Increasing experimental approaches in stream trout research – 1987–2006
Professor Emeritus Thomas G. Northcote. Current address: 10193 Morrison Close, Summerland, B.C. Canada VOH 1Z7.
J. Lobón-Cerviá, National Museum of Natural Sciences (CSIC). C/José Gutiérrez Abascal, 2. MADRID 28006-spain.
Abstract – This review of stream trout research literature for the 1987–2006 period covered >1300 papers dealing with 22 relevant topics, when compared with <400 papers on 18 topics in the previous one (1967–1986). The percentage of experimental approaches here quantified for both research reviews was 18% in the 1967–1976 period, increased to 21% in 1977–1986, to 39% in 1987–1996, and up to 43% in 1997–2006. Particular journals in the recent two decadal period published high percentages of experimental papers (The Journal of Animal Ecology and Canadian Journal of Fisheries and Aquatic Sciences; 62.9% and 62.4%, respectively), others, intermediate percentages (Nordic Journal of Fisheries Research, Transactions of the American Fisheries Society, North American Journal of Fisheries Management and Ecology of Freshwater Fish: 44.1%, 42.4%, 40.5%, and 35.6%, respectively); the remainder covered ≤33%. Research papers on stream ‘trouts’ published over the last two decades were classified into 22 subject areas for nine major journals separately and combined for 55 other journals. Subject areas dealing with the biology and ecology of trout redds, egg development, alevin emergence and onto fry and parr were the most heavily reported in nine major journals, as well as in all other journals combined. Habitat characteristics and cover use by trout were the next. Atlantic salmon and brown trout were the species with highest publication coverage in nearly all subject areas, with low coverage for all salvelinid (charr) species except brook charr. Research on Atlantic salmon in the UK used experimental approaches in nearly 60% of publications and in mid to high 30% for those on Atlantic salmon and brown trout in Scandinavian countries. Consideration is given to future research needs for stream trouts.
Why ‘trouts’ in a review of stream research?
Many may wonder why a review of stream research should focus so dominantly on the group of ‘trouts’ considered here, or for that matter why on fish at all, rather than more basic aspects of stream functioning such as their physical features (stream order, discharge variability, temperature and so on), their chemical characteristics and nutrient supply – for the later see Stockner & Ashley (2003), or their biotic structure (periphyton, benthic and drift invertebrates).
To limnologists having taken a ‘top–down’ view on the role that fish may play in the structure and functioning of freshwater ecosystems (Northcote 1988a), this present study focus on fish in streams came as second nature, but perhaps not necessarily so to many others, especially those used to bottom–up approaches in stream ecology. The emerging role for fish and riparian vegetation in structuring stream and forest food webs has been experimentally verified by Baxter et al. (2004), and examined earlier by others (Wallace et al. 1997; Nakano et al. 1999; Nakano & Murakami 2001; Kawaguchi et al. 2003).
There are broader and longer term reasons for taking this focus on fish, and in particular on many of the salmonid fishes in streams that are included here. In the first place there is a historical sequence to follow that focused on salmonids in streams, that in 1968 (Northcote 1969), 1987 (Lobón-Cerviá 1988) and the most recent one held in Spain (Ecology of Stream Fish II. 2007. Abstracts. León, Spain).
Trouts may use streams to complete their life cycle or as essential reproductive sites; and their anadromous forms and counterparts transport back to streams key nutrients gained elsewhere which are critical for a broad set of in-stream and riparian vegetation as well as associated terrestrial animals (Stockner 2003). In turn, trouts are used as food and sport for humans. In western North America, First Nation peoples at least from the 1700s onto the 1850s, and probably long before then, made heavy use of salmonids in streams and rivers for a major part of their food supply. Such salmonids and a few other species probably accounted for up to 75% of their annual diet in fresh, smoked or dried form. Northcote (1972, 1973) examined this for the Kootenay Lake watershed, as did Nisbet (1994) for other parts of the upper and middle Columbia River system for supplemental fish use by natives, early fur traders, and explorers from the late 1700s to at least the mid-1800s.
For the aristocracy of many European areas, Atlantic salmon in particular were a very desirable and sought after food source, within many flowing waters under their watchful jurisdiction. From the 1800s on Atlantic salmon use as food became more general. Angling and poaching in rivers and streams greatly increased, the former to the point where strong regulation of catches (seasons, numbers and methods) had to be invoked, a sequence followed with some differences in North America and other parts of the world where salmonids occurred or had been introduced.
As human populations began rapid increase throughout North America and Europe, so did their many types of intervention into streams and river functioning that supported salmonid populations. A brief review of this sequence for the Kootenay River and Lake system in British Columbia is given by Northcote (2006) and Ashley (2006), Hartman et al. (2006), and other chapters in the recent book examining the possibilities (not all that good) for wild Pacific salmon survival and restoration to the Pacific Northwest and California by the year 2100 (Lackey et al. 2006). Of course there are some success stories for stream restoration and management, but others only short term, and still others ineffective or worse. Here is not the place to go into this complex field, other than to note that the ‘trouts’ covered here have played major roles in both successes and failures. Reviews of native trout biology and management are given in Behnke (1992) for western North America, for brown trout in Europe by Elliott (1994), Baglinière & Maisse (1999) and Harris and Milner (2007); for brown trout and Atlantic salmon in Great Britain by Crisp (2000), and for Atlantic salmon more specifically in Buck (1993) and Mills (2003).
The importance of trouts in evolutionary studies has been pointed out by Behnke (1992), noting that the western North American trouts probably arose from a common ancestor some 2 million years ago, and evaluating the electrophoretic information on phylogenetic interpretations for that group up to the early 1990s, as did Bernatchez (2001), to improve our understanding of the complex evolutionary history of brown trout throughout its native Eurasian and North African distributional range. More recently Hendry & Stearns (2004) in their introduction and in the 13 following chapters have summarised the importance of trouts as a tool in evolutionary studies.
To some degree the general public interest in trouts, especially in stream habitats where they can be seen and angled more readily, has encouraged long-term studies of their behaviour, feeding, species interactions, competition, predation, reproduction and so on for several decades. The sequence of symposia on trouts in streams noted previously (Northcote 1969; Lobón-Cerviá 1988; Ecology of Stream Fish II, 2007. Abstracts, Leon, Spain) each spaced by about two decades illustrates the continuing scientific interest and productive research on trouts, now more of it becoming experimental on a number of key questions.
Furthermore, many scientists and watershed managers regard trouts as indicators of healthy ecosystems where clear and cool water with clean gravel and sub-surface water are not only aesthetically pleasing, but also necessary for safe water supplies. [Correction added after online publication 8 January 2008: insertion of sub-surface]. More recently, several populations and species have been considered ‘vulnerable’ or ‘threatened’ at national and international scales, whereas other trouts have been considered ‘invasive’ (Townsend 2003) and threat seriously the survival of numerous other species (McDowall 2006).
In September 1987, a symposium was held in Madrid, Spain to consider the research developing on ‘trouts’ in streams that extended over the many countries where such species occurred, either naturally or by introduction (Lobón-Cerviá 1988). This conference built on a similar focus covered nearly 20 years earlier at a symposium in Vancouver (Canada), the publication of which (Northcote 1969) still has requests for copies.
The 20 papers given at the Madrid symposium are not frequently cited, perhaps because the journal Polskie Archiwum Hydrobiologii is not well known at least in North America. The opening address (Northcote 1988b) at the Madrid symposium covered nearly 400 trout stream research articles published in 83 journals and 29 books, symposia, conference proceedings and theses. Literature dealing with stream trout research was spread widely and, at times, in journals that might not seem logical places to search for it. The review at the Madrid symposium indicated that there was an upsurge in experimental approaches to trout stream research, but did not document this quantitatively.
After the following two decades, 1987–2006, an attempt was made again to see if the trend for increasing experimental research in this field had continued. Most certainly the breadth of such research had expanded to >1300 relevant research publications covering an even broader range of topics (all references included in Appendix S1).
In the mid-1990s, several publications appeared on how in flowing water the ‘habitat provides the template upon which evolution forges characteristic species traits’ (Townsend & Hildrew 1994). This theoretical possibility was examined for the Upper Rhône River, first for fish (Persat et al. 1994), and then also for some 13 floral and faunal groups (584 taxa) where the theory was only very partially supported (Resh et al. 1994). See Fausch & Young (1995) as well as Gowan & Fausch (1996) for the only large-scale (250 m reaches), long-term (8 years), replicated (six streams), controlled and randomised manipulative experiment conducted to date on stream habitat enhancement for trout populations. Fausch et al. (2002) then covered what should be done to bridge the gap between research and conservation for stream fishes including trouts in landscapes to riverscapes. Fausch et al. (2002) show how a landscape to watershed viewpoint is required for stream–river systems to understand how processes interact at many scales for stream fishes and their habitat. More recently, Naiman & Latterell (2005) developed a set of eight ecological and social principles to better guide the management and conservation of inland water fish habitat with dimensions ‘ranging from pools to catchments to ecoregions, and from hours to seasons to centuries’. Such views are essential to maintain lotic habitats and our place as humans near them.
This literature review used most journals that were in the 1967–1986 coverage along with others obtained in reference sections of relevant publications over the most recent two decades. In this review, the definition of ‘experimental’ is used in a broad sense. As a consequence, we may have missed some experimental studies not indicated in titles, key words, abstracts and resumes of the publications covered. We apologise to all those workers who believe that they have used experimental approaches but whose papers were missed in this review.
For each publication the journal name, author/s, title of the work, species of trout covered (see later for details on trout names), occurrence in 22 subject areas and whether the study used experimental or nonexperimental methods, were considered. An experimental approach indicated that the author/s tried to test a well-defined hypothesis. Such research should not be followed to the exclusion of other approaches, as was pointed out at the Madrid symposium (Northcote 1988b). Research programmes may alternate observational and experimental approaches, using observational studies to generate hypothesis, design experiments to test them and observations again to validate them in the real world.
For this review, the country or country area where the research was conducted was also noted. In a few cases the latter information was not readily available, and in others the geographic area studied was very broad. Rather than give provinces or states for Canadian and USA research, these were divided into western and eastern portions at 100° longitude.
Largely following Nelson (1994), ‘trouts’ included Salvelinus alpinus, S. fontinalis, S. confluentus, S. malma, S. leucomaenis, but not S. namaycush (mainly lake dwelling). Included were the Atlantic trouts (Salmo salar and S. trutta) now introduced widely elsewhere (MacCrimmon & Marshall 1968), and the Pacific trouts Oncorhynchus clarki and O. mykiss also widely introduced (MacCrimmon 1971), and O. masou. According to Nelson (1994), O. rhodurus and O. masou and the most ‘troutlike’. The common names as given in Nelson et al. (2004) for this set were not used because doing so would have resulted in some charrs being called trouts, and other confusions.
Increased experimental research on trouts in streams
Over the last four decadal periods there has been a continued increase in experimental research on trouts in streams (Table 1). It almost doubled during 1987–1996 compared with the 1977–1986 period, and then has slowed down in the most recent decade. It now approaches 43% of all research reviewed.
Table 1. Assembly of research papers on ‘trouts’ in streams published in four decades from 1967 to 2006.
There are nine journals containing sizeable numbers of recent (1987–2006) research publications on trouts in streams (Table 2, all references in Appendix S1), notably the Canadian Journal of Fisheries and Aquatic Sciences (258), Ecology of Freshwater Fish (87) since starting in 1992, Freshwater Biology (30), Fisheries Management and Ecology (41), Journal of Animal Ecology (35), Journal of Fish Biology (238), North American Journal of Fisheries Management (42), Nordic Journal of Fisheries Research (34) regrettably stopped in 2002 and Transactions of the American Fisheries Society (139). These give publication coverage for 10 ‘trout’ species. Atlantic salmon and especially brown trout studies feature heavily in most of the nine journals noted and also frequently in the 81 other journals covered. Only in the NAJFM and TAFS do Atlantic salmon studies play a less dominant role, its position there being taken by publications on rainbow trout. The Journal of Animal Ecology and Canadian Journal of Fisheries and Aquatic Sciences contain over 60% of ‘trouts’ publications with experimental approaches, the Nordic Journal of Freshwater Research, Transactions of the American Fisheries Society and North American Journal of Fisheries Management over 40%, Ecology of Freshwater Fish and the Journal of Fish Biology near 33%, and the rest in the 20–30% range.
Table 2. A synopsis of experimental and nonexperimental studies on stream ‘trouts’* published in nine frequently used and 81 less frequently used journals from 1987 to 2006.
For the previous two decadal summary (Northcote 1988b), the number of trouts in streams publications by subject areas within the most prominent journals in which such work appeared were not determined, or were those in the other journals with small number of such papers. This is done in Table 3 with the addition of four more subject areas.
Table 3. A synopsis of stream ‘trout’ research studies classified into 22 subject areas published from 1987 to 2006 in the four most frequently used journals along with all other journals and reports (listed below with their sum of stream trout subject areas).
| 1 Life history traits, tactics||16||12||1||3||63||95|
| 2 Reproductive biology, behaviour||27||21||7||4||34||93|
| 3 Redds, eggs, alevins, young, parr ecology||108||102||31||20||116||377|
| 4 Food supply, feeding behaviour, energetics||36||27||9||12||72||156|
| 5 Sex, size, age, growth, maturity interactions||71||43||10||20||59||203|
| 6 Survival, mortality||31||10||7||5||25||78|
| 7 Movements, migrations, homing, residency||55||49||25||17||46||192|
| 8 Competition, aggression, territoriality||30||25||10||5||40||110|
| 9 Habitat, cover; diel, seasonal changes||91||35||54||13||107||300|
|10 Species interactions, predation, introductions, hybridization||29||22||7||10||37||105|
|11 Population size, structure, recruitment, production||55||28||27||8||90||208|
|12 Stock differences, status||7||9||2||7||22||47|
|13 Genetic differences, markers||37||27||12||9||36||121|
|14 Commercial, recreational fisheries||4||0||1||1||0||6|
|15 Acidification, toxic metals||16||9||3||1||12||41|
|16 Forestry, fire effects||14||1||2||1||11||29|
|17 Water velocity, discharge, impoundments, diversions||15||12||13||5||36||81|
|18 Silts, nutrient enrichment, urbanisation, industrialization||17||3||2||1||16||39|
|19 Agricultural activities, grazing||1||0||1||1||7||10|
|20 Management practices, hatchery stocking, stream restoration||44||22||6||10||60||142|
|21 Stream order, watershed, landscape interactions||40||0||17||15||28||100|
|22 Temperature and climate changes||4||8||13||0||25||50|
The top journals in this regard (Table 3) were the Canadian Journal of Fisheries and Aquatic Sciences (CJFAS) with 748 referrals over the two recent decades, followed by the Journal of Fish Biology (JFB) with 465, the Transactions of the American Fisheries Society (TAFS) with 260 and Ecology of Freshwater Fish (EFF) with 168 referrals. Subject area referrals for five other journals (Fisheries Management and Ecology, Journal of Animal Ecology, Nordic Journal of Freshwater Research and North American Journal of Fisheries Management) were in the 70s, with the others under 25.
More relevant are the number of referrals within each of the 22 subject areas (Table 3), which may be interrelated to varying degrees. The biology and ecology of trout redds, egg development, alevin hatching and emergence along with that for free-swimming fry and parr have been the most actively studied and referred to subject areas in trout stream research (Table 3; 22 subject areas). For each of the four journals covered separately this is almost always the case, as is for all the other journals combined (Table 3). This subject area probably lends itself more easily to experimental research than many of the others.
The next most actively studied subject area is that of habitat and cover use by stream trouts (Table 3, subject area 9), along with short-term diel changes in use as well as longer term ones, such as between summer and winter periods. This field, with a total of 300 referrals, has led to much experimental research, as well as to supportive observational work setting the stage for experimental approaches. Three more subject areas (Table 3, subject areas 5, 7 and 11) feature the size-age-growth area, the important aspect of moving or not to other habitats, and the effects that both of these have on population and production dynamics for stream-dwelling trouts, whether temporarily so or when moving to different habitats – large rivers or lakes, or for anadromous ones to marine waters. These three subject areas each have contributed about 200 research publications in the last two decades (Table 3). A closely related subject area one – that of food supply and feeding behaviour (subject area 4) – has contributed over 150 referrals.
Related groupings to those above are reproduction (Table 3, subject area 2), and those of another pair (Table 3, subject areas 8 and 10) in Table 3 that cover the broad subjects of social behaviour and species interactions – all with close to 100 referrals. A final subject area relevant to the preceding set in Table 3 is that of survival and mortality (Table 3, subject area 6) and possibly covered in part by the previously noted ones, hence its lower publication referral of 78 in the last two decades. These subject areas may culminate over evolutionary time into the first one (Table 3) – that of life history traits and tactics.
The next two subject areas – stock and genetic differences (Table 3, subject areas 12 and 13) – might have been combined but both are frequently used separately in publication key words, titles and abstracts. That the ‘genetic’ one is nearly treble that of ‘stock’ in many recent publications probably arises from mitochondrial DNA and related tools now frequently used to sort out and quantify stock differences. Such work can be very informative but is not necessarily experimental because of the definition for such research given previously.
Then follows a sequence of largely human-generated effects on stream trouts (Table 3, subject areas 14–20), and also including interactions with subject areas 21 and 22. The low number of six referrals for the first one that of commercial and recreational fisheries, suggests that many may have been missed, or alternatively that an inappropriate set of journals was examined, even though several included ‘fisheries’ and/or ‘management’ in their titles. The subject area sequence covered in Table 3– subject areas 15–18 – has more referrals (29–81) and includes both experimental and nonexperimental studies. The referral numbers found for agricultural-related effects on stream trouts was surprisingly low.
Subject area 20 in Table 3, dealing in large part with human-generated effects, considers the different management practices used to ‘restore’ streams and their trout populations with 142 referrals so categorised. Some restorations apparently were successful and others were not. The penultimate subject area 21 in Table 3 dealing with stream order, watershed and landscape interactions illustrates in a hundred referrals the importance that is being given to this field in all dimensional considerations of flowing water size, along with the inclusion of watershed and landscape approaches. This area must be extended to ‘airshed’ or ‘airscape’ thinking, as the frightening experiences in parts of central Europe, Scandinavia and eastern North America demonstrated a few decades ago with acid rain problems.
Subject area 22 in Table 3 covers that of changes (not many for the better) in temperature and climate at local, regional and global levels. There were 25 research referrals on this subject in three major journals and 50 overall in the other ones reviewed, showing climate change to be gaining wider consideration in trout stream research.
Coverage found for the previous 22 subject areas is briefly considered for five species of charr (salvelinids), two species of Salmo (Atlantic salmon and brown trout) and three species of oncorhynchids (Table 4). For charr species, subject area coverage is small with no referrals found in many areas, the exception being for brook charr, in part probably because of its wide introductions. In contrast, the two Salmo species, Atlantic salmon and brown trout, show high coverage in nearly every subject area, many having 25 or more referrals over the last two decades. This is also the case for rainbow trout but not for cutthroat trout, or for ‘troutlike’ Masu salmon, for which there are only three subject areas, each with a single referral (Table 4), though there are many in Japanese journals not covered.
Table 4. A synopsis by stream ‘trouts’ species for research subject areas over the 1987–2006 period in all Table 3 journals.
|Arctic char||1||2||5||7||6|| ||6||1||5||2|| ||4||4|| ||1|| || || || || || || |
|Brook char||3||4||17||11||4||3||16||5||14|| ||7||3||3|| ||7||5||3||3||1||1||10||3|
|D.V. char|| ||2|| || ||1|| ||2|| ||1|| ||1|| || || || || || || || || || || |
|Bull char||1||3||5|| || || ||3||3||7|| ||4||1||4|| || ||3|| ||1|| ||3||3|| |
|W-S char||1||1|| ||1||2|| ||6|| ||1|| || ||1|| || || || ||2|| || || || || |
|Atlantic salmon||18||24||136||39||73||27||35||25||50||9||38||11||23||4||14|| ||10||7|| ||21||20||8|
|Cutthroat trout|| ||2||7||4||4|| ||12|| ||15||2||5||4||5|| || ||1|| || || ||2||6|| |
|Masu salmon|| || || || || || ||1|| || || || || || || || || || || || ||1|| ||1|
In the last two decades (1987–2006) research publications for nine ‘trout’ species were assembled for 11 countries or country areas by their use of experimental or nonexperimental approaches (Table 5). For research publications on brook charr in western and eastern USA, Dolly Varden charr in Japan, and for Atlantic salmon in eastern Canada, eastern USA and the UK plus Éire, experimental approaches were used in over 40% of the publications examined. There was nearly as high a percentage for Atlantic salmon and brown trout (38% and 35%, respectively) in the Scandinavian countries, and combined with 50% for brown trout in Australia plus New Zealand (Table 5). Use of experimental approaches for rainbow trout research is high with over 50–70% of the publications covered in both western and eastern Canada, as well as in western USA, and for a third of those in Australia + New Zealand. More than 50% of the cutthroat trout publications reviewed for western Canada and USA also used experimental approaches (Table 5).
Table 5. A synopsis of 1987–2006 research publications on stream ‘trout’ species by research type – experimental, nonexperimental and per cent experimental if total are 10 or >10 – covered for 11 world areas: 1, western Canada; 2, western USA; 3, eastern Canada; 4, eastern USA; 5, Éire + UK; 6, Spain + Portugal; 7, Belgium + France; 8, Norway + Sweden + Finland + Denmark; 9, Rest of Europe + Russian Federation; 10, Japan; 11, Australia + New Zealand, plus overall total.
| Experimental||1|| ||5|| || || || ||7|| || || ||13|
| Nonexperimental|| || ||5||1||6|| || ||13|| || || ||25|
| Per cent experimental|| || ||50.0|| || || || ||35.0|| || || ||34.2|
| Experimental|| ||19||17||18|| || || ||1|| ||1||1||57|
| Nonexperimental|| ||23||30||26|| || || || || || || ||79|
| Per cent experimental|| ||45.2||36.2||40.9|| || || || || || || ||41.9|
| Experimental||1|| || ||1|| || || || || ||5|| ||7|
| Nonexperimental|| ||1|| ||1|| || || || || ||7|| ||9|
| Per cent experimental|| || || || || || || || || ||41.7|| ||43.8|
| Experimental||2||3|| || || || || || || || || ||5|
| Nonexperimental||6||24|| || || || || || || || || ||30|
| Per cent experimental|| ||11.1|| || || || || || || || || ||14.3|
|W-S char|| || || || || || || || || || || || |
| Experimental|| || || || || || || || || ||7|| ||7|
| Nonexperimental|| || || || || || || || || ||13|| ||13|
| Per cent experimental|| || || || || || || || || ||35.0|| ||35.0|
| Experimental||2|| ||45||17||63||2||2||44|| || || ||175|
| Nonexperimental||1|| ||49||20||43||9||8||71||7|| || ||208|
| Per cent experimental|| || ||47.9||45.9||59.4||18.2||20.0||38.3|| || || ||45.7|
| Experimental|| ||9||5||11||32||11||7||49||5|| ||16||145|
| Per cent experimental|| ||25.0||29.4||28.9||38.6||26.2||31.8||35.0||31.3|| ||50.0||33.9|
| Experimental||22||56||8||5|| || || ||3||2||2||4||102|
| Nonexperimental||11||54||3||14|| || || || ||2||1||8||93|
| Per cent experimental||66.7||50.9||72.7||26.3|| || || || || || ||33.3||52.3|
| Experimental||8||38|| ||1|| || || || || || || ||47|
| Nonexperimental||7||31|| ||1|| || || || || || || ||39|
| Per cent experimental||53.3||55.1|| || || || || || || || || ||54.7|
Over a hundred relevant reviews, book chapters and books have appeared over the last two decades that bear directly on the biology, ecology and related topics for trouts in streams or other flowing inland waters. These are not dealt with here but are included in the total set of over 1300 references used in this review, available on the online publication of this article as Appendix S1.
Future research for ‘trouts’ could be tied more strongly into Table 3 subject areas of life-history traits and tactics (1), stock differences (12), genetic differences (13) and into some of the potentially negative actions of management practices (20). Stream and population restoration activities could be tested in replicated stream sections using well-documented species and stock combinations as already have started in some regions. Suggestions for further subject area research combined with experimental approaches could include others of those in Table 3: feeding behaviour and interactions with cohabiting nonsalmonids; social behaviour, competition, aggression, territoriality, perhaps also with cohabiting nonsalmonids; experimental manipulation of riparian cover tied in with forests and forestry harvesting practices; experimental manipulation of agricultural activities and grazing effects, the latter building on some of the excellent approaches conducted in New Zealand; and also even experimental studies on climate change testing effects of simulated changes on replicated portions of trout stream watershed sections.
Dr D.T. Crisp, Dr K.D. Fausch and three anonymous referees reviewed an earlier version of this manuscript and provided excellent and helpful suggestions for improvement. Dr K.J. Hall and M.S. Greaven helped with the intricacy of getting reference information on-line from The UBC Library, or by the direct on-site assistance. L.A. Northcote provided much computer help to the senior author. Our wives Heather and Paloma dealt firmly but sympathetically throughout the time periods dedicated to this review.