Australian loranths are unique among mistletoes for their frequent resemblance to primary hosts (Fig. 6.1 A-C). Leaves in particular are so similar in form and presentation to a human eye several meters away that tree and parasite are indistinguishable, although closer examination often reveals less faithful coloration and texture. Amyema, Dendrophthoe, Diplatia, Lysiana, and Muellerina are mimetic mistletoes; their hosts are most often Acacia, Casuarina, and Eucalyptus. Parasitic mimics with compressed or terete foliage attack Casuarina (e.g., Amyema cambagei on Casuarina torulosa). Those with flat leaves (e.g., Amyema sanguineum and Dendrophthoe gla-brescens) infect Eucalyptus, other Myrtaceae, and phyllodinous Acacia. Several mangroves harbor mistletoes with comparably thickened and shaped foliage. Perhaps the most striking example of convergence involves Dendrophthoe homoplastica on Eucalyptus shirleyi (Fig. 6.1.B).
Resemblance is greatest among species pairs occupying arid rather than humid parts of Australia (Barlow and Wiens 1977). Excluding the mistletoes from everwet northeastern forests (which show relatively broad host specificity) and the two endemic terrestrials, 78% of Australian loranths bear foliage resembling that of their hosts too closely for coincidence. Barlow and Wiens and others before them have postulated that vegetative similarity constitutes genetically based mimicry which has arisen independently many times in response to selective pressures and confers fitness by hiding mistletoe foliage within matrices of less palatable forage. Leaf form is supposedly especially convergent in Australia because the two dominant tree genera -Acacia and Eucalyptus - have become so well defended (i.e., relatively unpalatable because of sclerophyllous leaves containing the essential oils of Eucalyptus and the phenolics of Acacia) that the parasites have become attractive food items. Mistletoes are therefore all the more liable to seek, as it were, safety in crypsis in order to escape their putative selective agent, the arboreal possum Trichosurus and other possible foliage feeders.
Support for the predator avoidance hypothesis was obtained from analyses of leaf Kjeldahl N, hence protein content and nutritive quality (Ehleringer, Cook, and Tieszen 1986). If mistletoe foliage is more nutritive than host foliage, cryptic mimicry could be especially advantageous. Of 22 Aus tralian mistletoes that exhibited mimicry, 17 bore foliage richer in N than that of hosts. (None of the five mimics of Eucalyptus had higher [N], but presumably "hid" because these hosts are so well defended.) Fifteen of 26 nonmimetic mistletoes produced foliage with [N] averaging about a third below that of their supports. Similar findings have been reported for a smaller group of New Zealand mistletoes (Banister 1989). Nitrogen-poor parasites might be expected to benefit less from cryptic mimicry than by standing out against the more nutritive food around them; vertebrates are quite adept at sorting out the most desirable herbiage.
Atsatt's (1983) morphogen hypothesis states that evolution of host specificity was sometimes accompanied by "genetic selection for hormonal compatibility." Concordant morphogens were especially important in hosts with "highly reduced or otherwise specialized foliage," and presumably an equally extraordinary set of growth factors that excluded incompatible mistletoes. But this proposal ignores the basic tenets of plant morphogenesis. The shape and size of determinate organs are established by the behavior of localized, temporary meristems in expanding primordia. Intercalary, plate, marginal, and other transitory mitotic loci remain active for precise intervals and in programed order as the appendage takes form. Uneven versus uniform activity along expanding embryonic margins creates a lobed rather than an entire lamina. Should a basal intercalary meristem shut down quickly, a sessile instead of a petiolate appendage results. Regulator and modifier genes are probably largely responsible for leaf shape, and ubiquitous mitogenic and growth-influencing hormones, not form-specific morphogens, serve as their mediators. Existence of as many specific translocat-able agents as there are different or unusual laminar shapes stretches credulity and draws no support from the literature. Hall et al. (1987) used radioimmunoassays to identify cytokinin bases, ribosides, 0-glucosides, and nucleotides in xylem sap of two Amyema species and three Eucalyptus populations that support them. Closer relationships were found between the type and concentration of cytokinins in one mimetic host-parasite pair than in two form-distinct combinations of tree and mistletoe, but assignment of cause to occurrence of these common growth factors would be premature.
Calder (1983) makes a plausible case for dispersal as the primary advantage of host simulation. If a mistletoe resembles its support well enough to be imperceptible at a distance, birds must learn to key on the more conspicuous (larger) host-crown signal to locate pseudoberries. Such a mechanism ensures regular recruitment of uncolonized supports because appropriate frugivores, some with guts already containing seeds, must explore all poten-
tial food-bearing trees to discover whether or not fruiting mistletoes are located there. Of course, advantage diminishes as host range widens, but so it would if protective concealment were the purpose. The fact that mimetic mistletoes, especially those native to relatively humid forests, are not all host-specific is attributed to relaxed selection pressure associated with late Tertiary and Quaternary extermination of herbivorous marsupials (Barlow and Wiens 1977). Changes in phytophagous insect fauna, if any, would be inconsequential; these predators usually cue on food source by olfaction rather than by sight.
It might seem surprising that no one has apparently entertained the notion that host-mistletoe mimicry is not really mimicry at all but a manifestation of functional convergence. After all, in warm arid zones especially, leaf proportions normally track regional meteorological conditions because inappropriate foliar morphology could have especially dire consequences for water economy and heat dissipation. Predominant leaf shapes in fossil floras have long been known to be useful indicators of paleoclimate. But such a proposal is not found in the literature, and for good reason. Variation among co-occurring species - those in a single temperate forest, for instance
- attest to the absence of fine-tuned selection for uniform shapes at the local level. In addition, gas exchange patterns, by influencing energy budgets and placing constraints on the evolution of leaf form, differ substantially between trees and their aerial parasites, as noted above. Finally, homoplasy in response to similar climatic stresses fails to explain the variety of leaf forms shared by sympatric, mimetic host-parasite combinations involving models with widely disparate leaf architecture (e.g., Acacia, Casuarina, and Eucalyptus) and their mistletoes. Microclimates surely do not vary much among crowns of these trees. More study is needed to test the first and third
- and possibly additional — hypotheses concerning vegetative form in Australian mistletoes. Scattered reports (Atsatt 1983) of comparable mimicry in other parts of the world should also be verified.
What few data are available on the deleterious impact of mistletoes in natural forests come from western North America, where each year Arceuthob-ium reportedly reduces wood yield by 500 million cubic feet in the United States alone (Hawksworth 1983). Mistletoes are second only to fire as damaging agents in certain Mexican forests where 10-24% of trees are infected. Significant economic hardship is experienced in many parts of the world by
Table 6.3. Some important disease-inducing mistletoe genera, their hosts, and regions of serious crop losses
Dendropemon Elytranthe Oryctanthus Phthirusa
Viscum several, including eucalypts and acacias several trees, including citrus rubber, cashew cocoa rubber, cocoa, erythrinas, citrus, mango, coffee, avocado cocoa citrus conifers rubber, mango, avocado, cocoa acacias, eucalypts coffee, avocado, teak, erythrinas, citrus, cocoa, various forest trees rubber, fruit trees, deciduous trees, walnut, persimmon, fir, pine
Costa Rica, Brazil, El
Salvador Ghana, Nigeria India
North America, Mexico, China, India, Pakistan, Kenya, Mediterranean area, Middle East South America Hawaii, Australia Bolivia, Trinidad, Central America, Mexico, United States, West Indies
Europe, Australia, China, Asia, Africa
Source: After Knutson 1983.
plantation owners growing such crops as Citrus, fig, mango, rubber, coffee, and cocoa. Table 6.3 lists important disease-causing mistletoe genera in both Loranthaceae and Viscaceae and the regions of greatest distress. Numerous xylem-tapping species are involved, but others forming diffuse endophytes are particularly destructive.
Symptoms of disease vary with the host-parasite combination. The literature contains frequent statements about visible damage, and there are occasional references to trees being killed by heavy infestations. Injury caused by Loranthus langsome can be so severe that heavily infected Lan-sium domesticum (mahogany) lives only seven to eight years. Fruit yield of English apple trees has been depressed 40% by Viscum album (Hawksworth 1983). Common reactions to mistletoe attack are localized hypertrophy at the attachment site, dieback of infected branches, deformation caused by death of the leader, reduced overall vigor, and formation of witches'-brooms. (Brooming is the profuse production of abnormal host twigs and foliage.)
Mistletoes are potentially more destructive than free-living epiphytes, no doubt owing to their direct use of host resources and possibly because of physiological consequences of the tap-in. Moisture stress is heightened during drought, damaging infected more than uninfected trees. Sometimes much host foliage is shaded by heavy infestations. There is also evidence of considerable hormonal involvement: Cytokinin levels in Arceuthobium douglasii have been found to exceed those in adjacent host tissue as much as tenfold. Brooming similar to that induced by Phoradendron and dwarf mistletoes is caused by growth factors where certain microbes are the infectious agents. Likewise, hypertrophy adjacent to the endophyte is undoubtedly under hormonal control, but whether the agents involved originate in tree or parasite is not known. Hormonal imbalance can be systemic and contribute to the general decline of diseased hosts; stressed foliage has been shown to contain reduced cytokinin and elevated abscisic acid activity, but these conditions may be more directly related to water deficits caused by the parasite's transpiration than to direct metabolic intervention. Like other obligate parasites, mistletoes lose, fitness if too virulent, and indeed many infections seem to create no serious problems for supports. Further inquiry might reveal that damage is greatest on occasional or unnatural (exotic) hosts or those trees already weakened by other factors. At this point, pathogenicity seems unrelated to host specificity.
Several means have been tried to control mistletoes in orchard crops and shade trees; In Europe, 2,4-dichlorophenoxyacetic acid (2,4-D), 2,4,5-tri-chlorophenoxyacetic acid (2,4,5-T), 4-(2-methyl-4-chlorophenoxy)butyric acid (2,4-MCPB), and dichloroethane killed Viscum album shoots on Abies with little host damage (Hawksworth 1983). Such less expensive agents as copper sulfate and diesel oil have been used against Dendrophthoe and other Indian mistletoes. Unfortunately, chemical control of Arceuthobium has not been successful. Increasingly, statutory restrictions will probably deter widespread application of broad-spectrum herbicides. Attempts to eliminate avian dispersers would be unwise. Acceptable alternatives might be based on insect predators or fungal pathogens, of which several are known. In the meantime, the ancient practice of pruning infected branches will probably have to suffice in most instances.
Was this article helpful?