THE 

 CONTINUOUS PLANKTON RECORDER 



By A. C. Hardy, m.a. 

 (Plates I-IV; Text-figs. 1-3 1) 



INTRODUCTION 



The first Continuous Plankton Recorder was used on the R.R.S. 'Discovery' in 

 the years 1925-7, and a brief note on its working was published during that 

 voyage (Hardy, 1926). A second machine of improved design was made in 1929, to be 

 followed by a number of others incorporating from time to time small modifications. 

 It has now been used as a routine instrument both in the Antarctic on the R.R.S. 

 ' Discovery II ' and for the last four years in a general plankton survey of the North Sea 

 on commercial steamship lines radiating from Hull. Whilst it may well undergo further 

 changes, and lines of possible development are suggested here, it seems desirable that 

 a full account should now be given of its construction, method of working and the 

 advantages and limitations of the results obtained by its use. This is the object of the 

 present paper; it will be followed, in an adjacent paper, by an account of the uneven 

 distribution of oceanic plankton demonstrated by its use on the 1925-7 voyage. A 

 report on the North Sea recorder survey is being prepared for publication elsewhere in 

 collaboration with my colleagues at Hull: Dr G. T. D. Henderson, Mr C. E. Lucas, 

 Mr J. H. Fraser and Mr K. M. Rae. A few examples of this work have already been 

 published as illustrations to a brief account of this method of survey (Hardy, 1935). I 

 am indebted to many people for technical assistance and encouragement during the 

 development of the machine ; acknowledgements will be made in different parts of the 

 paper and in a section on p. 502. 



PURPOSE OF THE MACHINE 



Whilst working in the North Sea before joining the Discovery Expedition of 1925, I 



had been frequently impressed by the patchy distribution of the plankton. 1 In making 



1 On one occasion (31 March 1922), when making a survey of post-larval herring and plankton in the 

 Southern Bight, using the Petersen Young Fish Trawl and Nansen tow-nets at the surface, midwater and 

 near the bottom, I decided to repeat the routine, so that four sets of observations were taken at one station 

 instead of one set as at all other stations in the survey. The object was to test the validity of the methods 

 employed in such a survey. The nets were used at 1900, 0700, 1430 and 1900 o'clock whilst the ship drifted 

 in the surface current for 24 hours. Wind was negligible. The numbers of post-larval herring taken at the 

 different times were 2448, 24, 7 and 341, and their average lengths were 25, 23, 22 and 26 mm. respectively. 

 (The figures have already been published by Wallace, W., 1924. Fishery Investigations (England), Ser. n, 

 vii, no. 4, p. 4.) The range of variation in the numbers caught at this one station was as great as 

 that found for all the other thirteen stations of the survey: i.e. 2-793. (The stations were on an average 

 about 20 miles apart.) The numbers of young sand eels and the volumes of zooplankton also showed con- 



