339. 1932. Ornithology of the Oneida Lake Region : With Reference to the Late Spring and Summer Seasons. Roosevelt Wild Life Annals, Vol. 2, Nos. 3 and 4, pp. 268-764. Torrey, Bradford 1903. Sand Swallows (Riparia riparia) Nesting in Sawdust. The Auk, Vol. 20, No. 4, pp. 436-437. Uchida, Seinosuke 1932. Studies of Swallows by the Banding Method. Bird-Banding, Vol. 3, No. 1, pp. 1 — 11. U. S. Department of Agriculture — Weather Bureau 1922. Summary of the Climatological Data for the United States by Sections. Reprint of Section 102 — Central New York, pp. 1-17. 1925. Climatological Data. Iowa Section. Vol. 36, No. 6, pp. 41-48: No. 7, pp., 49-56; No. 8, pp. 57-64. 1926. Climatoloeical Data. Iowa Section. Vol. 37, No. 6, pp. 41^18; No. 7, pp. 49-56; No. 8, pp. 57-64. 1927. Climatological Data. Iowa Section. Vol. 38, No. 6, pp. 41-48; No. 7, pp. 49-56; No. 8, pp. 57-64. 1931. Climatological Data. New York Section. Vol. 43, No. 5, pp. 33-40; No. 6, pp. 41-48; No. 7, pp. 49-56; No. 8, pp. 57-64. Wetmore, Alexander 1921. A Study of the Body Temperature of Birds. Smithsonian Misc. Coll., Vol. 72, No. 12, pp. 1-52. 1929 Birds and Automobiles. Bull. Northeastern Bird-Banding Assoc., Vol. 5, No. 4, pp. 141-143. THE ROOSEVELT WILD LIFE MEMORIAL As a State Memorial The State of New York is the trustee of this wild life Memorial to Theodore Roosevelt. The New York State College of Forestry at Syracuse is a State institution supported solely by State funds, and the Roosevelt Wild Life Forest Experiment Station is a part of this institu- tion. The Trustees are State officials. A legislative mandate instructed them as follows : "To establish and conduct an experimental station to be known as 'Roosevelt Wild Life Forest Experiment Station,' in which there shall be maintained records of the results of the experiments and investigations made and research work accomplished ; also a library of works, publications, papers and data having to do with wild life, together with means for practical illustration and demonstration, which library shall, at all reasonable hours, be open to the public.'' [Laws of New York, chapter 536. Became a law May 10, 191').] As a General Memorial While this Memorial Station was founded by New York State, its functions are not limited solely to the State. The Trustees are further authorized to cooperate with other agencies, so that the work is by no means limited to the boundaries of the State or by State funds. Provi- sion for this has been made by the law as follows : '"To enter into any contract necessary or appropriate for carrying out any of the purposes or objects of the College, including such as shall involve cooperation with any person, corpora- tion or association or any department of the government of the State of New York or of the United States in laboratory, experimental, investigative or research work, and the acceptance from such person, corporation, association, or department of the State or Federal government of gifts or contributions of money, expert service, labor, materials, apparatus, appliances or other property in connection therewith." [Laws of New York, chapter 42. Became a law March 7, 1918.1 By these laws the Empire State has made provision to conduct forest wild life research upon a comprehensive basis, and on a plan as broad as that approved by Theodore Roosevelt himself. Roosevelt Wild Life Bulletin, Vol. 6, No. 2. October, 1933. 1. The White-tailed Deer of the Adirondacks. Part 1. Preliminary Survey of the White-tailed Deer of the Adirondacks. M. T. Townsend and M. W. Smith. Part 2. Ecology of the White-tailed Deer in Summer with Special Reference to the Adirondacks M. T. Townsend and M. W. Smith. 2. Some Late Winter and Early Spring Observations on the White-tailed Deer of the Adirondacks Chas. J. Spiker. Roosevelt Wild Life Bulletin, Vol. 6, No. 3. July, 1935. 1. A Popular Account of the Bird Life of the Finger Lakes Section of New York, with Main Reference to the Summer Season Chas. J. Spiker. ANNALS Roosevelt Wild Life Annals, Vol. 2, No. 1. January, 1929. 1. The Red Squirrel: Its Life History and Habits, with Special Reference to the Adiron- dacks of New York and the Harvard Forest R. T. Hart. Roosevelt Wild Life Annals, Vol. 2, No. 2. October, 1929. 1. The Ecology of Trout Streams in Yellowstone Park Richard A. Muttkowski. 2. The Food of Trout Stream Insects in Yellowstone Park. . Richard A. Muttkowski and Gilbert M. Smith. Roosevelt Wild Life Annals, Vol. 2, Nos. 3 and 4 (Double Number). January, 1932. 1. Ornithology of the Oneida Lake Region ; With Reference to the Late Spring and Summer Seasons Dayton Stoner. Roosevelt Wild Life Annals, Vol. 3, No. 1. January, 1932. 1. Parasites of Oneida Lake Fishes. Part I. Descriptions of New Genera and New Species H. J. Van Cleave and J. F. Mueller. Roosevelt Wild Life Annals, Vol. 3, No. 2. October, 1932. 1. Parasites of Oneida Lake Fishes. Part II. Descriptions of new species and some gen- eral taxonomic considerations, especially concerning the trematode family Heterophyidae Justus F. Mueller and Harley J. Van Cleave. 2. Trichodina renicola (Mueller, 1931), a ciliate parasite of the Urinary bladder of Esox niger Justus F. Mueller. Roosevelt Wild Life Annals, Vol. 3, Nos. 3 and 4 (Double Number). July, 1934. 1. Parasites of Oneida Lake Fishes. Part 3. A Biological and Ecological Survey of the Worm Parasites. H. J. Van Cleave and Justus F. Mueller. Part 4. Additional Notes on Parasites of Oneida Lake Fishes, including Descriptions of New Species Justus F. Mueller. Roosevelt Wild Life Annals, Vol. 4, No. 1. December, 1935. 1. Studies on Some of the Small Mammals of Central New York M. T. Townsend.