Five species of southern ocean oreosomatid fishes (within four genera) are reviewed: Pseudocyttus maculatus Gilchrist 1906, an easily distinguished species widespread in the southern oceans, juveniles previously described as

Chinook salmon (Oncorhynchus tshawytscha) redd characteristics were measured at three locations in the glacially-sourced mainstem and two locations in small, primarily rainfed, tributaries of the lower Waitaki River. Mean redd size was 17.5 m 2 ; lengths of redds ranged from 2.4 to 10 m and their widths ranged from 1.2 to 9.5 m. Water depth (mean ± SD = 0.37 ± 0.09 m) and water velocity (mean 0.68 ± 0.27 m s -1 ) were significantly different between two of the mainstem locations but this may have resulted from differences in site conditions rather than preference by salmon. Substrate permeability, water flow rate through the substrate (termed apparent velocity herein) and oxygen were high in all redds and any "stress" to early fish life stages would likely result from excessive permeabilities (> 55.0 m h -1 ) and apparent velocities (> 0.15 m h -1 ) rather than

A review of the Australasian species Caranx platessa Cuvier, C. lutescens (Richardson), C. nobilis Macleay, and Usacaranx archeyi Griffin has shown these to be junior synonyms of C. georgianus Cuvier. This species has been redefined in terms of variation shown by specimens from temperate Australia, Lord Howe, Norfolk and Kermadec Islands, and New Zealand. JOURNAL OF THE ROYAL SOCIETY OF KEW ZEALAND Lengths of larger fish were read directly against a metric scale, but other measurements were made to an accuracy of 0.5 mm with fine-pointed calipers. Very small fish were measured using a binocular microscope. All measurements are from specimens fixed in formalin and preserved in alcohol. Definitions and abbreviations of measurements are as follows: Standard length (SL) from premaxillary symphysis to hypural crease; body depth (BD) taken just anteriorly to vent, is also greatest body depth; head length (HL) to end of fleshy operculum; snout length (SnL) to membraneous opening of orbit; eye diameter (ED) as diameter of cornaj"; upper jaw length(UJaw L) to posterior edge of maxillary; pectoral length (PL) greatest distance from insertion of fin to tip of longest ray; length of third dorsal spine (L 3DSp) ; length of first dorsal ray (L 1DR) ; length of curved portion of lateral line (LL Curv) shortest distance from its anterior commencement to end of curve or point where are, when protracted, intersects forward extension of straight portion of lateral line; length of straight portion of lateral line (LL Str) from intersection with curved portion to most posterior scute and parallel to longitudinal axis. Standard counts were made of the following: dorsal and anal fin rays and spines; scutes (only prominent scutes along straight portion of lateral line); gill rakers, on both upper and lower limbs of first gill arch; vertebrae, from radiographs, read to include the hypural vertebra. MATERIAL EXAMINED Collections held by the following institutions and totalling 14~specimens

Southern Buller's albatrosses (Diomedea bulleri bulleri Rothschild) are frequently associated with fishing vessels off southern New Zealand, and may depend substantially on discards from fishing vessels, particularly trawlers, during chick rearing. To test this hypothesis, the diet was determined from 188 chick regurgitations collected at The Snares (May 1996, July 1996 and 1997) and Solander Island (May and July 1997). Fish (mostly discards but also some small species obtained naturally) was the dominant prey category, occurring in 92% of samples, and forming 65% by weight of solid food consumed. Salps {Pyrosoma sp. and lasis zonaria) were the most abundant prey items (44% of all items), and second most important by frequency of occurrence (78%) and weight (24%). Cephalopod remains (mostly arrow squid (Nototodarus spp.) and Histioteuthis atlantica) occurred in 53% of samples, but contributed only 17% by number and 7% by weight. Other dietary items included crustaceans (chiefly Munida gregaria, Nectocarcinus antarctica, and Lepas anatifera), bird feathers, and flotsam. Fishery discards were present in 70% of samples, constituted 60% of the diet by weight, and comprised 91% of the weight of all fish consumed. Discard M99027

Brown trout (Salmo trutta Linnaeus 1758) and rainbow trout (Oncorhynchus mykiss Richardson 1836) populations in littoral zones of eight South Island, New Zealand alpine lakes were compared using gill and seine net sampling during summer. Lakes were selected to provide a matrix of lake level and water clarity variations and to assess how these variables influenced trout abundance (as reflected by catch rate), depth distribution, and size. Brown trout were small and in poor condition in three turbid lakes with shallow littoral zones and were scarce in one of these with a 14 m fluctuation. Although brown trout condition was generally higher in clear lakes, Lake Wanaka fish were an exception. Trout depth distribution was positively related to depth of the littoral zone. Brown trout were caught at most depths whereas rainbow trout showed variable depth preferences in different lakes. Rainbow trout catch rate, weight, and condition factor were unrelated to differences in depth of the littoral zone or lake level fluctuations, and it was concluded that spawning success and limnetic food supplies may be more important for this species. Spawning interactions may account for the different species composition between the stable clear Lake Wanaka, and the adjacent fluctuating clear Lake Hawea. Moderate and slow lake level fluctuations appear to have limited effects on trout in clear lakes, probably because trout are able to use deep littoral habitat and food supplies.