EVALUATION OF INTRODUCTIONS

The unbiased evaluation of introductions is virtually impossible because of the many different viewpoints that are brought to bear on natural ecosystems that are at the same time used for the production of food or some other value. At one end of the spectrum the strict conservationist is reluctant to see any change in 'status quo'. In this context aquatic ecosystems are particularly exposed. The innumerable changes in terrestrial ecosystems caused by the introduction of food crops have already been absorbed and have established the functional equilibria we equate with the landscape today. Introductions into aquatic environments, on the other hand are current, less integrated into the ecosystem and consequently more evident. At the other extreme the purely commercially oriented would wish to introduce species and modify ecosystems to maximize profit regardless of the damage to the ecosystem. These opposed views tend to coincide with different economic circumstances. Post-industrial societies, located mainly in the temperate zones, tend to place emphasis on the protection of the environment and thus approve only those species which produce minimum disturbance in the quality of the natural system. Conversely, rural societies of the tropics may be willing to tolerate quite considerable environmental change in favour of an immediate solution to food deficiencies. Because the units of measurement are so disparate it is inadvisable to use terminology which imply success or failure to an individual introduction but rather to consider a type of balance sheet of costs and benefits. Here one can only be guided by local perceptions as expressed in correspondence. Unfortunately this is highly subjective and does not permit comparability between assessments of introductions made by different countries. However, in certain species a clear consensus appears and in these cases some more definitive judgement can be made.

Examination of the introductions and comments reported in this document shows that six patterns of behaviour follow an introduction.

(i) The introduced species never becomes established, disappears without trace or its fate is listed as unknown. This happened in 298 (22.01%) of the introductions reported but this is probably a considerable underestimate due to failure to fully report such occurrences. Apart from the numerous cases where species were introduced into inappropriate climates and habitats, or the introduction was made in insufficient numbers as to ensure establishment, the reasons for the inability of a species to establish itself in the host waters remain obscure.

(ii) The introduced species does not breed under natural conditions and is maintained artificially or by continuous import. 246 (18.17%) of the introductions reported are of this type which is particularly common with species having commercial value for aquaculture, such as rainbow trout or the Chinese carps, and for species such as the grass carp which are used for controlling undesirable organisms. Ostensibly species which are maintained by artificial reproduction are easier to control and to eradicate from the natural waters when their purpose is served. However, continuous import heightens the risks of introduction of disease and grass carp have frequently been accused of environmental degradation. Continued failure to breed in the wild can not be assumed either as self breeding stocks of Ctenopharyngodon idella have appeared in both the Mississippi and Danube river systems.

(iii) The introduced species become locally established in an unusual habitat. Thirteen (0.96%) cases of colonization of thermally polluted waters in the temperate zone were reported from Canada, where at least six aquarium species occupied hot springs in Banff National Park, Alberta for a number of years; and from Australia, Hungary, Belgium, Netherlands and the United Kingdom, where a few species have been maintained in warm water discharges from factories or power stations. Such populations occupy areas inaccessible to native temperate fishes and persist so long as the favourable conditions are maintained but disappear when the thermal source is removed. Forty nine introductions were recorded where temperate species, usually the salmonids - Salmo gairdneri, Salmo trutta and Salvelinus fontinalis - are maintaining themselves in cold, high altitude waters at tropical latitudes. In these cases climatic conditions are more stable and the introductions can be regarded as permanent additions to zones whose native faunas are generally impoverished.

(iv) The introduced species establishes and maintains itself either as widespread populations at very low densities, or as isolated populations in individual water bodies or aquaculture installations. With 440 (32.50%) recorded instances this is the most common outcome of introductions, and, although it is improbable that a new element can be inserted into a fish community without its affecting native populations to some extent, the instances in this category have had effects that are so slight as to occasion no positive or negative comment.

(v) The introduced species establishes itself and may even increase rapidly in abundance to become a dominant element of the population. Later the species disappears or declines in abundance. The decline may occur naturally or may be deliberately eradicated. This type of behaviour was recorded in 36 (2.66%) instances. Of these eight were deliberately eradicated, five were tropical species established at the limits of their thermal tolerance and which died during particularly winters and one was eliminated by pollution. Other species established viable but low level populations which were unable to compete on a sustained basis and died out. A few species, notably Lepomis cyanellus, Oreochromis mossaabicus and O. niloticus showed the typical 'boom and bust' phenomenon. To a certain extent this might be expected of any species transported into waters where its preferred food is under utilized by existing species and the resulting population explosion is later brought into equilibrium with the available resources. However, it is difficult to explain why a once successful species disappears completely as has happened on several occasions.

(vi) The introduced species becomes a significant or dominant element of the host fauna as has happened in 321 (23.71%) records. Of these only 89 (6.57%) had sufficient impact on the local environment or fish stocks to cause serious concern. The other 232 (17.13%) were judged either neutral or even beneficial in their effect.

Several authors have described the ecology of introductions and have examined the possible reasons why species do or don't become established in any particular area (Stauffer, 1984; Taylor et al., 1984). Taylor et al. concluded that establishment of exotic fishes was most likely in

(i) aquatic habitats with mild year round temperatures;
(ii) disturbed or altered habitats, and
(iii) communities of low diversity.

While the present data are not of high quality and are possibly biased in favour of reporting species that have become established, it is the most extensive body of information on introductions on a global scale and should yield some information to clarify these ideas. There seems little doubt, for instance that the contention that introductions succeed in zones where species diversity has been kept low relative to similar areas elsewhere is correct. Apart from aberrations such as thermally polluted waters in the temperate zone two groups of habitat are particularly good examples of this: islands whose freshwaters are inhabited at low levels by secondary colonizers and highland lakes and rivers in the tropics. In both the establishment rate of introductions has been high; in tropical Islands 200 established out of 273 recorded introductions in Islands (73.3%) and in high altitude tropical waters the establishment rate of salmonids was 49 established out of 67 introduced (73.1%). Similar trends have also been noted in the impoverished communities of the Western drainage of the United States (Moyle, 1986). The statement that the likelihood of introduced species establishing themselves in disturbed habitats is less easy to demonstrate. There seems no doubt, however, that the particular success of common carp is due to environmental changes brought about by damming and that other introductions have been particularly successful in colonizing new reservoirs. However, this situation is very similar to that where fish communities are impoverished because in many cases the original inhabitants are unable to adapt to the changing regimes and no native species can colonize the altered system. In some circumstances an introduced species may establish itself at low density in a stream only to expand in numbers when conditions are changed by impoundment. The history of introductions in the United States, described by Moyle (1986), illustrates this interplay between introduced species and environmental changes. In this region, as in much of the rest of the world, species introductions have been made contemporaneously with increases in major environmental stresses. It is thus often uncertain whether the declines in local species and increases in abundance of introduced species are traceable to the introduction, to the environmental changes, or to both.

It has also been suggested that there are basic differences between the modes of colonization of tropical and temperate waters with 'r' selected species being more successful in temperate regions and 'K' selected species being more successful in the tropics (Stauffer, 1984). This idea originates from the supposition that complex tropical fish communities are in some way more 'saturated' than those of the temperate zone and that the probability of successful colonization by a new species is higher in simple communities with low diversity. A concept of species saturation has been developed in support of this whereby it is implied that any given environment can support only a limited number of species. This would suppose that the proportion of species becoming established in the complex communities of the tropics would be less than that in the temperate zone.

508 introductions were recorded from tropical and sub-tropical countries of which 318 (62.6%) were established, as against 573 introductions made into warm and cold temperate countries of which 324 (56.5%) became established. As to the relative effectiveness of different reproductive strategies in determining whether or not any particular species becomes established, it is difficult to neatly categorize species under this concept especially as some apparent 'K' strategists may switch to 'r' strategies when introduced into a novel environment. A broad classification, where species having one major spawning per year of many small eggs are grouped under 'r' and species showing some degree of parental care (nest builders, mouth brooders, live bearers etc.) are grouped under 'K' selection. Is presented in Table 14.

Table 14
Comparison between numbers of introductions of 'r' and 'K' selected species into temperate and tropical countries. Figures are presented as No of introductions/No of cases of establishment (%)

Category	Temperate	Tropical
'r'	356/235 (66.0)	84/44 (52.4)
'K'	177/123 (69.5)	406/337 (83.0)

This table excludes data on Chinese and common carps, which have been introduced into both climatic zones and are usually maintained artificially, and anadromous and catadromous species.

Table 15 shows the results of chi-square tests comparing establishment success in various ecological zones and of 'r' and 'K' selected species.

Table 15
Chi-square values derived from comparisons of establishment success

Type of introduction	Chi-square value	1%	5%
		(6.63)	(3.34)
Total tropical v total temperate	4.34	+	-
Island v high altitude tropical	0.03	+	+
Tropical: high altitude v low altitude	3.15	+	+
Islands v total tropical	9.50	-	-
Temperate: 'r' selected v 'K' selected	0.50	+	+
Tropical: 'r' selected v 'K' selected	35.90	-	-
'r' selected: tropical v temperate	6.04	+	-
'K' selected: tropical v temperate	12.72	-	-

Despite the poor quality and possible incompleteness of the data it is possible to conclude that success of establishment of species in tropical waters is very similar with introductions into the tropics being slightly more successful in terms of forming breeding populations that those made into temperate waters. The success of introductions made into ecosystems with impoverished native faunas (cold, high altitude streams in the tropics or islands) was very similar. Those made into islands were significantly more successful than those introduced into tropical mainlands, whereas introductions into tropical highlands were only significantly more successful than those made into tropical lowlands at the 5% level.

In temperate zones there was little difference between the success of 'r' and 'K' selected species but in the tropics 'K' selected species did significantly better. Because of the different nature of tropical and temperate communities 69.6% of tropical introductions were made with 'K' selected species as opposed to only 19.1% of temperate ones.

It is not possible to draw any conclusions from this data set as to which factors, if any, influence the success of introduced species in establishing themselves. Indeed the conclusion reached by the 1986 Royal Society Symposium on Biological invasions that there are no sure-fire criteria for predicting the fate of introduced species seem equally applicable here. On the other hand it is usually possible to identify the reasons for the successful establishment of a species after the event.

Evaluation of the desirability of introductions of individual species is highly subjective but in many cases the replies to the questionnaires have given an indication of how some species have been viewed which may serve to guide others who are considering similar introductions. The present listing is similar to Chat made by Welcomme (1984) but has been modified in some particulars as more information has become available. The majority of species have had impacts so slight that they occasion no comment. The rest can be classified into three groups.

The majority of species that are definitely described as pests (Table 16) are species, generally 'r' strategists, which respond to their new environment by forming dense, stunted populations (i). This behaviour makes them useless for anything but forage for other fishes and caused a disproportionate amount of environmental nuisance. Other species in this category are burrowers such as Eriocheir sinensis or Procambarus clarkii which cause costly and even dangerous damage to river and pond banks. Oreochromis mossambicus is also labelled as a pest despite its value to fisheries in Brazil and Sri Lanka because of the degree of disapproval it attracts. Its poor flavour, tendency to stunt and to eliminate competitor species are seen as overly disadvantageous especially as other tilapias give better results. Lates niloticus is included as a pest because of its drastic environmental and socio-economic effects and because it will probably lead to a diminution in the overall productivity of the water bodies to which it was introduced.

Table 16
Species widely classified as pests where introduced

Alburnus alburnus §	Lepomis cyanellus §
Blicca bjoerkna §	Lepomis gibbosus §
Clarias batrachus §	Lepomis macrochirus §
Eriocheir sinensis	Opsariichthys uncirostris §
Hemibarbus maculatus §	Orconectes limosus
Hemiculter eigenmanni §	Oreochromis mossambicus §
Hemiculter leucisculus §	Percottus glehni §
Hypseleotris swinhonis §	Pseudogobio rivulatus §
Ictalurus melas §	Pseudorasbora parva §
Ictalurus nebulosus §	Rhinogobius similis §
Lates niloticus	Rutilus rutilus §
Lepomis auritus §	Scardinius erythrophthalmus §
Serassalmus humeralis

Table 17
Species whose introduction is viewed with mixed feelings

Carassius auratus §	Oncorhynchus gorbuscha
Cichla ocellaris *	Oreochromis urolepis §
Cichlasoma managuense *	Oreochromis niloticus §
Ctenopharyngodon idella	Perca fluviatilis §
Cyprinus carpio	Poecilia latipinna §
Esox lucius *	Poecilia reticulata §
Gambusia affinis §	Procambarus clarkii
Micropterus salmoides *	Salmo gairdneri *
Odontesthes bonariensis *	Salmo trutta *
Oncorhynchus keta	Stizostedion lucioperca *
Oncorhynchus kisutch	Tilapia rendalli §
Tilapia zillii §

Another group of species are controversial in that they have created ecological problems of various kinds but are, at the same time, perceived as being extremely useful (Table 17). There is a high proportion of predatory species (*) which have been implicated in eradication of endemic fishes as well as species which form dense, stunted populations (§). However, nearly all have made a contribution to food production, control of unwanted organisms or sport fisheries which is perceived as greater than any damage caused.

Table 18
Species generally viewed as beneficial where introduced

Anguilla anguilla	Ictalurus punctatus
Anguilla japonica	Labeo rohita
Aristichthys nobilis	Limnothrissa miodon
Astatoreochromis alluaudi	Macrobrachium rosenbergii §
Barbus javanicus	Mylopharyngodon aetiops
Carassius auratus §	Mylopharyngodon piceus
Catla catla	Oncorhynchus tshawytscha
Colossoma macropomum	Oreochromis aureus §
Coregonus lavaretus	Oreochromis esculentus
Dorosoma petenense §	Oreochromis niloticus §
Heterotis niloticus	Pacifastacus leniusculus
Hypophthalmichthys molitrix	Trichogaster pectoralis

Other than disease related problems these species have not been identified with any major environmental impact. Some show a tendency to stunt (§) but this is viewed as a minor problem.

Of the 237 species forming the subject of the 1354 introductions examined here only 71 made sufficient environmental impact to warrant mention by the correspondents. Of these 25 (10.7%) were judged pests and thus wholly negative in their impact. The remaining 47 (19.8%) were considered beneficial despite varying degrees of negative effect. The majority (69.5%) of introduced species, therefore were judged innocuous.

It also can be concluded that, apart from disease related effects which may be independent, serious impacts on the environment can be anticipated from two main classes of species:

(i) those whose reproductive pattern enables them to fora stunted populations and

(ii) major predators, especially where these are introduced into communities which lack ichthyophages.

The evaluation of international introductions of inland aquatic organisms can, perhaps, best be summarized in a statement made by the 4th Session of Committee for Continental Fisheries of Latin America (COPESCAL) Working Party on Fishery Resources:

"The introduction of new species into areas from which they were absent constitutes a valuable tool for the management of inland wafers for capture and culture fisheries. However, such introductions involve considerable risks to the health and ecological integrity of existing populations of fish through transfer of diseases, degradation of the ecosystem and the formation of undesirable hybrids. It is therefore necessary to examine proposals for the introduction of further species with extreme care in order to minimize the risks of unforeseen consequences".