Changes in levels of chlorophyll' a' and phaeopigments and in the species composition of phytoplankton populations at station Ei during the summer months are described. After the spring diatom outburst in April, diatoms continue to be abundant during May, first in the surface water and then below the thermocline. From June to August dense populations of dinofiagellates in the thermocline are the dominant feature. In September diatoms are again abundant as the thermocline is eroded. Factors controlling the growth of these different algal communities, including grazing, are discussed. P. M. HOLLIGAN AND D. S. HARBOUR III fluorometer was used to detect chlorophyll ' a ' fluorescence. The time delay between water entering the hose and reaching the flow cell of the fluorometer was 60-90 s, depending on the length of hose in use. The hose inlet was raised or lowered at a rate of 2 m min" 1 for routine observations, and mixing within the hose tended to smooth out any sharp discontinuities in chlorophyll fluorescence. Thus, whenever the hose inlet was held at the depth of subsurface maxima, maximum fluorescence values were higher by as much as 50 % than those observed in normal profiles.Plant pigments in discrete filtered samples were determined spectrophotometrically (Strickland & Parsons, 1972). The filters were placed immediately in 90 % acetone or stored for as short a time as possible (generally < 24 h) in the dark at -20 °C prior to extraction. Where appropriate, corrections were made to chlorophyll' a' values for interference by phaeopigment. Rates of carbon dioxide fixation were determined as described by Pingree et al. (1976).Water samples for phytoplankton counts were collected by bottle or from the outflow of the fluorometer and preserved with Lugols Solution (10 g I + 20 g KI in 200 ml distilled water + 20 ml glacial acetic acid) at a final concentration of 0-5 %. Cell counts were made by UtermohPs inverted microscope technique (Lund, Kipling & Le Cren, 19585 Lovegrove, i960), using a Wild M40 microscope that was fitted with phase contrast equipment, at magnifications of x 187, x 375 and x 750. Each organism was identified to the lowest possible taxonomic rank and nomenclature follows the check lists of Hendey (1974) for diatoms and Parke & Dixon (1976) for other algal groups. Authorities are given only for diatom species that are not listed by Hendey. The category ' flagellates' includes all cells that had recognizable flagella and/or plastids (excluding dinoflagellates and diatoms), and was dominated by forms with a mean diameter in the range 2-10 finx. A copy of the complete taxonomic data has been deposited for reference in the library of the Marine Biological Association.The volume of samples used for sedimentation varied between 50 and 265 ml, according to the overall abundance of phytoplankton as shown by the fluorometer response. At least 100 cells of each of the more abundant species were enumerated and, to obtain some measure of accuracy for the counting procedure, 10 sub-samples of a larger, well ...