☰  Menu

The Flowering Plants of Hawaii (beta)

Part 50 Violaceae - Viscaceae - Vitaceae - Zygophyllaceae


Isodendrion pyrifolium
Isodendrion pyrifolium. Gerry Carr photo.

Although the Violaceae are a modestly sized family, 700 species in 24 genera, and enjoy a cosmopolitan distribution, by far the most well known member is the genus Viola with well over half of the family's species. The genus is a very common contributor to early spring color in most of temperate North America as well as Europe. The Hawaiian Islands are home to members of two genera, the endemic Isodendrion, and Viola. Isodendrion, aupaka in Hawaiian, consists of three rare species, and one (I. pyrifolium) that has not been collected since 1870 and is presumed extinct. Our representative of the genus here is Isodendrion longifolium (see image), a native of deep valleys on O`ahu and Kaua`i.

Hawaiian Viola consists of seven species according to the Manual, nine according to a recent paper on evolutionary relationships within the group (Havran et al. 2009). Some of them are considered rare, or very rare in the case of V. helenae, which is only known from a small population on Kaua`i. Others are also highly localized; V. wailenalense occurs only in the vicinity of the Alaka`i Swamp (Kaua`i), for example. Visits to the Pëpë`öpae Bog on Moloka`i over the years have been rewarded by finding V. maviensis (see image) in flower. This species also occurs in the Kohala Mountains on the Big Island and on Maui, as its specific epithet states. A second species, V. chamissoniana (see image), known as pämakani or `olopü in Hawaiian, occurs in the Pëpë`öpae Bog area as well as on Kaua`i, O`ahu, and Maui. These two species hybridize in the vicinity of the Pëpë`öpae Bog, but I have not seen a sufficient number of individuals to make any comment on intermediate forms.

Viola maviensis
Viola maviensis
Viola chamissoniana
Viola chamissoniana

A DNA-based examination of most of the Hawaiian members of Viola plus 40 species representing areas from which progenitors of the island's species might have come was undertaken by H. E. Ballard of Ohio University and K. J. Sytsma of the University of Wisconsin (2000). These workers sought answers to five questions: (1) Have all Hawaiian violets arisen from a single colonization? (2) Since woodiness is generally considered to be an ancestral characteristic, at least in continental settings, do the island species, which are woody, represent an ancestral lineage within the genus? (3) Are the island species derived from a group of Central American and Andean ancestors as suggested by some earlier workers? (4) Is their woodiness really an ancestral feature, or could it have been secondarily derived as has happened in other unrelated groups of island plants? (5) When did colonization occur?

The DNA data provided unequivocal answers to all five questions. The data clearly showed that Hawaiian violets are the result of a single colonization, with that event occurring no more than 3.7 million years ago, with colonization first on Kaua`i and subsequent radiation to O`ahu. Migration to Maui, Moloka`i, and Läna`i would have occurred later still, likely at a time when all three of those islands were conjoined to form Maui Nui. Migration to the newest island, Hawai`i, would have been the most recent event. Ni`ihau and Kaho`olawe probably never had habitats that were high enough or wet enough to accommodate violets. Data from the more recent study (Havran et al., 2009) resulted in a different scenario. It is now thought that colonization first occurred on Maui Nui no earlier than 1.2 to 2,000,000 years ago. Interisland colonization of O`ahu and Kaua`i occurred more recently, with migration to the Big Island still more recently.

These data also placed the island violets among the most advanced members of the genus, rather than as extant members of an ancient lineage. Woodiness, then, is a secondarily derived feature of these plants. Many members of the Hawaiian flora, the silversword alliance among others, which are known to have herbaceous ancestors on the mainland, are woody shrubs or trees. Woodiness as a survival strategy has been discussed by Carlquist (1974, 1980).

In most DNA-based studies the search for ancestral relationships of island species have resulted in identifying a likely genus or, in the case of large genera, a group of genera, as the source of the original propagules. The outcome of the Hawaiian violet study is remarkable in that it revealed not just the geographic area of a likely ancestor, but identified a species as the closest relative of the island group, in this case V. langsdorfii. Viola langsdorffii occurs from Japan to the Queen Charlotte Islands (Haida Gwaii), British Columbia, in what is known as an amphi-Beringian distribution. This finding reduced the ancestral home significantly, but we are still dealing with the entire northern Pacific rim. An examination of the chromosome numbers of V. langsdorffii allowed these workers to limit the ancestral range even more. Plants ranging from Japan to the Kamchatka Peninsula have been shown to be decaploids (n = ca. 50), those from Kamchatka to be hexaploids (n = ca. 30, and those from the Queen Charlottes to be dodecaploids (n = ca. 60). No data were available for plants from the American Arctic, but the numbers were assumed to be in the n = 30 – 50 range, which would presumably have a group of n = ca. 40 populations in the middle. Chromosome numbers for Hawaiian species of Viola have been recorded as n = 38, 40, and 41-43. Ballard and Sytsma suggest that the ancestral V. langsdorffii, the missing n = 40, will be found in the American Arctic. Dr. Ballard informed me in an e-mail conversation on 20 April 2009 that the missing n = 40 violet has still not been located. He did raise an interesting point, however: he suggested that the Hawaiian violet was formed in situ as a result of hybridization between two colonists. The mystery continues.

It is useful to emphasize the importance of the DNA data in establishing that the island species are the product of a single colonization event, that is to say that they constitute a monophyletic group. All share a 26-base pair deletion in the ITS sequence studied. Alterations of a DNA sequence of this magnitude would not be expected to occur randomly in unrelated species.

It seems hardly necessary to point out that an Arctic origin of the island violets was not anticipated. The Arctic connection with an element of the Hawaiian flora was a first. The authors commented on the obvious bias of workers who usually search for tropical, or at least warm temperate, origins of Pacific island plants. They made the additional point that there may be other Hawaiian plants with a northern connection, a reasonable view considering the very significant bird traffic between the two areas.


Korthalsella sp.
Korthalsella sp.
Korthalsella sp.
Korthalsella sp.
Korthalsella sp.
Korthalsella sp.

Viscaceae, the mistletoe family, are home to plants that would be familiar to most readers, but probably not by their formal botanical names, the North American genera Arceuthobium and Phoradendron. Both of these have species that are common parasites on, but not restricted to conifers. Phoradendron is the mistletoe of commerce in North America. An interesting addition to the collection of mistletoes in North America is Viscum album, the European mistletoe, which was introduced into California in about 1900 by Luther Burbank (Hawksworth and Wiens, 1993). This mistletoe is less choosey as to host plant, parasitizing hard and soft woods alike.

The Hawaiian contribution to Viscaceae is the genus Korthalsella otherwise known from such diverse places as northeastern Africa, Madagascar, the Mascarene Islands, Asia from the Himalayas to Japan, and south to Australia and New Zealand (Mabberley, p. 457). With this wide a range one might expect it to be a reasonably large genus; it is not; according to Mabberley, the genus consists of a mere eight species. But it has not always been so. In the latter years of the 19th century botanists recognized over 60 species in Korthalsella and two closely related genera (neither mentioned by Mabberley). The most recent formal treatments of the genus by B. H. Danser (1937, 1940; references in Molvray et al., 1999) resulted in a reduction of the number to 23, with 10 varieties. In her doctoral research (at Tulane University) Mia Molvray performed a morphometric examination of a wide collection of specimens and concluded that only eight definable species are supportable by her data. Authors of the Manual do not accept this strictly delimited view of the genus, stating in their introduction that there are about 20 species worldwide, with four of them endemic on the Hawaiian Islands along with two indigenous species. Species identification can be difficult owing to extensive morphological variation within the genus. Three illustrations are included here to demonstrate some of the variants (see images). The Hawaiian names for this group of species are hulumoa and kaumahana. Hulumoa is the more descriptive of the two, translating as chicken feathers (hulu is feather; moa is chicken).

Cissus nodosa
Cissus nodosa

DNA sequence data place Viscaceae within Santalaceae, thus including these parasitic and hemiparasitic plants in a single family.


The grape family, Vitaceae, has a single representative on the islands, Cissus nodosa, the grape ivy (see image). This species, native to Java, was brought to the Hawaiian Islands as a decorative plant and has now become naturalized on Kaua`i, O`ahu, and on the Big Island.


The last of the dicot families in our tour through the Hawaiian flora are the Zygophyllaceae, commonly referred to as the creosote bush family. That name comes from the most familiar member of the family, at least in North America, the creosote bush, Larrea divaricata subsp. tridentata (or just L. tridentata in some treatments). Creosote bush is an extremely common component of the vegetation of the southwestern deserts. A single bush in the Mohave Desert, reproducing by suckering clones, has been aged at 11,700 years.

Tribulus cistoides flower
Tribulus cistoides flower
Tribulus cistoides
Tribulus cistoides fruit.

The family's contribution to the Hawaiian flora consists of two species Tribulus terrestris, known as puncture vine or goat head; and T. cistoides (see images), known to Hawaiians as nohu or nohunohu. The former grows in disturbed habitats on Kaua`i, O`ahu, Maui and on the Big Island, the latter mostly in coastal settings. The photograph of nohu was taken on a windy headland on the northern coast of Moloka`i. I have also seen it growing in sand on the northern coast of Läna`i. This is an Old World species that has become a pantropical weed.

Preparations made from T. terrestris are reputed to be useful in addressing certain "male problems" and are available from a number of sources. Reports of their efficacy vary, including some that claim an increase in testosterone levels, and others that say that hormone levels are not affected. This disparity is not unusual in untested natural preparations. As in the case of many naturopathic preparations, care should be exercised in their use.

Literature cited…

Ballard, H. E., Jr. and K. J. Sytsma. 2000. Evolution and biogeography of the woody Hawaiian violets (Viola, Violaceae): Arctic origins, herbaceous ancestry and bird dispersal. Evolution 54: 1521-1532.

Carlquist, S. 1974. Island Biology. Columbia Univ. Press, New York.

Carlquist, S. 1980. Hawaii, a natural history. Pacific Tropical Botanical Garden, Lawa`i, HI.

Havran, J. C., K. J. Sytsma and H. E. Ballard, Jr. 2009. Evolutionary relationships, interisland biogeography, and molecular evolution in the Hawaiian violets (Viola: Violaceae) American Journal of Botany 96: 2087-2099.

Hawksworth, F. G. and D. Wiens. 1993. Viscaceae. Mistletoe Family. In The Jepson Manual. Higher Plants of California, pp. 1092-1097. J. C. Hickman (ed.) University of California Press, Berkeley, CA.

Molvray, M. 1997. A synopsis of Korthalsella (Viscaceae). Novon 7: 268-273.

Molvray, M., P. J. Kores and M. W. Chase. 1999. Phylogenetic relationships within Korthalsella (Viscaceae) based on nuclear ITS and plastid trnL-F sequence data. American Journal of Botany. 86: 249-260.

October 25, 2012

Table of Contents

Dicot Families:

Monocot Families:

Copyright 2016 Floridata.com LLC | Tallahassee, FL | Crestview Hills, KY