Appendix G: Underreporting of Hybridization
Theory
|
Eugene M. McCarthy, PhD
There is good reason to believe that underreporting of hybridization is widespread, that is, there are probably many types of hybrids that do occur, but that have not yet been reported. In addition, many other natural hybrids that are reported are probably more common than generally supposed. Various factors contribute to this underreporting.
In many cases, hybrids are common within a particular region (the hybrid zone) but occur only rarely elsewhere (e.g., the already mentioned hybrids between the Black-headed and Rose-breasted grosbeaks). Surveys taking into account the entire ranges of the parental forms are therefore likely to convey the impression that hybrids are rare, when in fact they are common in regions where the parent types come into contact. On the other hand, although a new form would be expected to arise within the confines of a hybrid zone, once established, it might spread far beyond the limits of the zone so that its connection with its parents can no longer be inferred on the basis of spatial distribution. It is known that many natural allopolyploids have ranges larger than those of their parents. Under such circumstances it might not occur to an investigator that the organism in question was of hybrid origin. The possibility might therefore go uninvestigated.
Short and Robbins (1967: 542) assert that “the remarkable appearance of intergeneric [avian] hybrids renders them more likely to be noticed by collectors and banders than the usually less obvious hybrids between congeneric species.” In other words, hybrids between very distinct parents are often very obvious. But the kind of hybrids that probably occur most frequently, those between similar somatypes, are more difficult to detect and therefore less likely to be reported. For example, Clausen et al. (1945) note that the allopolyploid Townsend's cordgrass (Spartina townsendii) has been mistaken for both of its parents, smooth cordgrass (S. alterniflora) and small cordgrass (S. maritima = S. spartina). They also say a figwort known as Plumas County beardtongue (Penstemon neotericus) has been confused with both of its parents, mountain blue penstemon (P. laetus) and azure penstemon (P. azureus).
Similarly, Short and Horne (2001: 453) state that two birds, the Thick-billed and Lesser honeyguides (Indicator conirostris and I. minor), are so similar that hybridization is hard to detect. Hybrids between the similar Eastern and Western meadowlarks (Sturnella magna and S. neglecta) are another example. This difficulty probably accounts, at least in part, for the lack of reported hybrids in certain taxonomic categories. Thus, according to Holyoak (2001), the dearth of reports of hybridization among caprimulgids (nighthawks, whip-poor-wills, etc.) may simply result from difficulties in recognizing hybrids between birds that are themselves often hard to distinguish.
Despite the fact that most hybrids are intermediate in appearance, those from certain crosses closely resemble one parental form, even when those forms greatly differ. One hybridizer of begonias (Brilmayer 1960: 188) notes, for example, that wax begonias Begonia semperflorens (familiarly known as "semps"), "are such forceful parents that, no matter what other type they are crossed with, the result is another, even though slightly different, semp."
Stebbins (1950: 309) notes that some hybrids "may resemble one or the other of their parental species so closely that they have not been recognized as distinct by systematists." He gives as an example Nasturtium microphyllum, which so closely resembles ordinary watercress, N. officinale, that it was thought to be an autopolyploid form. However, Stebbins (ibid) states that it is now thought to be a hybrid between N. officinale and some species of a different but related genus, Cardamine. He also mentions (ibid) Madia citrigracilis, which he says "was at first thought to be a form of M. gracilis, but later … was synthesized artificially via the triploid hybrid between M. gracilis and the very different diploid species M. citriodora." Clausen et al. (1945) list a variety of plant hybrids that have been confused with their parents due to similar morphology. Grant (1981: 304) notes that many similar examples have been added since 1945, and that "this pattern of variation can be said to be a common one."
An example among birds is the population on Mangere Island, New Zealand, which was thought to be the only extant population of Forbe's Parakeet (Cyanoramphus forbesi). Using genetic techniques, Chan et al. (2006) showed that at least 81 percent of the birds in that population are hybrids between Forbe's Parakeet and the Red-crowned Parakeet (C. novaezelandiae), even though many of the birds in the population appeared to be C. forbesi on the basis of morphology. Another example is the hybrid between Egyptian Goose (Alopochen aegyptiacus) and the Ruddy Shelduck (Tadorna feruginea) Finn (1907: 18) says this not uncommon hybrid is very similar to T. ferruginea and that "the very marked characteristics of the Egyptian goose disappear almost completely except in [the pink color of] the legs."
It may well be that this phenomenon, of F1 hybrids looking almost identical to one of their parents, is itself underreported since such hybrids would only be detected if their parentage were known. Who knows how many hybrids masquerade as pure individuals in this manner?
In crosses producing partially fertile hybrids, backcross hybrids are often highly similar to the parental type to which the backcross occurs. As Stebbins (1950: 310) notes, if "a polyploid has originated from hybridization between two closely related species, with chromosomes partly homologous to each other, some derivatives of this polyploid may be hardly distinguishable from autopolyploids of one or the other parental species. And if through backcrossing, such a polyploid acquires a preponderance of genes derived from one or the other of the parental species, it may fall entirely within the range of variation of the latter" (such misclassification is especially likely given the fact that hybrid specimens are often mistakenly used in defining the range of variation characteristic of their pure parents). When hybrids between Long-tailed Finch (Poephila acuticauda) and Black-throated Finch (P. cincta) backcross to either parent, the resulting hybrids are almost identical to that parent.
Many animals that appear on the basis of their morphology to be pure American bison (Bison bison) contain a significant percentage of cattle nuclear genes. The Black-throated Green Warbler (Dendroica virens) and Townsend’s Warbler (D. townsendi) hybridize in southwestern Canada (Alberta). Dunn and Garrett (1997: 305) say genetic analysis showed a bird was hybrid even though it “was exactly identical to Townsend’s in appearance.” Vallender et al. (2007) genetically analyzed 48 birds with plumage identical to that of a Golden-winged Warbler (Vermivora chrysoptera). They were taken from a region where Golden-winged Warblers regularly hybridize with Blue-winged Warblers (V. pinus). Of the 48 birds, seemingly pure on the basis of plumage, 11 (23%) turned out to be hybrids.
Many types of organisms are difficult to study because they breed in inaccessible habitats (e.g., tropical rainforests, remote islands, cliffs, the depths of the sea) or have habits that make them harder to investigate (e.g., are nocturnally active). Hybridization among such organisms, although it may occur frequently, would likely go unobserved. Many organisms are also poorly known because their diet makes them difficult to keep in captivity. Gramivorous birds (e.g., finches) are easily kept and have been extensively hybridized in captivity. In comparison, captive hybridization is very poorly studied for insectivores such as tanagers and wood warblers. In addition, many hybrids are observed, but not formally reported. For example, according to the Birds Australia Rarities Committee, there are several “reliable, but unpublished reports” of hybrids in Australia between the Sooty and Pied Oystercatcher (Haematopus fuliginosus and H. longirostris).
Biologists working in the field or classifying specimens have tended to think of hybrids as very rare. But an observer does not look for something she does not expect to see. Over the years, then, many encounters with hybrids have likely gone unreported, written off as sightings of aberrant individuals or morphs. However, knowledge of avian hybridization is on the increase. Randler (1998, 2001b) suggests that a growing awareness of the existence of hybrids and improved identification literature have brought about the recent large increase in avian hybridization reports. Higgins and Davies (1996: 307) say certain new sandpiper hybrids lately reported probably reflect “increasing interest in plumages, hybridization and field identification of waders rather than changes in breeding habits and distribution.” The common occurrence of hybridization in plants is now widely recognized, at least among botanists, agricultural breeders, and horticulturalists. Even today, though, it is not unusual to encounter the assumption that natural animal hybrids are so rare that they are unlikely to be seen. This attitude tends to prevent consideration, for example, that a collected specimen might be hybrid, or that a given form of life might be of hybrid origin, and in the long run results in systematic reporting bias.
The writer's own experience with birders and ornithologists reveals that individual variation is often overlooked in the rush to classify specimens and sighted birds. While many hybrids are obviously intermediate between two types, many others not so exactly intermediate will end up classified as pure individuals of the type to which they tend. For example, two songbirds, the Tufted Titmouse and the Black-crested Titmouse (Parus bicolor and P. atricristatus), have a hybrid zone in the south central United States. A USGS website notes that “there is a tendency for observers to specifically identify all titmice detected on routes in or near this hybrid zone.” This predisposition to record hybrids as pure parentals guarantees underreporting. Dittman and Cardiff (2003: 12) say that although there may well be natural hybrids between Cave Swallows (Hirundo pelodoma) and Cliff Swallows (Hirundo pyrrhonota) "we will have to assume that birds of mixed-species colonies are 'pure' until demonstrated otherwise." One might well question why Dittman and Cardiff feel compelled to make such an assumption, particularly when they themselves point out that some individuals classified as Cliff Swallows may well be hybrids of this sort. Regarding this propensity of birders to overlook hybrids, famed ornithologist David Sibley remarks (1994: 164) “the human brain has a remarkable capacity to ignore glaringly inappropriate features while categorizing a bird based on other, more familiar, features. My experience is that an unexpected hybrid does not stand out at first glance.”
In fact, there is a tendency for writers to talk of variation “within a species,” even when much of the variation is known to be the product of hybridization with some other type of organism. For example, it is well known that most polyploids are derived from hybridization. But polyploids are often treated as conspecific with their diploid relatives, which obscures the fact that hybridization has occurred. Lewis (1980b) lists many examples. Kaufmann (1990) says that in the field hybrids between the Red-naped and Yellow-bellied sapsuckers (Sphyrapicus nuchalis and S. varius) can't be identified with certainty because “the parental forms are similar enough that a bird which seems intermediate could just be an extreme variant of one form or the other.” Actually, since these two birds are in fact known to hybridize extensively (McCarthy 2006), any bird that seems intermediate is almost certainly a hybrid, not just "an extreme variant." However, in all those cases where hybridization between two types of organisms has not yet been recognized, this tendency to describe hybrids as extreme variants of one or the other of the parental forms will certainly obstruct recognition of the true situation and is bound to decrease the number of hybrids reported.