THE REALITY OF BIRD SPECIES.
BY LEVERETT MILLS LOOMIS.
In 1858, in volume IX of the ‘Reports of Explorations and Surveys ... from the Mississippi River to the Pacific Ocean,’ Ammodromus samuelis Baird and Melospiza fallax Baird appear as full-fledged species. In 1874, in ‘A History of North American Birds,’ Land Birds, volume II, these so-called species are reduced in rank, being designated respectively Melospiza melodia, var. samuelis, Baird and Melospiza melodia, var. fallax, Baird. In 1886, in the first edition of the A. O. U. ‘Check-List,’ these names are altered, in accordance with earlier lists by Mr. Ridgway and Dr. Coues, to Melospiza fasciata samuelis (Baird) and Melospiza fasciata fallax (Baird), pure trinomials and the term subspecies having come into vogue. In 1910, in the third edition of the A. O. U. ‘Check-List,’ the two names are amended to Melospiza melodia samuelis (Baird) and Melospiza melodia fallax (Baird).
Owing to his lack of knowledge of geographic variation, Professor Baird gave to each of these geographic variations of the Song Sparrow an entity which they did not possess, and this entity, having gained a foothold in the literature, is perpetuated today in the subspecies (‘incipient species’). As no one can foresee the future of these variations of the Song Sparrow, it is not known whether they are the beginnings of species or not. Nevertheless, it may be urged that bird history repeats itself, and that the record of past events warrants the conclusion that bird species are now in process of evolution through geographic variation. Theorize as we may, the fact remains that we do not know what part geographic variation or other agencies played, or did not play, in the origin of existing bird species, the modus operandi of the evolution being unknown. But we do know that geographic variation is one of the common variations occurring within the bounds of a bird species of today, and that it is not the only variation in which geography is a factor.
Independent of individualism, age, sex, season, or climatic conditions, there exists a type of variation known as dichromatism, which perhaps originated in mutations. It is well exemplified in the Jaegers, Albatrosses and Petrels, Herons, Hawks, and Owls. In some species there is a difference in the geographic range of the phases, but it does not correlate with environment as in geographic variation. Instances to the point are found in the Wedge-tailed Shearwater, Red-tailed Hawk, and Screech Owl.
More than thirty years ago, when our knowledge of variation was far less than it is now, Dr. Stejneger had the discernment to interpret Colaptes auratus (Linnæus), Colaptes cafer (Gmelin), and Colaptes hybridus (Baird) to be dichromatic or trichromatic phases of one species, and not two species that hybridize on a gigantic scale.[21] None of the characteristics of dichromatism are wanting in these extremes and intermediates. They are similar in general character to the extremes and intermediates of well-known dichromatic species, of the Wedge-tailed Shearwater, Neglected Petrel, and Rough-legged Hawk for example. They are not individual and are not dependent upon age, sex, season, or environmental conditions. Moreover, intermediates crop out sporadically in the Eastern States, where the auratus phase is dominant. It is well to bear in mind that these variations of the Flicker are not greater than certain other normal variations; as the age variation of the Western Gull, the sexual variation of Williamson’s Sapsucker, the seasonal variation of the Marbled Murrelet, and the dichromatic variation of the Parasitic Jaeger.
The question naturally arises, whether dichromatism has often been misinterpreted and made the basis of apocryphal species and their supposed hybridization on a grand scale. In the alleged Junco species, for instance, possibly dichromatism or polychromatism, originating in mutations, obtains along with geographic variation.
Vermivora leucobronchialis (Brewster) and Vermivora lawrencei (Herrick) are not overlooked in this discussion. The evidence thus far presented tends to prove that they are hybrids between two species rather than intermediates of one dichromatic species.[22] Be this as it may, hybridization between unquestionable species of birds is an abnormal and relatively rare occurrence.
To affirm that bird species are concepts, is to ignore the facts in the case. Ammodromus samuelis Baird and Melospiza fallax Baird are concepts, but Melospiza melodia with all its geographic variations is a reality. It is absolutely separated from Melospiza lincolni and Melospiza georgiana and all other existing bird species. Colaptes auratus is likewise a reality. In spite of its great dichromatic variation, it does not intergrade with any other woodpecker. It is confidently stated that the great majority of the A. O. U. ‘Check-List’ species are also realities, and the remainder time-honored concepts based on inconstant variations, like Fulmarus rodgersi Cassin, which is merely an extreme white phase of Fulmarus glacialis (Linnæus).[23]
In a word, absence of intergradation among birds results in a definite entity, the existing bird species.
GEOGRAPHICAL VARIATION IN
THE BLACK-THROATED LOONS.
BY A. C. BENT.
Dr. Jonathan Dwight’s interesting paper in ‘The Auk’ for April, 1918, describing a new species of Loon from northeastern Siberia, has opened up a subject to which I have given considerable study without having been able to come to any satisfactory conclusion. After examining directly or indirectly some seventy specimens of Black-throated Loons, including the entire series in several of the largest collections in this country, I came to the conclusion that the necessary material was still lacking to settle satisfactorily the true status of this group.
I have long recognized the existence of a large, Green-throated Loon in the Bering Sea region; but I have postponed publishing anything on it until I could obtain enough breeding birds from somewhere in that region, to establish a more or less definite breeding range in which a more or less constant form is to be found. Now that Dr. Dwight has seen fit to open up the subject, I feel called upon to publish what incomplete data I have on the whole group.
It seems to me that there are only two alternative theories into which the known facts may be made to fit. The first and most likely theory is that there is but one circumpolar species, divided into three, or possibly four, subspecies, as hereinafter designated. To support this theory we need more material from Siberia and eastern Europe to show complete intergradation between the two intermediate subspecies, arctica and suschkini, though what material we have seems to indicate that such intergradation exists. An argument against this theory is the fact that the two extreme subspecies, viridigularis and pacifica, apparently breed side by side in northeastern Siberia and northwestern Alaska.
The second theory is that there are two species, arctica in Europe, with viridigularis as a Siberian subspecies occupying a subarctic area, and pacifica in North America, with suschkini as a Siberian subspecies occupying the Arctic coast. This theory would explain the breeding of the two extreme forms in the same or in contiguous areas; but it would be upset by the discovery of more complete intergradation, unless such intergrades could be regarded as hybrids. A final choice between these two theories cannot be made until more material is available showing the distribution and relationships of the forms to be found in Siberia, a vast and little known region.
I will now attempt to state, roughly and in general terms, the main known facts in this complicated case and let the reader judge for himself how they fit in with the above theories. There are apparently three or four fairly well marked subspecies of Black-throated Loons, as follows:—
1. Gavia arctica pacifica (Lawrence), the smallest of all, in which the hind neck or nape is much lighter gray than the crown or forehead, nearly white in some cases, the black throat patch terminates below in a straight line and the metallic reflections of this patch almost always appear purplish in any light. This form occupies a breeding area which includes the whole of northern North America (which need not be more definitely outlined here), the Arctic Islands west of Greenland and the Arctic coast of Siberia for our unknown distance westward.
2. Gavia arctica suschkini (Sarudny), intermediate in size between arctica and pacifica, but nearer the latter, in which the colors are nearly as in pacifica, but with a slight tendency towards arctica. This form probably has a breeding range somewhere on the northern coast of Asia, but is known only from specimens taken in winter or on migrations in the Ural and Turkestan regions.
3. Gavia arctica arctica (Linnæus), intermediate in size, but nearer viridigularis than pacifica, in which the crown and nape are uniform dark gray, the black throat patch terminates below in a point and the reflections of this patch appear either purplish when held away from the light and greenish when held towards it, or wholly purplish in any light, with considerable individual variation. This form inhabits northern Europe, and northern Asia for an unknown distance eastward and southward in Siberia.
4. Gavia arctica viridigularis (Dwight), the largest of all, but intergrading perfectly with arctica, in which the crown and nape are colored as in arctica, the black throat patch terminates below in a point and the reflections of the throat are usually more greenish than in the others. I have yet to see a specimen in which more or less purple reflections could not be found. Even Dr. Dwight’s type shows “slight purplish tints.” This form, if it is a good subspecies, has no well defined habitat; but what specimens I have seen would seem to indicate a breeding range on both sides of Bering Sea, which may extend for a considerable distance westward into the interior of Siberia.
The above arrangement may appear satisfactory to the casual observer, but the trouble with it is that all of the above characters, particularly those on which Dr. Dwight bases his new species, are decidedly variable and inconstant. Size is the most satisfactory character but even this shows intergradation or overlapping and greater individual variation in each group than the differences in averages between the groups. The measurements, in inches, of the four forms, which I have taken or had sent to me, are as follows:—
The other characters are equally confusing. The nape is lightest and almost constantly so in North American pacifica; it is darkest in viridigularis and more or less intermediate in many specimens of the other two forms.
The black throat patch terminates below in a straight line almost invariably in North American pacifica; I have seen but one exception to this rule; but in Siberian pacifica this character is less constant. In viridigularis this patch terminates below in a decided point, in all specimens that I have seen. In European arctica about half of the specimens I have seen have the patch decidedly pointed below and the others have it nearly straight or only slightly pointed.
The colored reflections of the black throat-patch are the most variable and inconstant of all the characters. In viridigularis three of the specimens examined show mainly greenish colors but even these show some signs of purple; and in one, a bird in my own collection, the colors are about equally divided. In European arctica about half of the specimens show mainly purplish reflections, while fully half show both purplish and greenish. In North American pacifica the purplish reflections predominate, but five specimens out of twenty-two show more or less greenish in certain lights. Mr. Waldron DeWitt Miller, in sending me descriptions of Pacific Loons in the American Museum, used the following terms in designating the colors of the throats; greenish-blue, bluish-green, dark greenish-blue, violaceous and dark violet. It can be easily seen from the above that the colors are very variable.
Dr. Dwight says, in his diagnosis of viridigularis:—“The green coloration of the throat is the essential character that sets this species apart from arctica and its races, which all have purple throats.” In the light of the facts stated above this “essential character” disappears and his new species must be reduced to the rank of a subspecies at least. Even a subspecies must prove to be fairly constant in a more or less definite range. The range of viridigularis is very imperfectly known; the four specimens, referable to this form, that I have seen were taken at Nijni Kolymsk, Siberia, St. George Island, Bering Sea, Nome and Saint Michael, Alaska; Dr. Dwight’s specimens all came from northeastern Siberia. The Nijni Kolymsk bird, referred to above, is somewhat intermediate between viridigularis and arctica; if it had been taken in Europe it would probably be referred to the latter. I also have a perfectly typical pacifica from the Kolyma River, Siberia.
I have seen birds from Victoria, B. C., from Finland and from Norway which closely approach this new form, viridigularis, in size and color characters. If we had a larger series of arctica from Europe and Asia available for comparison, we could perhaps match these birds exactly and we could certainly show, if I have not already demonstrated it, that viridigularis is merely a subspecies of arctica. To use Dr. Dwight’s own terms, the green throat seems to be a quantitative rather than a qualitative character.
REASONS FOR DISCARDING A PROPOSED
RACE OF THE GLAUCOUS GULL
(LARUS HYPERBOREUS).
BY JONATHAN DWIGHT, M. D.
In discussing the moults and plumages of the Glaucous Gull, a dozen years ago I took occasion to bury “Larus barrovianus” among the synonyms of Larus hyperboreus (then known as glaucus) because the alleged characters seemed to me to afford insufficient grounds for recognizing even a subspecies (Auk, XXIII, 1906, p. 29). Later, in the 1910 edition of the A. O. U. ‘Check-List,’ the Committee on Nomenclature and Classification adopted my view of the case and discarded “barrovianus”; but recently Dr. H. C. Oberholser has seen fit to dig it up and it is revived, somewhat impressively, as a subspecies of hyperboreus (Auk, XXXV, 1918, p. 472).
If it were not for certain aspects of the matter I would merely reaffirm my convictions of 1906; for it is a question whether Dr. Oberholser has added anything new to the original claims made by the describer, Mr. R. Ridgway (Auk, III, 1886, p. 330). This does not seem to be the case, for his diagnosis is virtually a restatement of Mr. Ridgway’s, except that a supposed character of the bill is discarded on evidence I submitted in 1906. My measurements had shown that this character, namely, “depth through the angle never less and usually decidedly greater than through the base,” was not diagnostic, but this was not my only “evident reason” then for rejecting “barrovianus” as Dr. Oberholser now wrongly assumes. What I said was that this form “is scarcely 3% smaller [than glaucus] in size and 4% smaller in bill” and furthermore, I said; “It is true that the largest specimens of barrovianus never quite reach the dimensions of the largest glaucus, but overlapping of size is so considerable even when careful comparison of sexes is made that without first reading the labels one cannot, except in a very few cases, tell whether a bird is from Greenland or Alaska. The variation in the size and shape of the bill in gulls is very great and a few millimeters difference in wings that are as long as one’s arm is hardly ground on which to rest a subspecies, much less a full species.”
These conclusions may be contrasted with Dr. Oberholser’s recent diagnosis which reads, “Similar to Larus hyperboreus hyperboreus, but smaller, the bill particularly so and relatively as well as actually more slender; mantle decidedly darker; and the line of demarcation between the white tips to the primaries and the pale grayish basal portions usually more evident.” I would here call attention to the fact that the “line of demarcation” is not a distinct character but a corollary of the preceding, for the color of the mantle in the Glaucous Gull regularly runs over, so to speak, into the wings, and a darker mantle would mean darker bases of the primaries and therefore greater contrast as a matter of course. Consequently, in the final analysis there are two characters and only two on which “barrovianus” rests,—(1) darker mantle and (2) smaller size, especially of the bill. I will invite attention to a new estimate of the value of these characters.
Fig. 1. Diagrams showing relative measurements in millimeters of 31 adult specimens of Larus hyperboreus and its alleged race. Top line shows actual length in largest birds, middle line shows average, and bottom line shows smallest of the series.
1. As for the color of the mantle, which Mr. Ridgway calls “somewhat” and Dr. Oberholser “decidedly” darker, I can only say that my series fails to support either of these statements. I find that if comparison of like stages of plumage be made, birds from Greenland are quite as dark as Alaska specimens and conversely Alaska birds are as pale as those from Greenland. It is, perhaps, a matter of more than passing interest that the majority of adult Greenland birds in the collections I have seen are in worn faded plumage while most of the Alaska material is in fresh dark plumage. One might easily get the impression that the darker birds represent a race unless due allowance is made.
It may not be generally known that the adult Glaucous Gull moults twice in the year, a complete postnuptial moult beginning toward the last of July and extending over nearly two months and a prenuptial in March and April which involves most of the body feathers but not the wings nor the tail. Between moults the mantle fades and looks even paler than it is in color because of the worn and whitened feather edges. There is some individual variation in the depth of color in freshly moulted specimens, whether from Greenland or Alaska, but both may be equally dark and they may become equally pale after the lapse of a few months. I have examined birds taken nearly every month in the year and I am at a loss to understand how Dr. Oberholser finds a “decidedly darker” race unless he has unwittingly compared birds of unlike stages of plumage.
2. As for size, this is a question of relative dimensions that permits some latitude of opinion, so that a new presentation of the facts seems desirable.
My early table of measurements (Auk, XXIII, 1906, p. 28) based on 31 adults (14 of them males and 17 females) is accepted by Dr. Oberholser “except for dimensions of the bill which have been remeasured for the present use.” I have reproduced all of these measurements by the graphic method [(Fig. 1)] and anyone may see, almost at a glance, what the variations of size in the Glaucous Gull actually are. The diagrams are drawn to scale, the upper horizontal line representing the actual size of the largest specimens, males and females, the middle line the mean or average size and the lower line the smallest specimens. The oblique solid lines represent hyperboreus, the broken lines “barrovianus” and the dotted lines Dr. Oberholser’s remeasurements of the bill. His “depth of bill” for “barrovianus” is the same as mine and therefore cannot be separately plotted. He does not tell us from what series he made the remeasurements that do not tally with mine, but the figures suggest that it may have been a small one and with an unusual proportion of very large and very small birds, possibly wrongly sexed in some cases.
The original series that I measured was composed of breeding birds from Greenland and from Alaska which formed a small part of the 200 specimens I had then gathered together for comparison. Although they are now widely scattered, some of them (as well as new specimens) are still either in my collection or in that of the American Museum of Natural History. A reëxamination and remeasurement of them (68 in all, 39 being adults) confirms to a surprising degree my earlier measurements and conclusions. Individual variation is greater than the supposed subspecific values and the overlapping of size is marked. Birds as large as these Gulls, it must be remembered, may not be measured with unfailing accuracy, especially when different persons attempt it, for specimens are often greatly worn, the wings or tail are sometimes not quite grown and often the feathers are bent and broken. It is not unusual to find a variation of five to ten or more millimeters between the right and left wing of the same bird, due to the make-up of the skin, while tarsi and toes of opposite legs may be bent very much out of shape in drying. Where such variation exists, one may to advantage measure each wing or foot separately and strike an average as I have done in many cases.
Turning finally to the bill, I would call attention to the sketch [(Fig. 2)] which shows the average adult bill of the male of hyperboreus contrasted with that of “barrovianus.” When one realizes that the variation in the bills of all female gulls is much greater than that of the males and that young birds only very slowly acquire adult dimensions, it becomes evident that “barrovianus” is not “very readily recognizable by its usually smaller size and particularly smaller bill.” One may guess cleverly that large birds belong to one race and small ones to another, but without reference to the labels the guesses may be astray by a continent’s width.
Fig. 2. Bill of average Larus hyperboreus, male, life size, drawn to scale. The broken line shows the bill of the alleged race.
So far as I can see the case of barrovianus stands where it did in 1906 and it is a pity that there should have been any need of reopening it. Fortunately the merits of this and similar cases do not rest upon individual bias, but they are determined by the A. O. U. Committee which, as far as North American birds are concerned, acts somewhat as a supreme court rendering verdicts according to evidence presented. Let us hope they will give us “safe and sane” subspecies rather than the shadowy indefinite groups of averages that too often are named as geographical races. It should be remembered that while a name is a handle to a fact, too many handles would make a door or a basket perfectly useless. Ornithology will become a wilderness of handles if every difference is named at sight,—a wilderness of subspecies founded more on hasty opinions than on digested facts. A step farther and we shall have the psychological subspecies in which the expectant mental attitude of the subspecialist (if I may be pardoned the word) will play the most important rôle. In our gropings after the truth it is wasteful of too much time to spend so much of it stumbling over names of groups so poorly defined that they convey only a vague meaning to a few specialists and none at all to everybody else. Decking the subspecies in all the glittering panoply of diagnosis, dimension, and distribution makes of it an impressive spectacle, but this does not necessarily make of it a good subspecies.