ALIEN FERNS IN HAWAI'I

Considerable attention has been paid to the advent of alien plants to Hawai'i and to their establishment and spread in the Islands. W. L. Wagner et al. (1990) listed 861 species of naturalized angiosperms in Hawai'i, representing 47% of flowering plants treated in their manual. Among significant publications dealing with the immigration and establishment of alien plants in Hawai'i are: a study by Smith (1985) of the impact of alien plants on the native Hawaiian biota; a 1989 paper by Loope and Mueller-Dombois (1989) examining the characteristics of islands subjected to biological invasions, paying particular attention to Hawai'i; and, most recently, a volume that includes 44 papers on alien invasions (Stone et al. 1992) . Studies have shown that the factors allowing for the successful establishment of large numbers of alien plants in the Hawaiian Islands are: the widespread disturbance of native habitat by humans, introduced animals, or natural causes; human activities, either deliberate or unintentional, that introduce large numbers of plant species; and the ideal growing conditions found in the Islands.

With very few exceptions, the recent attention paid to the problems presented by the naturalization of alien plants has not included pteridophytes. The Hawaiian fern flora is estimated to include about 170 native and 30 naturalized species (Wagner 1995) . The naturalized ferns represent fewer than 16% of all the pteridophytes, a figure significantly lower than that for the flowering plants. Since 1950, nine new immigrant ferns and fern allies have been found naturalized in Hawai'i. The invasion by alien pteridophytes has been gradual but steady, averaging about three new species per decade. Twenty five of the 30 naturalized species are frequently cultivated; only two of them are not known to be cultivated in the Hawaiiand Islands (Lindsaea ensifolia and Selaginella galeottii). Undoubtedly, Hawaiian gardens have provided the principal port-of-entry for the introduction of alien pteridophytes. Botanical gardens have, on more than one occasion, served as a source for naturalized ferns. A recent inventory of the cultivated ferns and fern allies in Hawai'i listed more than 260 species of alien cultivated pteridophytes, most of them in botanical gardens and arboreta (Imada et al., in prep.). These plants are well situated to supply additional alien species to the local flora.

It would be advantageous to be able to predict which alien species would escape from cultivation and spread in the Hawaiian flora; this would allow us to guard against these species and attempt to exclude them. However, data on growth behavior of ferns in other regions provides only limited information about their probable growth patterns in Hawai'i.

Recent floristic studies of two Pacific islands allow for comparisons with Hawai'i. G. Brownlie (1977) incuded only two naturalized fern species, Ceratopteris thalictroides and Salvinia auriculata, in his Fiji fern flora. In New Zealand, however, P. J. Brownsey and Dodsworth (1989) found 22 naturalized pteridophyte species, and three additional species have been found since then (Brownsey, pers. comm.). Only four species now naturalized in New Zealand are also established aliens in Hawai'i: Adiantum raddianum, Cyrtomium falcatum, Cheilanthes viridis, and Selaginella kraussiana. Adiantum hispidulum, naturalized in Hawai'i, is indigenous to New Zealand. Azolla filiculoides is listed by Brownsey as indigenous to New Zealand, but he points out that these plants are frequently referred to as A. rubra R. Brown or A. filiculoides var. rubra (R. Brown) Strasburger, which is different from the plants found in Hawai'i. Two ferns indigenous to Hawai'i, Adiantum capillus-veneris and Pteris cretica, have become naturalized in New Zealand.

Of the species naturalized in New Zealand and also in Hawai'i, only Adiantum raddianum has spread widely in the Hawaiian Islands and is crowding out the indigenous A. capillus-veneris. In contrast, in New Zealand, A. raddianum is naturalized only in a few places around Auckland, Taranaki, and Wellington (Brownsey 1981; Brownsey 1988) . The aggressive growth of Nephrolepis multiflora in Hawai'i might have been anticipated, but its early introduction and escape, and because of the uncertainty about its identity, and even whether it was indigenous or introduced, did not apparently arouse concern. The inpact of Angiopteris evecta and Cyathea cooperi on the Hawaiian native flora could problably not have been entirely anticipated; they are not serious problems elsewhere.

During the period between 1996 and 2001, three additional species have been found to be naturalized in Hawai'i. Marsilea crenata is a known weed in rice paddies in southeastern Asia. In Hawai'i it has been found only in one area, a small cultivated taro plot. There is no information about how it came to be introduced, but it has not been reported to be in cultivation. Its spread in Hawai'i is to be anticipated. Selaginella umbrosa is not reported to present special problems where it is naturalized. It is widely cultivated and its continued spread may be anticipated.

In 1996 I remarked that "Salvinia molesta D. S. Mitchell, a well known pest of waterways, is in cultivation in Hawai'i; perhaps its escape into the Hawaiian ecosystem should be expected." (Wilson 1996) In 1999 Salvinia molesta made its appearance as an escapee, already well established in different lakes on O'ahu. Manual eradication efforts were begun immediately and must be continued if its growth and spread is to be controlled. Biological control projects have not yet been reported, but these would involve the introduction of insects to the Islands with possible unpredictable consequences. Moran (1992) discussed the biology of this fern as well as the biological control methods that have been used in efforts to prevent its continued spread.

Because of the isolation of the Hawaiian ecosystem and the absence of many selective pressures during its evolutionary development, this ecosystem is particularly susceptible to invasion; its fragility is well documented (see Loope and Mueller-Dombois 1989) . Notwithstanding the difficulties in predicting the behavior of alien ferns when introduced into Hawai'i, when a challenge to the ecosystem is demonstrated, control and eradication efforts should be mounted. Often, however, these are initiated too late and inadequately to eliminate the threat. The cultivation of serious pest ferns, such as Angiopteris evecta and Cyathea cooperi, in nurseries, botanical gardens and home gardens, particularly in areas near native forests, invites their continued spread into the native ecosystem. Control progams must involve cooperative efforts not only of local, state, and federal agencies, but also of the general public. Considering the historical record in the prevention and control of the introduction and spread of the far more threatening seed plants, it is difficult to anticipate better results in limiting pteridophyte invasions, but an effort must be made.