CHAPTER FOUR

Nomenclatural Reform and the Quest for Standards and Stability

DISCIPLINING ORNITHOLOGY

Systematics—the naming, classification, and description of organisms—represented the core activity of scientific ornithology in the United States throughout the second half of the nineteenth century.1 Nowhere is this more boldly asserted than in the introduction to Robert Ridgway’s magnum opus, The Birds of North and Middle America, a multivolume technical manual that represented the fruit of more than three decades of intensive labor.² According to Ridgway, there were "Two essentially different kinds of ornithology: systematic or scientific, and popular." The first dealt with "The structure and classification of birds, their synonymies, and technical descriptions,” while the latter touched on birds’ “habits, songs, nesting, and other facts pertaining to their life-histories.” Although Ridgway admitted that scientific and popular ornithology were “closely related and to a degree interdependent,” as a primary architect of late nineteenth-century American ornithology, he left no doubt about where his allegiance ultimately lay:

Popular ornithology is the more entertaining, with its savor of the wildwood, green fields, the riverside and seashore, bird songs, and the many fascinating things connected with out-of-door Nature. But systematic ornithology, being a component part of biology—the science of life—is the more instructive and therefore more important. Each advance in this serious study reveals just so much more of the hidden mysteries of creation, and adds proportionately to the sum of human knowledge.³

By the time he penned these words in 1901, a number of scientific ornithologists and many more in the broader American ornithological community took strong exception to Ridgway’s rigid characterization of their endeavor.⁴ But in terms of identifying what had been at the heart of serious ornithological research for the preceding half-century and what remained its central defining activity, Ridgway was essentially correct.

One need only compare typical publications from the first and second half of the nineteenth century to see striking evidence of a remarkable transformation, a narrowing of a much broader ornithological gaze. Volumes once considered the acme of North American ornithology—like Alexander Wilson’s American Ornithology (1808-1814) and John James Audubon’s Birds of America (1827-1838)—contained not only a complete enumeration of all the birds known to inhabit the continent, but also exquisite hand-colored drawings and detailed accounts of the behavior and life histories of each individual species.⁵ However, the appearance of Baird, Cassin, and Lawrence’s Pacific Railroad Survey Report in 1858 marked the beginning of a new era. While this landmark study still contained colored plates of the new species discovered in the West, it entirely ignored descriptions of the life history and behavior of birds that had been central to the earlier compilations of Audubon and Wilson. The text of the publication that provided the inspiration for the next generation of scientific ornithologists contained only exhaustive synonymies, technical descriptions of plumage variations, and reports of geographical distribution.

This profound change in the content of ornithological publications reflected an equally profound shift in their intended audience. Where Wilson and Audubon sought to make their work accessible to a broad audience that included scientists and novices alike, the dreary tomes that characterized scientific ornithology in the second half of the nineteenth century were increasingly aimed at the small number of technically oriented specialists. A complete inventory of North American species remained an overarching goal throughout the century—it was what one ornithologist called the “alphabet of ornithology”—but following the death of Audubon in 1851, scientists increasingly narrowed the focus of their investigations.⁶ Research into the external morphology and geographical distribution of North American birds, largely based on the vast skin collections amassed during the second half of the century, increasingly dominated American ornithological science.

In limiting themselves to taxonomic and faunistic studies, American ornithologists were clearly taking their cue from abroad. According to historian Paul Farber, between the years 1820 and 1850 European ornithologists succeeded in firmly establishing their field as a scientific discipline characterized by “an international group of recognized experts, working on a set of fruitful questions, using an accepted rigorous method, and holding a common goal.”⁷ That goal was the production of an exhaustive catalog of the birds of the world, organized to reflect their natural relationships.⁸ But as Farber points out, substantial progress in this ambitious endeavor exacted a heavy toll. In their rush to construct a complete inventory of the world’s birds, scientists increasingly ignored the lives of the individual species comprising that inventory.

American ornithologists worked both with an awareness of and in curious opposition to this European tradition. They wholeheartedly embraced the challenge to create a complete inventory of bird forms, but, unlike their counterparts abroad, who had access to cosmopolitan collections gathered from the far corners of expanding empires, Americans limited their vision almost entirely to their own continent. Americans also accepted the need for uniform standards to govern the process of naming and describing birds, but they sought to reform the widely accepted Strickland Code, the code of nomenclature adopted by the British Association for the Advancement of Science in 1842.⁹

By the second half of the nineteenth century, achieving the goal of a single, authoritative list of North American birds remained elusive. The very factors that would seem necessary to construct a definitive list—an increase in the number of ornithologists and the systematic exploration and gradual settlement of the continental United States—had instead resulted in a proliferation of claims for inclusion in the catalog of North American birds. By the early 1880s two divergent lists, each of which gained many devoted followers, had been created in separate attempts to sort out these competing claims. But because neither list was able to supplant the other, and neither author was willing to back down, nomenclatural chaos continued to threaten the integrity of ornithological research in this country.

The ostensibly scientific goal of nomenclatural reform also had a strongly political dimension. The American nomenclatural reform movement was informed by personal ambition, individual rivalries, nationalistic pride, and the promotion of a particular research agenda: what I call the subspecies research program. The backbone of this program was the call for formal recognition of distinct geographical races (or subspecies) by the addition of a subspecific designation to the more traditional binomial name. Although the idea of using trinomial nomenclature was clearly not without precedent in Europe, its promoters almost invariably presented it as a distinctly American innovation.¹⁰ One ornithologist’s boast is typical: “Trinomialism is known as the ‘American School’ of ornithology, and the central idea is the ‘American idea.’”¹¹

The subspecies program not only set American ornithologists apart from many (though not all) of their European colleagues, it also became a source of tension in the United States. Many members of the extended ornithological community not only rejected the need to recognize the minute distinctions upon which subspecific differences were characterized, but even the need for a specialized scientific nomenclature. The issue was one of several that divided the relatively small group of technically oriented specialists from the larger ornithological community. For those primarily oriented toward collecting and displaying birds—and not toward the production of scientific knowledge—the pages of ink spilled over nomenclatural quibbles seemed entirely wasted.

A primary reason for the creation of the AOU in 1883 was to sort out the competing lists of North American birds and provide institutional sanction to the subspecies research program. Within three years of the organization’s founding, AOU leaders published an official Code and Checklist that embodied their distinct nomenclatural vision. Despite significant opposition both in the United States and abroad, this document soon gained the authority its creators desired. In 1901, the same year that Ridgway declared systematic ornithology superior to popular ornithology, the delegates gathered at the Fifth International Congress of Zoology in Berlin ratified an International Code of Nomenclature that embodied many principles first enunciated in the AOU Code, including the provision for trinomial nomenclature.

The larger goal of nomenclatural stability remained elusive, however, as North American ornithologists continued to create new subspecies based on smaller and smaller distinctions and to discover prior names for previously accepted forms. By the turn of the century, even several original proponents of trinomialism began to question whether its practitioners had gone too far.

THE GEOGRAPHY OF SPECIES

Theory and practice were inextricably linked in the development of the American version of the subspecies concept. The particular form of the concept embodied in the AOU Code and Checklist first became the subject of sustained discussion in the United States in the early 1870s, but the idea had its roots in the pervasive interest in geographical distribution in the years surrounding the publication of Darwin’s Origin of Species.¹² As early as the 1840s MCZ founder Louis Agassiz began to sketch his theory of “biological provinces.”¹³ In broad strokes, Agassiz delineated the boundaries for what he believed to be the unique, large-scale associations of animals inhabiting different portions of the world.¹⁴ The only way to gain accurate knowledge of the extent of these associations was by gathering as much information as possible concerning the distribution of their component species.

Agassiz also believed that geographical distribution data was crucial in determining the natural boundaries of species within a genus. His classic Essay on Classification (1857) characterized the higher taxonomic categories—basic types, classes, orders, families, and genera—using strictly morphological differentia.¹⁵ But Agassiz rejected using morphology alone to delineate the boundaries of species, which he believed were to be defined based on the fact that they “belong to a given period in our globe” and “hold definite relations to the physical conditions then prevailing and to the animals and plants then existing.” First among his lengthy list of examples of “definite relations” exhibited by species was geographical range.¹⁶ Agassiz considered this data to be so important that he once claimed “every new fact relating to the geographic distribution of well known species is as important to science as the discovery of new species.”¹⁷

Agassiz’s interest in geographical distribution also found expression in the printed circulars he regularly issued to collectors, sport hunters, and naturalists in an effort to maintain a steady stream of specimens flowing into his museum. Among the advice on collecting, preserving, and shipping natural history objects contained in these circulars was the reminder that all specimens, even of the most common species, were welcomed if their geographic origin was properly indicated: “A great mistake, constantly made by those who desire to contribute to the increase of the Museum, consists in the supposition that only rare specimens are desirable. It should be understood that beyond a 150 miles from Cambridge, every specimen, which may be sent in with a label indicating its origin, will be greatly received and acknowledged.”¹⁸

Agassiz’s rival museum builder, Spencer Fullerton Baird, was also vitally interested both in determining the geographical distribution of individual species and in reconstructing the broader associations of organisms inhabiting larger geographic areas. Like Agassiz’s circulars, Baird’s printed instructions for collectors highlighted his desire to achieve wide geographical representation for the species in the Smithsonian collections: “As the object of the Institution in making its collections is not merely to possess the different species, but also to determine their geographical distribution, it becomes important to have as full series as practicable from each locality.”¹⁹

Figure 12. Smithsonian Institution Bird Gallery, 1885. This photograph shows birds preserved as old-fashioned taxidermie mounts and as more modern study skins.

Baird’s statement hints at one result of the increasing theoretical emphasis on geographical distribution: changing collecting practices. Until the middle decades of the nineteenth century, private collectors and museums generally aimed to acquire a single pair, male and female, of each desired species.20 Additional examples of the same species were usually considered duplicates to be traded, sold, or discarded. An increased interest in geographical distribution and a separate but related growth in the study of variation within species led naturalists to amass increasingly large series of specimens. Although serial collecting had begun in the United States prior to the appearance of the Origin of Species, Darwin’s book, and the subsequent interest in evolutionary studies it engendered, greatly accelerated this trend.

As I have already noted, serial collecting resulted in a tremendous increase in the size of individual and institutional collections. This growth, in turn, forced changes in exhibition practices. While museums had previously tended to display all of the specimens contained in their collections, the large amount of material amassed through serial collecting resulted in pressure to create separate research collections removed from the public eye. In ornithology, serial collecting also resulted in a movement away from the practice of mounting birds (fig. 12) and toward the production of “study skins,” which were much more convenient to store and less prone to damage from dust, light, and insects.²¹

Accompanying the growth of serial collecting was a move toward increased precision in specimen description. In this respect, Baird’s ornithological volume of the Pacific Railroad Survey Reports was clearly an epoch-making work.²² Baird’s contribution was one of a series of reports presenting the scientific findings of the survey teams sent out in the mid-1850s to evaluate possible routes for the transcontinental railway.²³ But rather than limiting his report simply to a description of the avian species collected by the naturalists attached to the survey teams, Baird used the opportunity to produce a complete survey of North American avifauna, providing exact diagnoses of all known species and genera, with copious synonymy and critical commentary.

Of course, the idea of publishing a complete enumeration of the North American bird forms was hardly new. Alexander Wilson, Charles Lucien Bonaparte, John James Audubon, and others had attempted exhaustive surveys in the first half of the nineteenth century.²⁴ But Baird presented much more precise and detailed information than his predecessors. Not only did he give exact descriptions of each species, he also provided a complete list of all the specimens on which his descriptions were based. These tabular lists contained data on each of up to fifty individual specimens, including museum catalog number; sex; information on when, where, and by whom the specimen was originally collected; and a standard series of up to ten different measurements of each bird, taken as closely as the nearest hundredth of an inch (fig. 13). Because of his innovations, Baird’s Pacific Railroad Survey Report became a model for American naturalists struggling to bring more precision to natural history description.²⁵

Thus in the middle decades of the nineteenth century, American naturalists showed a keen interest in geographical distribution and the practices necessary to support that interest—large series collections of species taken over widely ranging geographic areas and increasingly precise techniques of measurement and description capable of distinguishing even the most minute individual differences. It was only a small step to recognize that species might exhibit regular variations that correlated with the changing environmental conditions across their range.

THE “AMERICAN” SUBSPECIES CONCEPT

After carefully examining an extensive series of box turtles (Cistudo Carolina) obtained through his collecting network in the 1850s, Agassiz recognized four separate and apparently constant North American forms of the species. Each of these four forms seemed to coincide roughly with a unique geographic area.26 The forms were so distinct that Agassiz was unsure whether they represented several species or a single species with different well-marked geographic forms. Although he was soon to deny the existence of any kind of constant variation from the species type, in 1857 Agassiz hinted at a research agenda that was to occupy many American naturalists for the next several decades: “The differences noticed may indicate different species; but they may also mark only varieties. There is, however, a remarkable circumstance connected with the specimens that come under my observations: their variations are limited to particular regions of the country.”²⁷

Figure 13. Baird’s list of red-winged blackbird specimens, 1858. An example of Baird’s efforts to introduce more rigor into the practice of American natural history.

As early as 1868 Joel A. Allen accepted his mentor’s challenge to investigate the possibility of regular patterns of geographic variation within certain species. Following an unsuccessful bid to obtain bird specimens through a widely circulated appeal, in the winter of 1868-1869 Allen mounted an expedition to Florida.28 He hoped that by comparing a large series of birds of the same species obtained in Florida with enough examples collected in Massachusetts, more than one thousand miles to the north, he could shed light on the vexing problem of whether geographical varieties actually existed in nature.

Alien’s research came in the wake of Agassiz’s recent denial of the existence of constant variation from the species type. Where Agassiz once entertained the possibility that certain well-marked varieties might correspond with a given geographical location, more recently he had rejected the possibility because of his strong opposition to Darwin’s Origin of Species.²⁹ Apparently Agassiz had come to recognize the evolutionary implications of his earlier position.³⁰

Allen was also responding to several of Baird’s recent publications that had touched on the subject of geographical variation. Beginning as early as his Pacific Railroad Survey Report of 1858, Baird had hinted that changes in size, form, and coloration of a species might correspond with the climatic differences of particular geographic regions.³¹ One year later he again called attention to these climatic differences and announced his first general law of geographic variation: a gradual decrease in size of individuals of the same species with decrease in latitude or altitude of their birthplaces.³² By 1866 Baird published a series of “laws” and “tendencies” of geographic variation. In addition to his previous correlation of size with latitude and altitude, he also noted a tendency toward absolute increase in the size of the bill in birds born farther south (even though they were subject to a general diminution of bulk across this same range) and that change in longitude from east to west led to increase in the length of tail and a general darkening of the color of birds.³³

Apparently Baird did not consider his generalizations concerning geographical variation to be terribly monumental or original. They occur as passing references within much longer books and articles largely devoted to other subjects. Moreover, he provided only limited data to back up his assertions. And he was careful to cite a German predecessor, Constantin Gloger, who had noted regular geographical variation in European birds nearly three decades previously.³⁴

While Baird’s references to changes in birds across their geographic ranges remained brief, largely undocumented assertions, Allen’s 1871 paper presenting the results of his Florida expedition contained a lengthy discussion of the phenomenon of geographical variation. Allen first showed that the level of individual variation in the size and color of species was much larger than had been previously appreciated; in some cases different specimens of the same species were found to vary as much as 20 percent in size. Because taxonomists had frequently relied on size to differentiate species and even genera, the failure to recognize the high level of individual variation had led to the introduction of “numerous strictly nominal species.”³⁵

Allen then differentiated between this background level of individual variation and larger trends in that variation reflected in population-wide averages. Because he believed the characteristics responsible for these trends were directly induced by temperature and humidity, Allen frequently referred to this latter type of variation as “climatic variation,” a term he used interchangeably with geographical variation.³⁶ Allen demonstrated that morphological variation in populations could be roughly correlated with the location of the population by employing the same tabular presentation of data that Baird had first used in his Pacific Railroad Survey Report. However, Allen introduced a crucial innovation into his lists of specimens: tables with summaries showing the minimum, maximum, and average sizes of the specimens originating from various geographic locations (fig. 14). Allen thus provided a wealth of empirical data to substantiate his generalizations about the level of regular geographical variation in several bird species.

Figure 14. Alien’s list of red-winged blackbird specimens, 1871.

Allen ended his discussion with a consideration of some of its implications. The original description of most American birds had been made by European ornithologists, often based on a single specimen. However, with access to a large series of specimens collected across their continent, American ornithologists had begun to show that many of these apparently distinct forms intergraded through specimens from intermediate localities. They were not distinct species, but only different “varieties, races, or simply forms” of a single species. Nomenclaturally, the appropriate way to handle these variant forms was to use the first name applied to any of them and to include with its specific description a statement of its tendency to vary with locality and the degree to which that tendency was developed.³⁷

Ironically, although Allen failed to advocate the use of trinomials in the designation of geographical forms in 1871, his “Mammals and Winter Birds” paper did more to promote the wide-scale adoption of the practice in America than any other single work. It did so by providing evidence for a clear and simple distinction between species and subspecies (or what he then usually called geographical forms or varieties): subspecies are forms that intergrade across their range, while species do not. It was this distinction that provided the “guidance to a methodical and consistent trinomialism” promoted by the “American school.”³⁸ Trinomial nomenclature had occasionally been used in both American and European ornithological works, but before Allen had formulated his “test of intergradation,” its use was sporadic and often unsystematic.³⁹

In a June 1871 review of “Mammals and Winter Birds of Florida,” Elliott Coues challenged Allen’s claim that geographical variants should all be referred to a single species. Instead he recommended that trinomial nomenclature ought to be used to designate such forms.⁴⁰ Coues also suggested that the distinct geographic forms Allen had described probably represented an initial stage in the formation of species.⁴¹ In doing so he linked the geographic variation studies of Baird and Allen with the wider stream of American evolutionary speculation initiated by Darwin’s Origin of Species.⁴²

A year later Coues published his Key to North American Birds, the first comprehensive listing of American birds that systematically applied trinomials using Allen’s newly formulated test of intergradation.⁴³ Coues recognized distinct geographic forms (or subspecies) of a species by giving them all the same specific name and providing each with a different third name, with the abbreviation “var.” inserted between the specific and subspecific epithets.⁴⁴ One of the first artificial keys in zoology designed for both novice and advanced naturalists, Coues’s Key was an immediate commercial and critical success.⁴⁵ As a result, both the idea of using trinomial nomenclature to recognize subspecies and the particular names contained in the Key gained a large following.

Between the time he published his 1871 paper and the time he assumed editorship of the Nuttall Club Bulletin in 1876, Allen had also become a proponent of recognizing geographical variation with trinomial nomenclature. His first use of that nomenclature, with the abbreviation “var.” inserted between the specific and subspecific epithet, came less than one year after his “Mammals and Winter Birds” paper.⁴⁶ Five years later, Allen endorsed the use of pure trinomial nomenclature without intervening abbreviations. He noted that the discovery that “certain strains of deviation of pronounced types (i.e., geographic variation) occurred in a large number of species in different families” led to the recognition of “the subspecific relationship of many forms which when first made known seemed unquestionably of specific rank.”⁴⁷ Now, he argued, ornithologists ought to provide nomenclatural recognition of such forms:

The next step, and apparently a wholly logical one in the revolution, will doubtless be the general adoption of a trinomial system of nomenclature for the more convenient expression of relationship of what are conveniently termed “sub-species,” so that we may write, for instance, Falco communis anatum in place of the more cumbersome Falco communis subsp. anatum. This system is already, in fact, to some extent in use here, though looked upon with strong disfavor by our transatlantic fellow-workers, who seem not fully to understand the nature of the recent rapid advance ornithology has been made in this country, or to appreciate the thoroughly substantial nature of the evidence on which it is based.⁴⁸

Allen then pointed out that recognition of geographical variation and the subspecies was made possible by the successive waves of exploration undertaken by American collectors, which resulted in the acquisition of “hundreds and often thousands of specimens of a single species, representing the gradually varying phases presented at hundreds of localities.”⁴⁹

As editor of the Bulletin of the Nuttall Ornithological Club, Allen promoted the use of trinomial nomenclature in two ways. First, beginning with the initial volume in 1876, he regularly published papers with trinomial subspecies designations. Second, the Bulletin became a forum for frequent outcries for nomenclatural reform. In one extensive paper discussing the issue, Robert Ridgway argued that the binomial system created by Linnaeus was no longer adequate now that most naturalists recognized that species were not stable and fixed. Ridgway then issued what was becoming an increasingly familiar call for the recognition of subspecies “by a suitable amendment of the rules of nomenclature.”⁵⁰ Under Allen’s guidance, the subject of trinomial nomenclature failed to receive criticism in the Bulletin, even though there was significant opposition to the innovation both in the United States and abroad.⁵¹

Meanwhile, in 1880 Robert Ridgway published a catalog of North American birds that adopted trinomial nomenclature, but differed greatly from the checklist issued by Coues six years previously.⁵² Throughout their lives, the two naturalists clashed on a variety of issues, but the animus between them was ultimately related to differences in temperament and the fact that in 1874 Ridgway secured a permanent appointment on the staff at the Smithsonian Institution, something that Coues had long desired but never achieved.⁵³ Ridgway revised the list and arranged for its separate publication a year later under the title of Nomenclature of North American Birds (1881 ).⁵⁴ Following the publication of Ridgway’s list, one prominent ornithologist complained:

I see by Ridgway’s new catalogue that he has made a great change in our nomenclature, if it is accepted by ornithologists in general. It never will be in Europe and will cause much confusion to ornithologists who do not confine themselves to the study of a comparatively small tract of country. We shall also be finding varieties of varieties a few years from now and will see something like the following[:] Passer domesticus domesticus americanus longerostris nigricaudatus &tc &tc with a string of authorities following.⁵⁵

But with the imprimatur of one of the nation’s most prestigious scientific institutions behind it, Ridgway’s list quickly won converts from Coues’s earlier checklist.⁵⁶

Never one to back down from controversy, Coues answered Ridgway’s challenge by issuing his own revised checklist a year later.⁵⁷ The move prompted C. Hart Merriam to comment impatiently: “It seems very peculiar that our leading ornithologists can not agree upon the names by which we shall know our common birds. The contrary is, to say the least, extremely unfortunate.”⁵⁸ Scientific ornithologists were not the only ones disturbed by the prospect of two contradictory lists. In a short notice appearing in the collecting-oriented Ornithologist and Oologist, the author complimented Coues’s workmanship but decried the chaos it might introduce into specimen exchanges: “It will have extensive circulation, but it will be unfortunate if a portion of our collectors should recognize its numbers, as it will create confusion in every transfer, unless specimens are doubly numbered.”⁵⁹ Within the next year, the controversy led to the call for the formation of the AOU to settle the issue.

THE AOU AND NOMENCLATURAL REFORM

Nomenclatural reform and the subspecies research program were foremost on the minds of the three AOU founders from the time they first began to contemplate the organization. As early as January 1883, Coues’s thinly veiled call for a national association of ornithologists hinted at the pressing need to achieve nomenclatural uniformity as a primary reason to institutionalize ornithologists.60 And the formal invitation to the inaugural AOU meeting issued by Coues, Brewster, and Allen in early August 1883 announced that a special object of the new organization would be the “revision of the current lists of North American Birds, to the end of adopting a uniform system of classification . .. carrying the authority of the Union.”⁶¹

In responding to the invitation to attend the founding meeting of the AOU, Robert Ridgway announced his strong support for the proposed organization, and especially “the prospect of gaining unanimity in the matter of nomenclature.”⁶² In a letter to Allen a week later, Ridgway stressed the importance of keeping the matter of nomenclatural reform firmly in the hands of a few carefully selected scientific ornithologists:

The great difficulty which I see in the way of having the question of nomenclature discussed by the “convention as a whole,” is that this would involve the value of the views of a considerable number of amateurs, some of them with the crudest possible information on the subject, as opposed to the intelligent discussion of the subject by a few well informed specialists. The only way, in my opinion, would be to have the matter put into the hands of a committee composed exclusively of working ornithologists, and have this committee hold a special session for the purpose of coming to an agreement. ... I cannot too strongly urge, however, the necessity of limiting such a committee to a very carefully selected number of working ornithologists who are known to possess intelligent views on the subject.⁶³

But on this point Ridgway was preaching to the choir. The three AOU founders never intended to open up the nomenclatural issue for consideration by the larger ornithological community, which had not even been invited to the first AOU meeting. Rather, they worked to keep the issue under the control of a few of the most technically proficient ornithologists who held views similar to their own.

The Committee on Classification and Nomenclature was the first of several committees created on the second day of the AOU’s inaugural meeting in September 1883.⁶⁴ As had been arranged before the meeting, temporary AOU Chairman Coues appointed himself, Allen, Brewster, Henry W. Henshaw, and Ridgway to the committee, which held its first series of seven sessions in Washington on 11-19 December 1883.⁶⁵ After reaching a consensus on the scope of its work, the basic nomenclatural rules to be incorporated into its proposed code, and several species to be included in its official list of North American birds, the committee divided itself into two subcommittees to hammer together a final report. Allen and Coues placed themselves on the subcommittee to compose the AOU’s code of nomenclature, while Ridgway, Henshaw, and Brewster were charged with formulating the official checklist based on the new code. During these and subsequent deliberations in March and September 1884 and April 1885, the committee sought the aid of several other prominent American naturalists, including Theodore Gill, Leonhard Stejneger, C. Hart Merriam, Charles B. Cory, and, of course, Spencer F. Baird.

The committee presented its proposed code and checklist to the AOU Council for formal approval in April 1885, and the work was finally published nearly a year later, in March 1886.⁶⁶ Although largely based on the Strickland Code, which had first been adopted by the British Association for the Advancement of Science in 1842, the new AOU Code also contained several novel provisions that distinguished it from its predecessors. Among the major innovations claimed by the framers of the AOU Code were the adoption of the tenth edition of Linnaeus’s Systema Naturae (1758) as the starting point of the law of priority (instead of the twelfth edition [1766]) and allowing names previously employed in botany to be available for zoology.⁶⁷ But the most important innovation was Canon XI, which advocated the use of pure trinomial nomenclature to designate intergrading subspecific forms.⁶⁸ Allen used the lengthy “Remarks” section that followed Canon XI to clarify the committee’s intent regarding the application of trinomial nomenclature:

Trinomials are not necessarily to be used for those slightly distinct and scarcely stable forms which zoologists are in the habit of calling “varieties”; still less for sports, hybrids, artificial breeds, and the like; nor indeed to signalize some grade or degree of difference which it may be desired to note by name, but which is not deemed worthy of scientific designation. The system proceeds upon a sound scientific principle, underlying one of the most important zoological problems of the day,—no less a problem than that of the variation of animals under physical conditions of the environment, and thus of the origin of species itself. The system is also intimately connected with the whole subject of geographical distribution of animals; it being found, as a matter of experience, that the trinomial system is particularly pertinent and applicable to those geographical “subspecies,” “races,” or “varieties” which have become recognizable as such through their modification according to latitude, longitude, elevation, temperature, humidity and other climatic conditions.⁶⁹

Allen continued by pointing out that it was not the “kind or quality, nor the degree or quantity, of difference” of one organism in relation to another that determined whether it should be labeled with a binomial or trinomial designation. Differences that were “constant,” no matter how small, might be considered as distinct species and hence named using strictly binomial nomenclature. However, any difference “no matter how extreme in its manifestation, that is found to lessen and disappear when specimens from a large geographical area, or from contiguous faunal regions ... is to be provided for by some other method than that which formally recognizes ‘species’ as the ultimate factors in zoological classification.” Allen then succinctly summarized the American basis for the application of trinomial nomenclature: “Intergradation is the touchstone of trinomialism.”⁷⁰

In the concluding section of his commentary on Canon XI, Allen again revealed that the American promotion of trinomial nomenclature for intergrading forms was a source of nationalistic pride: “It is gratifying evidence, therefore, of the progress of Ornithology, and of the position attained by that branch of science in America, that the members of an American Ornithological Association have it in their power first formally to enunciate the principles of the new method, the practicality of which they have already demonstrated to their fellow workers in Zoology.”⁷¹

As the editor of the Auk, Allen consistently promoted the adoption of the new AOU Code and Checklist. For example, he arranged for David Starr Jordan, a prominent ichthyologist and president of Indiana University, to write a sympathetic review of the publication for the Auk.⁷² Although Jordan had some minor quibbles with two of the code’s many canons, he was generally laudatory and hopeful that the AOU’s handiwork would soon become the standard nomenclatural authority not only for American ornithology, but for “zoology and botany” as well.⁷³ In the early years of his editorship, Allen also routinely edited manuscripts submitted to the Auk to make them conform to new AOU Code and Checklist.⁷⁴

AOU leaders initiated a series of related reforms to promote the subspecies research program and consolidate their discipline. One of these reforms involved the push to introduce uniform standards into the practice of scientific ornithology. In 1888 the council of the union appointed a special committee to devise “some uniform method of measuring birds.”⁷⁵ Because scientists often created new subspecies based on small average differences in the lengths of birds and their appendages, consistency in the way those measurements were taken was crucial to the subspecies research program. Not surprisingly, the committee found that “systems employed by different authors varied widely” and two years later issued a report recommending standard methods of taking measurements of the bill and tail.⁷⁶ But for reasons that remain unclear, shortly after submitting what was supposed to be the first of several reports, the committee disbanded. It was over forty years later before there was another serious attempt to address the problem of inconsistency in bird measurement.⁷⁷

In a separate call for reform in measurement practices during this same period, several Americans advocated abandoning “the confusing and irrational system of inches and hundredths in the measurement of birds and eggs” for the more “rational” and widely adopted metric system. For example, Merriam argued that the founding of the AOU marked the “commencement of a new era in American ornithology.” Now that the union was about “to establish a stable nomenclature,” it ought to give up “the barbarous scale of our forefathers and join the men of science of all nations in adhering to a system of weights and measures that is uniform throughout the world.”⁷⁸ Again, this reform achieved only modest success. Although the AOU Council decided not to require the use of the metric system in the Auk, a number of authors began to use it routinely.⁷⁹

In 1886, the same year as the appearance of the AOU Code and Checklist, Robert Ridgway published a nomenclature of colors in an effort to standardize this notoriously subjective attribute.⁸⁰ Containing named examples of some two hundred colors, the first edition of Ridgway’s Nomenclature of Colors for Naturalists came to be used widely by scientific ornithologists both in the field and in museums. Ridgway, an accomplished artist, remained interested in the subject of color nomenclature throughout his lifetime, and in 1912 he published an expanded version of his reference work. The second edition of Ridgway’s nomenclature contained over a thousand named colors.⁸¹

PLAIN ENGLISH

By the time the AOU issued its official checklist in 1886, collectors had begun to speak out against the introduction of trinomial nomenclature. The issue was part of a larger discussion about the amateur’s inability to keep abreast of developments in an increasingly technical scientific ornithology. I have already mentioned the resentment that many AOU associate members felt at being asked to support the Auk, which was often unintelligible to all save a few knowledgeable specialists. Associates were especially critical of articles that failed to include the common names of birds, which collectors generally preferred over scientific names.82 For the most part, they and the larger collecting community also rejected scientists’ claims about the need to designate by formal names the increasingly minute morphological differences upon which most subspecific distinctions were based. Rallying under the banner of "Plain English,” collectors sought to maintain ornithology’s wide accessibility.

Underlying the controversy were fundamentally conflicting visions of the scientific endeavor. One was a more inclusive model that sought to reconcile the needs of scientists with those of the larger community of amateur practitioners. The other model presented the construction of boundaries between the technically oriented specialist and the novice as a necessary step in the continued development of science. Unlike many scientific disciplines that tended to move firmly in the direction of exclusivity and increasingly rigid boundaries as the nineteenth century came to a close, most scientific ornithologists continued to be guided by a combination of both visions. It was a tension that plagued ornithology well into the twentieth century.

As it did with several other divisive issues, the Ornithologist and Oologist provided a regular forum for opposition to the AOU. Recall that in November 1883 the periodical’s editor and owner, Joseph M. Wade, had ridiculed scientific ornithologists’ preoccupation with nomenclature in a suppressed editorial protesting the exclusive membership policy of the new AOU.⁸³ A few months earlier, Wade had published a short letter entitled “Plain English,” which praised his editorial policy of rendering articles in a more accessible “popular style.” According to the letter’s author, Montague Chamberlain, too much scientific writing was rendered in overly jargon-ridden prose, which made it inaccessible not only to the public, but even to informed amateurs like himself, who studied “a branch of science for the pleasure to be derived from it or for a relaxation from more engrossing labor.” Wading through the morass of obscure, polysyllabic appellations contained in most scientific treatises tried the patience of even the most dedicated amateur ornithologists:

These [amateurs] are obliged, in order to keep informed of the latest discoveries, to read the determinations of the leading observers in their chosen departments, but it is in much the same spirit as that with which they submit to the manipulations of a dentist, that they worry through the tedious pages filled with unattractive and often obscure sentences, with Latin and Greek terms and names which are hard to spell, hard to pronounce, hard to remember, and harder still to understand.⁸⁴

Scientific publications “could be written quite as easily and with as exact precision without the constant use of these technicalities.” But authors purposefully used obscure prose to make their writing unintelligible to all but a select few, and thereby “to throw over science that veil of mystery which is so dear to the savant, and beneath which he delights to pose as the custodian of knowledge too profound for ordinary mortals to comprehend.”

In a letter appearing a month later, Chamberlain also questioned the right of the self-styled “American School” to ignore the British Association’s authority in nomenclatural matters. According to Chamberlain, the trinomialism promoted by the “American School” was not only contrary to the letter of the British Association’s Strickland Code, but also uniformly condemned by European naturalists.⁸⁵

For several years after 1886, much of the discussion of nomenclatural matters in the Ornithologist and Oologist centered on the new AOU Code and Checklist. For example, in May 1886 an anonymous editorial noted impatiently that with the recent appearance of the AOU’s official inventory, there were now five disparate catalogs of North American birds.⁸⁶ The author hoped that this latest list would finally bring stability to nomenclature, but he doubted that the AOU list would immediately come into wide use since its $3.00 price tag placed it “beyond the reach of many who desire several lists for use in labelling and exchange.”⁸⁷ After a brief editorial again decrying the steep price of the AOU list in 1888, a year later the periodical called the continued use of several contradictory systems “embarrassing” and suggested that collectors and dealers label their specimens, catalogs, and exchange lists using a fraction composed of the AOU number in the denominator and the Ridgway number in the numerator until all could agree on a single uniform system.⁸⁸

During the next few years, the Ornithologist and Oologist pursued a different tack in its criticism of the AOU and its system of trinomial nomenclature: satiric verse. The first venture in this genre was a twenty-five stanza gem that openly ridiculed the practice of creating subspecies based on barely discernible differences in size or plumage color.⁸⁹ The poem described an amateur ornithologist’s excitement upon receiving a common English sparrow mailed to him in jest by two close friends. After carefully inspecting his new specimen and comparing it to others of the same species already in his collection, the ornithologist was delighted to discover that his new sparrow was “just one shade darker, on the edge of the breast” and therefore could be designated as a new subspecies. The poem then described how the act of bestowing a subspecific epithet transformed the previously humble collector into a vainglorious ornithologist, whose dubious achievement was certain to result in his election as an honorary member of the AOU:

When next we did meet him, three weeks to a day,

You scarcely would know him, so changed was his way;

A sense of his GREATNESS, his wisdom and skill,

His manner, his action, his life seemed to fill.

Now that his name is known wide and far,

He’ll hereafter smoke naught but a ten-cent cigar;

And at the next meeting (the hints are not few)

There are prospects of election as an “H” A.O.U.⁹⁰

For those readers unable to discern the author’s critical tone from the body of the poem, the concluding stanza left no doubt as to his attitude toward the proliferation of subspecies:

Kind reader, we call to your careful attention

This wonderful growth of a modern invention.

Lord knows where we’ll end if this craze increases,

Such a trotting out yearly of created SUB-SPECIES.

Several months later the magazine published a second poem, which again simultaneously ridiculed the needless creation of subspecies, scientific ornithology, and the AOU.⁹¹ The protagonist in the story was the editor of an unnamed ornithological periodical who, while sorting through his correspondence late one evening, came upon two curious letters. Both were from collectors who had recently captured several shrikes and had carefully noted the stomach contents of their specimens. The first collector found that the diet of his birds had consisted entirely of “insects, bugs, and the like.” The second collector, who had shot his shrikes only after witnessing them prey upon other birds, not surprisingly found only the remains of “sparrows, and other small birds” in the stomachs of his specimens. The remainder of the poem detailed how the clever editor resolved the dilemma presented by these contradictory accounts of the shrike’s diet by splitting the species into two subspecies:

Now the editor, he was ambitious too,

He wanted his share of the fame;

He wanted to see at some future time

An alphabet tacked to his name.⁹²

So he said to himself, I will settle this thing,

I’ll make two sub-species of these;

I’ll take Mr. A’s, which feed upon bugs

Which they glean from the ice-covered trees,

And call it the L. b. insec-ti-vorous⁹³

And the other will then follow on,

In shall be the L. b. ornitho-vorous

And the thing shall be regarded as done.

Now all ye august and mighty A.B.’s

Who hold our fates in your hands,

Assemble your conclaves, and get out your “Keys”

And loosen the tightly drawn bands.

Regard this poor mortal thus thirsting for fame,

He’s anxiously waiting the Tail

Which a letter will bring, to affix to his name

By the next “U.S.S.D.” mail.

When the Ornithologist and Oologist fell victim to the depression of 1893, the AOU lost one of its most constant critics. Most other collecting-oriented periodicals tended to use common nomenclature in their pages and to ignore the issue of scientific nomenclature entirely. Typical of this strategy was the O & O’s chief rival for the patronage of collectors, the natural history dealer Frank H. Lattin’s Oologist. In the first year of publication, 1884, Lattin announced that since “ninety-nine one-hundredths, if not all collectors” abhorred Latin names, he would use “plain English” whenever possible.⁹⁴ When the AOU published its code, the review appearing in the Oologist was generally sympathetic, but it discussed only the changes in common names, not the scientific names it sanctioned.⁹⁵

Where Lattin had tended to ignore the constant name changes in scientific nomenclature, Richard M. Barnes, the lawyer and egg collector who purchased the magazine in 1909, repeatedly raised the issue. Barnes attacked the tendency to create subspecies based on increasingly fine distinctions discernible only to specialists with access to extensive research collections. In one typical editorial, Barnes also decried the widening gap between “professional Ornithologists and amateurs.” According to the editor of the Oologist, most of the “so-called ‘professional’ ornithologists” made a living from taxes assessed on the public, which included a great many amateur bird students. Yet these same “professionals” completely ignored the potentially valuable contributions of amateurs, while seeking to bolster their reputation by “discovering alleged geographic races of birds [i.e., subspecies], the delineation of which to the tax-paying public off of whom they live is, and always will be an unfathomable mystery.”⁹⁶ Often Barnes condemned the “scientific snobbery” exhibited by those who insisted on using only Latin names when common names were much more accessible and often more stable.⁹⁷ Writing in the early decades of the twentieth century, Barnes had a vision for ornithology that was more in keeping with the heterogeneous community in the pre-AOU days than with the increasingly differentiated community of his time.

Discussion of amateur discontent with trinomialism was not confined to collecting journals. From the time of the first issue in January 1884, the Auk contained regular references to the popular outcry against nomenclatural recognition of subspecies. Again it was Montague Chamberlain, an earlier champion of the use of “plain English” in scientific treatises, who asked “Are trinomials necessary?” By this point, Chamberlain, who had recently been elected to the AOU Council and appointed associate editor of the Auk, was clearly acting with the knowledge and encouragement of AOU leaders.⁹⁸ They hoped that by airing the issue under carefully controlled circumstances, they might win converts to the subspecies program. As someone who had recently spoken out against excessive use of jargon and scientific nomenclature in ornithological writing, Chamberlain could credibly present himself as the champion of the amateur.

Chamberlain began by pointing out that in asking if trinomials were necessary, he was not trying to persuade advocates of the practice to repudiate it. Rather his goal was to “have the whole matter plainly set forth, and, if possible, an end put to the opposition to this system” among amateur ornithologists: “Let me state here, that I do not wish to assert that this opposition occurs in the ranks of the more advanced of American students—the ‘scientists’ . . . indeed so far as I am aware, it is found only among a portion of my brethren of the ‘amateur element.’” ⁹⁹ Montague also noted that opposition to trinomialism came from a number of “savants of Europe,” who, with the sole exception of Henry Seebohm, had uniformly denounced the system.¹⁰⁰

In his reply to Chamberlain’s query, Allen detailed the logic behind trinomial nomenclature as it was promoted by the “American school” and then attempted to document support for the system abroad. In an effort to boost the authority of the practice he had helped create, Allen claimed that the trinomial system did not do “violence” to the earlier Strickland Code, but that it was a means “to meet simply and completely, a condition of things unknown and unsuspected when that . . . admirable system was conceived [i.e., evolution].” He also cited important European precedents for trinomialism, including Hermann Schlegel of the Leyden Museum, who had begun using trinomials as early as 1844. A few years earlier, American ornithologists, anxious to make a name for themselves, had ignored Schlegel’s earlier use of trinomials. Now Allen attempted to use the name of the Dutch ornithologist to boost the authority of the AOU’s actions. But in doing so Allen was careful to point out that ornithologists in the United States had begun advocating trinomials independent of their knowledge of Schlegel’s earlier use: “While he antedates Americans in the systematic use of trinomials for intergrading forms, we are in position to know that the ‘American school’ was the spontaneous outcome of our studies of American birds, and that the use of trinomials was forced upon us by conviction of their utility and necessity.”¹⁰¹

Apparently unconvinced by Allen’s reply, three months later Chamberlain called for further clarification on several points. According to Chamberlain, “American amateur ornithologists” had no particular opposition to the use of three terms to distinguish varieties from species. Rather, they denied the “necessity of recognizing varieties by any distinctive appellation”:

We harbor no “Dr. Dry-as-dust” “craze” for a purely binomial nomenclature, but we do protest against the propagation of any system which unnecessarily creates obstacles to the study of science, instead of simplifying it; we do ask that our leaders shall not take a step backward and force upon us something which is . . . not only no improvement, but a palpable injury; that we not be dragged into a “craze” for trinomialism by following the lead of an “American school.”¹⁰²

According to Chamberlain, the “amateur element” had several complaints with trinomialism. First was the belief that “recognition of varieties” tended “to create confusion, in classification and nomenclature,” and increased “the difficulty of identifying specimens.” Moreover, if the argument for designating subspecies was carried to its logical conclusion, “every variation from a given type must receive a distinctive name; necessitating not alone the recognition of varieties of species, but also of varieties of varieties almost without limit.” And finally, Chamberlain argued that if forms were distinctive enough to deserve different names, they should be designated as separate species, instead of mere varieties.

Allen replied by admitting that if doubters could see a large series of specimens firsthand, they would soon be convinced of the necessity for trinomial nomenclature. Any difficulties associated with general questions of classification and nomenclature and with the creation of subspecies more particularly were not due to the adoption of trinomial nomenclature, but were “necessarily inherent in the subject.” According to Allen, there was a “philosophic principle” that undergirded the practice of bestowing trinomial nomenclature, and even the “unbelievers” of which Chamberlain spoke were “not to be presumed to be so skeptical as to ignore the modern doctrine of evolution.” Varieties, Allen asserted, were “incipient species” which were “still in the process of evolution.” This fact had to be taken into account by the system of classification and nomenclature advocated by most American ornithologists. Experts might occasionally err in their judgment on where to draw the line between species and subspecies, but on the whole they were the only “safe guides.”¹⁰³

Following this initial set of exchanges, Auk editors generally avoided entering into direct discussion on the issue of continued amateur discontent with trinomialism. But the problem was repeatedly mentioned in passing within the context of the more general nomenclatural controversy that raged in the pages of Auk for the next fifty years.

As I have already suggested, amateur opposition to trinomialism was but one small part of that community’s larger discontent with a scientific ornithology that seemed increasingly unconcerned with its needs. Yet scientists continued to depend on this larger community for specimens, legitimation, and financial support. Nowhere is this tension between professional dependence and amateur frustration more apparent than in a letter from an amateur ornithologist from Milwaukee, Walter B. Hull, to Auk editor J. A. Allen in 1897:

Have the Editors of the Auk ever considered the advantage to be gained in publishing the common names of the birds, particularly the more common, in connection with the Latin? The habit of writers using Latin entirely seems to be contagious. For a purely scientific journal, supported by scientists alone, I could take no exceptions to that course. But for the Secretary to send out a sheet every fall, urging members to interest their friends, or in plain English, get their money to help run the Auk, and as a reward, send them a quarterly with descriptions of they know not what, while these same subjects of discussion may be flitting about their lawns and they none the wiser, though perfectly familiar with all our common birds. If the manager cannot take the time to insert the [common] names in case authors neglect so to do, then do not ask the layman to help support the publication of scientists, as the management has done ever since I have been acquainted with their methods—1889.¹⁰⁴

Hull went on to state that he and most amateurs would rather see “plain English” in the Auk than the expensive color plates that regularly adorned its pages. He also claimed that he had experienced little luck in getting participants in local bird study classes to join the AOU. Although initially interested, when provided a sample copy of the Auk, these enthusiasts replied “Why don’t you have it all in Latin, it would mean just as much to us.” Hull indicated that he would withdraw his AOU membership if the situation did not improve soon, and concluded with the suggestion that “for business reasons, if no other, would it not be advisable to accommodate the layman.”

TRINOMIAL WOES

As the nineteenth century drew to a close, the three major innovations promoted by the framers of the AOU Code—the establishment of the tenth edition of Linnaeus’s Systema Naturae (1758) as the beginning point for the law of priority, the separation of zoological and botanical nomenclature, and the use of trinomial nomenclature to designate subspecies—gained wide acceptance among technically oriented zoologists in the United States and abroad.¹⁰⁵ As one indication of increasing European support for the practice of trinomialism, in 1898 the delegates gathered at the Fourth International Congress of Zoology in Cambridge, England, appointed a fifteen-member committee charged with drafting a single international code of nomenclature to supersede the various competing national and specialist codes.¹⁰⁶ After careful review of the existing codes, the committee incorporated the most important American innovations—including the designation of subspecies by use of a third term added to the traditional Linnaean binomial—into their proposed International Code of Zoological Nomenclature.¹⁰⁷ Three years later, the zoologists attending the Fifth International Congress in Berlin ratified a modified form of this draft code, and it was finally published in French, English, and German text as Règles internationales de la Nomenclature zoologique in 1905.¹⁰⁸

Despite widespread support for trinomial nomenclature, the larger goal of establishing a single, stable list of North American bird forms remained elusive. The AOU Checklist was increasingly accepted as the sole authority for the scientific names of North American birds, but the names contained in the various editions of the list (2d ed., 1895; 3d ed., 1910; 4th ed., 1931) and their numerous supplements were subject to constant revision.¹⁰⁹

One major factor responsible for the regular shuffling of bird names was rigid enforcement of the law of priority. American promoters of the practice had hoped its consistent application would lead to nomenclatural permanence by fixing a particular bird form to its first properly published name after the 1758 starting point.¹¹⁰ During the first half-century following the publication of the first AOU Code and Checklist, however, the law of priority had failed miserably to deliver on this promise. Ornithologists continued to dredge up obscure ornithological publications containing earlier names, and the decision about whether a given previous name applied in a particular case was often complex and subject to differing interpretations.

The constant stream of name changes necessitated by the strict adherence to the law of priority (and other similar guidelines established in an effort to promote nomenclatural stability) did not pass unnoticed. One of the most notorious cases of nomenclatural variability during this period was the ubiquitous American robin, the red-breasted portent of spring that one author has recently described as “the most familiar and best-loved songbird in North America.”¹¹¹ Though known to every schoolchild as the robin, the bird appeared in the first and second editions of the AOU Checklist under the scientific name Menda migratoria, in the third edition as Planesticus migratorius, and in the fourth edition as Turdus migratorius!¹¹² As scientific ornithologist and popularizer Frank Chapman pointed out, in this case and many others during the previous half-decade, the lowly common name, long eschewed by many technically oriented ornithologists, had proven more enduring than its scientific counterpart.¹¹³ Not surprisingly, in the face of the continued variability of technical nomenclature, many amateur ornithologists and collectors who had long championed common over scientific nomenclature felt vindicated.

Continued instability in technical nomenclature also disturbed many scientific ornithologists who noted the relative fixity of common names. For example, in 1905 William Leon Dawson, the author of a series of state bird books, sent an open letter to the AOU Committee on Nomenclature decrying the constant jumbling of technical names demanded by strict adherence to the law of priority. Dawson predicted that if the current trend continued, "we shall need to issue daily bulletins or publish diagrams of the nomenclatural barometer, after the fashion of the morning papers.”¹¹⁴ Four years later William Brewster repeated this complaint:

Just as eels are said to have become reconciled to being skinned alive, so most ornithologists are learning, I suspect, to regard with resignation or indifference, not unmingled with disgust, the ever-increasing and apparently quite hopeless instability of their technical nomenclature. Fortunately there are the English names of birds to which one may turn with blessed sense of relief because of their comparatively fixed and stable character. For they have changed but little since the days of Wilson and Audubon, although purists have not failed to suggest that they should be critically looked into and perhaps extensively emended. Heaven forbid that this ever come to pass! It would mean universal chaos in ornithological nomenclature. Surely we have enough of trial and tribulation to bear with this ceaseless tinkering of the scientific names.¹¹⁵

Even more than application of the law of priority, the creation of new subspecies—by combining (“lumping”) or, more often, breaking up (“splitting”) forms previously designated as full species—was responsible for the overwhelming number of changes in nomenclature during the forty-five years between the first and fourth editions of the AOU Checklist.¹¹⁶ As figure 15 shows, the number of full species recognized by the AOU nomenclature committee grew at a very modest rate during this period, beginning just below and ending just above the eight hundred mark. But the number of subspecies increased over threefold in the first four editions, from 183 in 1886 to 609 in 1931! In a review of the 4th edition, Joseph Grinnell, a California ornithologist and well-known splitter, predicted that at the current rate of increase, by the year 2000 the total number of North American bird forms would exceed 2,050, and that most of these (1,160) would be subspecies.¹¹⁷

The continued splitting of North American species was precisely the kind of taxonomic practice against which critics of the trinomialism railed. I have already noted the amateur community’s discontent with the tendency to grant subspecific status to forms based on what they considered unnecessarily fine distinctions. “Feathersplitting,” as this increasingly pervasive practice was often referred to by its critics, also disturbed many scientists. One response within the scientific community was humor. For example, a caricature that appeared in Condor in 1900 (fig. 16) depicted the “American Splitters’ Union Headquarters,” complete with a magnifying glass, microscope, color charts, and five prominent North American ornithologists known for their propensity to divide species into subspecies. During this period a joke began to circulate in ornithological circles concerning the creation of “north light subspecies,” forms based on differences so slight that they seemed to be more the result of bird skins having been viewed through a window on the north, rather than the south, side of a building, than any appreciable difference in the specimens themselves.¹¹⁸ And in 1910 William Brewster, who one year previously had decried the instability of technical nomenclature, wrote a biting letter to AOU President E. W. Nelson, in which he ridiculed the practice of designating subspecies based on increasingly minute distinctions:

Figure 15. North American bird forms, 1814-1931. This graph indicates the total number of species and subspecies listed in various ornithological publications. Data from Coues, The Coues Check List of North American Birds, 2d ed. (1882), and Grinnell, “Some Inferences from the New Check-list” (1932).

Figure 16. The American Splitters’ Union headquarters, 1900. This cartoon depicts several notorious creators of new subspecies, including (from left to right) Richard McGregor, Joseph Grinnell, Harry C. Oberholser, Edgar A. Mearns, and Wilfred Osgood.

I have just read Grinnell’s latest and it fills me with hope that the Yellow warbler that breeds in my garden may be a good and new subspecies. After I have determined this by slaying him and ascertaining that his left wing is one-thousandth of a millimeter shorter than his right (owing, no doubt, to the fact that as the garden is small he has to fly always in a circle when taking his daily exercise) I shall name him Dendroria aestiva hortensis.119

As the predictions of critics—who feared that the practice of designating subspecies using trinomial nomenclature was a Pandora’s box certain to result in a confusing proliferation of new names—seemed increasingly to have been fulfilled, even some original American proponents of the subspecies program began to complain of the recent trends in its application. The first major call for a careful reassessment of the practice of designating subspecies based on increasingly minute distinctions came from J. A. Allen in 1890. In an article entitled “To What Extent Is It Profitable to Recognize Geographical Forms among North American Birds?,” Allen provided a brief history of the splitting and lumping North American bird forms.¹²⁰ He argued that there had been “at least three well-defined oscillations of the ornithological pendulum.”¹²¹ In the period leading up to 1870, the tendency had been to create too many new species and genera from the multitude of forms sent back by the naturalists attached to the various western surveys active during the period. In the years immediately following 1871, however, there had been a swing in the opposite direction, and taxonomists had tended to reduce previously accepted species to the rank of geographical races or subspecies.¹²² This roughly five-year period of lumping was followed by a nearly a decade of relative “equilibrium” during which the AOU Code and Checklist had been constructed. But according to Allen, during the previous five years systematists had “started the pendulum again in the direction of finer discriminations and excessive splitting.”¹²³ In an argument that explicitly challenged the legitimacy of a completely separate sphere for experts, Allen warned that recent taxonomic tendencies had led to an unnecessary and dangerous proliferation of subspecies:

[G]reat caution should be exercised in bestowing trinomials, in order to guard against drawing too fine distinctions. Very little is gained by naming races distinguishable only by experts, aided by a large amount of material, or where the differentiation is largely a matter of slight average difference between forms contiguous in habitat—forms which nine out of ten ornithologists of average acuteness and experience, and with only ordinary resources, will be more or less unable to satisfactorily distinguish. . . . Only the exercise of due discretion can prevent the reduction of “our beneficial system of trinomials” to an absurdity.¹²⁴

In 1903, on the twentieth anniversary of the first AOU Committee on Nomenclature and Classification, Allen again sounded the alarm. In concluding a brief review of changes in nomenclature contained in the first two editions of the AOU Checklist and its numerous supplements, Allen referred to the “‘hair-splitting’ tendencies of the day, of which complaint is more or less prevalent.” And again he expressed not only his anxiety about the practice of designating subspecies, but also his fear that scientific ornithology was becoming increasingly divorced from the needs of the larger ornithological community:

The degree of difference necessary for formal recognition in nomenclature is ever likely to be a bone of contention, its decision being, in the nature of the case, more or less a matter of temperament as well as of opinion. The danger of excessive splitting is greater now than ever before, since we have reached a point where comparatively few strongly marked local forms remain to be discovered and named, while the number of enthusiastic young workers is steadily increasing. Plainly, not every degree of differentiation that can be recognized by the trained expert needs recognition by name, and not every slightly differentiated form that can be distinguished readily on comparison of large series of specimens should be considered as entitled to a place in a list of North American birds. The trinomial system unfortunately lends itself readily to abuse, and can easily be made to bring the whole system of naming subspecies into disrepute.¹²⁵

During this period an increasing number of other prominent American scientists spoke against the practice of splitting species based on minute distinctions. For example, in 1899 Theodore Gill, an internationally renowned ichthyologist and one of the scientists consulted during the construction of the first AOU Code and Checklist, complained of the “embarrassing” and “undue prominence” American ornithologists were giving to subspecific forms. Gill feared that continued “splitting” would lead to “an interminable number of subspecies.” Like early critics of trinomialism, he argued that scientists should describe species in “generalized terms, that is, including all the variants, and the diversification into subspecies indicated in terse phraseology immediately after the diagnosis of common characteristics.”¹²⁶ A few years later Leverett M. Loomis, curator of ornithology at the California Academy of Sciences in San Francisco, echoed Gill’s sentiment when he maintained that the geographical variation that had led to subspecific designation should be treated the same as variation due to sex, age, season, and the like: the tendency should simply be noted in the specific description of the birds.¹²⁷

In a steady stream of critical articles, with titles like “The Exaltation of the Subspecies,” “The Fallacy of the Tendency towards Ultraminute Distinctions,” and “The Tyranny of the Trinomial,” other well-known scientists decried the abuse of the subspecies system and proposed various solutions to resolve the dilemma.¹²⁸ Besides the complete elimination of nomenclatural recognition or the toning down of the prominence of the subspecies, several ornithologists proposed a less radical solution to mitigate some of the confusion associated with the nearly constant change of names: issuing official checklists only once a decade without the almost yearly supplements.¹²⁹

Despite the protests of their more conservative colleagues, technically oriented ornithologists continued to create new subspecies, while most amateurs continued to protest what they considered the unnecessary division of the species. The issue was one of several that reflected the increasing distance between the two groups. By the early twentieth century, scientists began to suggest that the split was a perfectly normal result of the continued development of science. Published in the same year that Ridgway declared the superiority of “systematic” over “popular” ornithology, the sentiments of the scientific ornithologist and mammalogist Wilfred H. Osgood were typical of those who began to argue for the legitimacy of a separate sphere for the scientist:

The tendency to revolt among the “lay” class against the so-called splitting seems to be not so much because it is thought to be based on unsound principles, but more because it brings about a multiplication of names which are hard to remember and because it makes identification of individual specimens difficult. The popular ornithologist, following in the footsteps of other popular scientists, has reached the point where he cannot keep pace with the man who gives up his life to technical work. There was a time when country gentlemen of the Gilbert White type were able to keep fairly abreast of all branches of natural science, but now to be expert in any one branch requires almost a lifetime of study. The question arises—is this a deplorable condition or is it the outcome of a vast increase in the quantity and quality of material, a corresponding increase in facilities for work, and a convenient access to useful contributive results of investigation in other branches of science? Is it strange that the careful ornithologist should continually add named and labeled facts to the sum of knowledge as well as the astronomer with his telescopes discover new stars and the histologist with his new methods of preservation find unexpected conditions?¹³⁰

Perhaps what is most significant about Osgood’s statement is the tentativeness with which he proposed that the increasing distance between scientific and lay ornithologists was to be expected and welcomed. Osgood’s arguments—which are posed as questions, not positive declarations—reflect the continuing ambivalence that many scientific ornithologists felt about their increasing divergence from other bird enthusiasts. Although desirous of increasing their autonomy, scientists were still unable, and often unwilling, to divorce themselves completely from the larger ornithological community from which they had emerged and upon which they continued to depend for legitimation and funds. The attempt to find a way to reconcile the needs of expert and novice bird enthusiasts became even more problematic with the development of the bird protection movement at the end of the nineteenth century.