INTRODUCTION

There has been a resurgence of interest in the relationship between fish population dynamics and oceanographic processes since the highly publicized Pacific salmon crisis off the west coast of British Columbia and the collapse of the Atlantic cod fishery. A relationship between "ocean climate" changes and fish populations can be inferred, but in the Canadian context, real data are meager. Shifts in the position and intensity of the Aleutian Low-Pressure System, (over the northern North Pacific) exert a major control on fisheries recruitment off British Columbia by influencing stratification and coastal upwelling that can alter plankton abundance (Brodeur and Ware, 1992). Unfortunately, the nature of long-term oscillations in this system and the accompanying effect on fish stocks are not known. Thus, the analysis of a long record of fish abundance data and accompanying ocean-climate "regime shifts" in coastal British Columbia is of strategic value to the Canadian west coast fishing industry. For example, the collapse of the Atlantic cod fishery has been partially linked to North Atlantic temperature/climate variation (Smith and Page, 1996).

Fishery collapses (in addition to the recent problems experienced by the salmon fishery) have occurred before on the west coast of North America. For example, the northern Pacific sardine fishery, initiated in the 1920s, began to collapse in the early to mid-1940s. By the early 1950s the central Californian fishery had also collapsed, followed by the southern Californian fishery in the 1960s. This progressive collapse from north to south was apparently aided by a significant change in the pelagic habitat in the California Current system and was associated with large-scale climate change in the Pacific during the 1940s (Francis and Hare, 1994). The gradual return of sardines to waters off Vancouver Island since 1990 is associated with the gradual recuperation of the populations off southern California (Baumgartner et al., 1992). A gradual expansion of the population probably began soon after a large-scale climate shift of the Pacific in 1977 (Robinson, 1994). Thus, although the historical fisheries record is limited for the Pacific coast of North America, sardine populations are known to expand and contract over time-scales of several decades or more and over distances that encompass the entire North American Upwelling Zone (Baumgartner et al., 1992).

We now know that many of the oceanic drivers that affect fish population dynamics in the NE Pacific (Ware and Thomson, 1991) have greater than decadal scale return times, and since the settlement history of the region, particularly British Columbia, is relatively short, available anecdotal and commercial fishing records are of limited use in assessing long-term variation in fish populations. Therefore, other methodologies are required for this research (Leaman, 1993). The use of osteological and squamatological (scale) fish remains as proxies to estimate paleofish populations was pioneered in the 1960s to better understand clupeid (sardine, herring, anchovy) population dynamics off the coast of southern California (Soutar, 1966; Soutar and Isaacs, 1969; Soutar and Isaacs, 1974). Subsequently research has been expanded all along the coast of the Americas (O'Connell and Tunnicliffe, 2001), most recently in Effingham Inlet on Vancouver Island (Dallimore, 2001). Effingham Inlet has proven to be an ideal area for this research because varved sediments deposited in the quiet and anoxic bottom waters of the contained basins archive a very well-preserved record of osteological (bone) and squamatological (scale) fish remains and other biologic proxies, such as diatoms, dinoflagellates, and foraminifera (Patterson et al., 2000).

Despite the existing body of research on fish remains available, it has proven surprisingly difficult to identify fish remains, not because of poor preservation, but due to a lack of accessible resource material. Most of the few research groups involved in this line of research have collections available on hand but there have been relatively no attempts at publishing definitive comparative references, with the exception of a few non illustrated keys (see Casteel, 1976, for summary). In this paper we aim to rectify this deficiency by presenting a detailed atlas of squamatological material collected from living representatives of common pelagic fish species found in southern British Columbia coastal waters. In addition, as some important types of scales (e.g., sardines, herrings, and anchovies) become altered and fragmented in the sedimentary record, making identification difficult, we provide comparative examples of selected fossilized material obtained from Effingham Inlet (Figure 1).