The Flowering Plants of Hawaii
Part 21 Goodeniaceae - Gunneraceae - Haloragaceae
One of the most interesting groups of plants on the Hawaiian Islands belongs to the mainly Australian family Goodeniaceae. The type genus Goodenia consists of one species endemic to Java and 183 either endemic or largely restricted to Australia. The family is represented on the Hawaiian Islands by the genus Scaevola most of whose 100 or so species occur in Australia. Nine species (possibly 10, more on that later) of Scaevola occur on the Hawaiian Islands, eight (or nine) of which are endemic. The exceptional species, S. sericea (some authorities prefer the name S. taccada), is a coastal plant that occurs widely in the Pacific and Indian Oceans. This is the species most casual visitors to the islands are likely to see since it is frequently planted by resorts and hotels as a combination decorative hedge and beach stabilizer. Alternatively, visitors can find naupaka kahakai in its native habitat such as along the northern coast of Moloka`i as shown in the photograph. Naupaka is a general Hawaiian word for members of this genus, kahakai refers to the beach habitat. Another name for this plant, one that exemplifies this beautifully imaginative language, is huahekili which is interpreted as ‘thunder fruit’ in reference to the resemblance of the white fruits of naupaka to hail stones.
Other members of the genus with similar flowers are S. chamissoniana (see images) a species that occurs in the mountains of Lana`i, Moloka`i, Maui, and on the Big Island and hence known in Hawaiian as naupaka kuahiwi (kuahiwi is mountain or high hill); S. procera (see image) from Kaua`i and Moloka`i; S. kilaueae (image) from the Big Island (named after Kilauea volcano); and S. coriacea (see image) known as the dwarf naupaka. Dwarf naupaka, rare and protected in Nature, was photographed at the Maui Nui Botanical Garden. Other species characterized by unsymmetrical flowers (not pictured) similar to those above are S. gaudichaudiana (Kaua`i and O`ahu), S. gaudichaudii (all of the main islands), and S. mollis (Kaua`i, O`ahu and Moloka`i). All of these species occur at middle to higher elevations and are also referred to as naupaka kuahiwi.
The odd species is S. glabra (see images), an endemic species of wet forest on Kaua`i and O`ahu. I am most familiar with this species along the Pihea Trail on Kaua`i where it is very abundant. The Hawaiian name for this species is `ohe naupaka, where `ohe is the word for flute, pipe, hose, or tube. Recognition of this species as a naupaka despite its profoundly different flower structure compared to the flat face flowers of the other species is intriguing. If this is a pre-contact name—a name assigned prior to European botanical activities on the islands—it suggests a very sophisticated appreciation of technical floral structure by the indigenous people. The other possibility is that, having learned from haole botanists that the yellow tubular flowers really were from a Scaevola, the local name, naupaka, was modified appropriately with the Hawaiian descriptive word for tube (`ohe).
An examination of the origin, or origins, of Hawaiian species of Scaevola requires first a look at the distribution of the 130 species that comprise the genus. Ninety of them occur in Australia; two are widespread strand plants, S. sericea in the Pacific and Indian Oceans, and S. plumieri, in tropical Americas and Africa, with the ranges of the two overlapping in the Indian Ocean. Most of the rest of the species occur in the Pacific Basin; one species occurs on Socotra (Yemen) and one occurs on Cuba. The nine (possibly 10) species on the Hawaiian Islands constitute the largest assemblage on a single archipelago. Adding to the puzzle is the observation that S. glabra is a tetraploid whereas all other species, S. sericea and the island endemics, are diploids. The tubular flower, which resembles that of S. coccinea from New Caledonia, and the unusual chromosome number set S. glabra off from the others.
Opinions have varied as to the number of colonizations necessary to account for the Hawaiian species. F. R. Fosberg thought that only a single colonization occurred, while Paterson (1990, 1995), who studied morphological data, suggested that two or possibly three colonizations would have been necessary, although the larger estimate was based on a limited sampling of species. A possible fourth colonization was suggested when a single herbarium specimen, whose relationship with Scaevola had been overlooked, was recognized as the ninth island endemic. This species, S. hobdyi—named after Robert Hobdy, a well known Hawaiian botanist and forester—had been collected on Maui, but the species is now thought to be extinct. Its single bud and leaf features are unlike any others in the Hawaiian group which led to the suggestion of a fourth colonization. As we will soon see, this suggestion is not supported.
Using DNA sequence data from the nuclear ribosomal ITS region, Dianella Howarth and coworkers (2003; see also Howarth and Baum, 2005) studied the relationships among species representing the entire range of the genus. The data support the existence of three colonizations leading to three distinct sets of species on the archipelago, and point to the places of origin of colonizers: (1) Scaevola sericea, the beach naupaka, is most closely related to Polynesian members of the species; (2) the tubular flowered Scaevola glabra is most closely related to Australian species; and (3) the diploid species, including the newly discovered S. hobdyi, are probably most closely related to S. plumieri from the Americas. This relationship does not come as a surprise: in 1969 Sherwin Carlquist had pointed out the similarity between S. plumieri and the Hawaiian coastal S. coriacea. A very close relationship between S. hobdyi and S. kilaueae was indicated by their identical ITS DNA sequences. As the authors noted in their discussion, finding that the endemic island species are not closely related to Polynesian species was surprising. It should be noted that this is the first documented example of a multi-species Hawaiian genus that has involved multiple colonizations. Recall that in two large assemblages in the islands, the silverswords and the lobelioids, single colonizations—an idea which historically had been summarily rejected—resulted in very high levels of structural and ecological variation. But there is more to learn about naupaka.
Morphological variation among the diploid naupakas is not sufficient to generate a robust phylogeny (evolutionary history). The adaptive radiation of these species resulted in a set of organisms separated primarily by ecological specialization (differences in elevation, rainfall, substrate) with less energy expended in development of breeding barriers. Thus, species with some degree of morphological similarity, which would likely cross if opportunities existed, don’t cross because of their distinctive niches; in other words, they never meet one another in nature. This scenario is common in the islands where ecologically separated species will readily cross when brought together in a common garden environment. Thus, morphological differences among the diploid species of Scaevola are not sufficiently distinct to construct a robust phylogeny. Overlap in some features between species pairs, in fact, led Carlquist (1974, 1980) to hypothesize that S. procera was a hybrid between S. gaudichaudiana and S. mollis because these two species are known to hybridize whenever their ranges overlap. The recent study by Howarth and Baum (2005), however, revealed instead that S. procera is a hybrid product from S. mollis crossing with S. gaudichaudii.
The Howarth and Baum study was an attempt to find genes that would give a more reliable phylogeny of the diploid species. In addition to the ITS data previously gathered, these workers added sequence information from three additional nuclear genes: leafy (LFY), nitrate reductase (NIA), and glyceraldehyde 3-phosphate dehydrogenase (G3PDH). Analysis of the combined data sets yielded a more detailed picture of relationships among the seven species. Two clades emerged, the first consisting of the two dryland species, S. coriacea and S. gaudichaudii; with the second consisting of three species from wetter habitats, S. gaudichaudiana, S. mollis, and S. chamissoniana. The second clade was split further with S. gaudichaudiana and S. chamissoniana emerging as sister species; with that pair being sister to S. mollis. Note: the term clade is used to identify a group of biological taxa (species in the present case) that includes all descendants of one common ancestor.
The remaining two species, S. procera and S. kilaueae did not emerge cleanly from the analysis suggesting that they are products of hybridization. The origin of one of these, S. procera, was mentioned above. The other odd species out in the DNA analysis was S. kilaueae, whose hybrid origin from S. coriacea and S. chamissoniana was strongly supported. Scaevola kilaueae occurs on lava flows on the youngest island (Hawai`i) and is thought to be of very recent origin (recent within the geological time scale). Support for a recent origin comes from identical or nearly identical sequences in three of the genes studied. Stated another way, too little time has passed for the accumulation of significant, or any, changes in these particular gene sequences. Both putative parents, S. chamissoniana and S. coriacea, occur, or occurred, on the island, at the requisite time. Scaevola chamissoniana occurs on the Big Island, while S. coriacea no longer exists on the island owing to habitat loss related to human activities.
For our final comments on Scaevola we leave hard science and enter the realm of Hawaiian legend where we meet a beautiful (what else?) Hawaiian princess named Naupaka and her ill fated lover Kaui. Since he was a mere commoner he had no future with the princess owing to the kapu (taboo) on inter-class relationships. A kahuna (priest) at a local heiau (temple) suggested that they pray, although there was nothing he could do for them. Recognizing that their situation was hopeless, Naupaka took a flower from behind her ear, tore it in half, gave one half to Kaui and sent him to live on the coast, while she remained in the mountains. Naupaka plants, seeing how sad the lovers were, began to bloom in half flowers, as they continue to do today. Should the lovers ever reunite, the story sometimes goes, the half flowers will once again be united. Whether the reunited flower would still qualify morphologically as Goodeniaceae remains to be seen.
Gunneraceae consist of the single genus Gunnera with 40 species distributed primarily throughout the Southern Hemisphere, with two on the Hawaiian Islands, G. kauaiensis, restricted to Kaua`i, as its name suggests, and G. petaloidea, which occurs on O`ahu, Maui, eastern Moloka`i, and in the Kohala Mountains on the Big Island. Both species live in very wet, steep habitats. Some of the species feature the largest leaves of any dicot often attaining diameters of up to three meters (by contrast, other Gunnera species have very small leaves). The plant in the photo is a specimen planted by Vancouver Parks; my attempt to find a Hawaiian specimen on my own was highly unsuccessful. I was assured that I could find specimens of G. petaloidea on Moloka`i in a gully in the Kamakou Preserve accessible from the Pëpë`öpae Trail, a trail with which I was familiar from earlier visits. The gully, which lies beyond the alpine bog on the way to the Pelekunu Valley overlook, was easy enough to find, but the challenge was getting from the trail to the plants. Running water, deep mud—overtopping my boots immediately—no reliable hand holds on the slippery banks, and the vegetation provided nothing in the way of support. In short, it was not a good place to be. Skirting the gully was an option that we soon rejected because of the slippery mud banks and treacherous footing in general. And, one must ask, how much environmental destruction is a photograph worth? At least it wasn’t raining, so the trip wasn’t a complete loss. In fact, the day was a very successful one considering that we found three more island endemics, two species of violet and an orchid! We’ll meet them a little later.
Much more successful in getting close to a Hawaiian Gunnera was Ken Marr (who we met above). One of his collecting trips took him to the Blue Hole, an area that lies at the base of the steep eastern face of Mt. Wai`ale`ale, Kauai’s second tallest peak at 5,148’ (ca. 1,570 m) elevation ( Mt. Kawakini is 95’ taller, ca. 29 m). The waterfalls that grace the cliff face are fed by the rain that falls on the peaks above, reputedly, the wettest spot on earth. Growing conditions are perfect for plants that like it hot and damp, among which is Gunnera kauaiensis, `ape`ape or hähä on Kaua`i. These massive plants, with leaves that can attain widths of six feet, can be seen in the middle of the photograph. The equally impressive flowering structure can be seen in the next photo.
The question as to the origin of the islands’ Gunnera species has been addressed by Livia Wanntorp (2002, 2003) and her associates at the University of Stockholm. Species occur in East Africa, Madagascar, Malaysia, New Guinea, Tasmania, New Zealand, Mexico and Central America, South America (both eastern and western parts), the Falkland Islands, the Juan Fernandez Islands (Chile), and in the Hawaiian Islands. Initially, then, there are several possible sources for the ancestors of the island species. Morphological and other features place the Hawaiian species in a large group whose range stretches south from Mexico to southern South America (technically, section Panke). Application of sequence data from the nuclear ribosomal DNA ITS region and two chloroplast genes, rbcL and the rps16 intron, revealed that the two island species form a well supported group (clade in formal terms) and that that group is sister to, that is to say, shares a common ancestor with, the other members of section Panke. A specific candidate for most likely ancestor is not possible on the basis of the information available.
Although almost any plant could, in theory at least, escape from cultivation and become an aggressive pest. The reason for introducing the idea of aggressive aliens here is that I learned only recently that a species of Gunnera has become a problem on the west coast of Vancouver Island, an area of significant tourist value in British Columbia. The situation is reminiscent of other decorative species that behaved themselves well for a period of time—often years—and then erupted spreading rapidly into, usually, disturbed sites nearby. The local situation involved four plants each of G. manicata and G. tinctoria planted in the Tofino Botanical Garden (Tofino, B. C. coordinates: 49.05N, 125.51W) in 1998. A few seedlings were sighted two years later some few hundred meters from the parents. In 2005 three young seedlings of G. manicata were planted about 40 meters from the edge of the highway well buffered by other vegetation. In September 2008, 250 seedlings, measuring as large as 45 cm in width, were removed from the area in which the 2005 seedlings had been planted. One year later, in September 2009, there were ten well-established plants growing in a drainage ditch across the highway. As the author (Osborne, 2009) said in her report, “Something in the past two years has been just right for Gunnera” (emphasis hers). She also discussed the rampant nature of G. tinctoria in New Zealand and on the coast of western Ireland. In New Zealand, at least, it is now illegal to display, plant, or propagate gunneras of any kind knowingly.
Two naturalized species within Haloragaceae occur on the islands, Gonocarpus chinensis in wet forest in the Hawai`i Volcano National Park; and Myriophyllum aquaticum (see image), the water milfoil, in taro paddies where there is permanent water. One might expect to find this plant in the extensive taro fields in the Hanalei Valley in northern Kaua`i. Water milfoil is a major menace in fresh water lakes in North America where it is readily moved from lake to lake by means of whole plants or plant parts adhering to water craft.