SAMPLING ANCHOVY LARVAE WITH A PLANKTON PURSE SEINE' 



Garth I. Murphy- and Rorf.rt I. Clutter' 



ABSTRACT 



A miniature purse seine (100 ft by 21 ft) was constructed of 333 fi Nitex. It was used 

 in Kaneohe Bay, Hawaii, together with a 1-m plankton net constructed of the same ma- 

 terial in order to evaluate the sampling efficiency of the towed plankton net on anchovy 

 larvae (Stolephoriis purpureus). The results show that during the day, the purse seine 

 is at least an order of magnitude more efficient for larvae over 5.5 mm in length. The 

 largest larva caught by the plankton net was 14.5 mm and by the purse seine 29.5 mm. 

 At night the plankton net was relatively more effective than during the day, catching 

 about 60% as many larvae as the purse seine over the interval 8.5-19.5 mm. The maximum 

 size taken increased to 21.5 mm, but the maximum taken by the purse seine increased 

 to 50 mm. An attempt was made to rationalize the difference between the day plankton 

 net and purse seine catches by a geometric model involving alarm distance and larval 

 swimming speed. The results are' moderately satisfactory. 



More limited data on anchovy larvae catches by a 10-ft Isaacs-Kidd trawl, Vi-m, Vi-m, 

 and 1-m plankton nets are presented and discussed in the framework of the meter net- 

 purse seine data. These analyses suggest one or more paradoxes in the larval escape 

 problem, or that the data are inadequate. 



The classical approach to sampling zooplankton 

 is the towed net. This has two disadvantages: 

 an unknown fraction of the organisms in the 

 path of the net escape by dodging and the net 

 integrates the organisms living along a transect 

 of considerable length — making it difficult to 

 consider the catch from a tow as representing 

 an assemblage of coexisting organisms. The 

 dodging problem has been considered by (among 

 others) Fleminger and Clutter (1965) with re- 

 spect to planktonic Crustacea, Ahlstrom (1954) 

 and Isaacs (1965) with respect to fish larvae, and 

 McGowan and Fraundorf (1966) with respect 

 to zooplankton. A general review of the prob- 

 lem is given in UNESCO (1968). 



In recent years there have been attempts to 

 apply community theory to the pelagic realms, 

 e.g., Fager and McGowan (1963) and Venrick 



^ Contribution No. 390, Hawaii Institute of Marine 

 Biology, University of Hawaii. Supported in part by 

 NSF Grant GB 5698 and the Marine Life Research Pro- 

 gram, Scripps Institution of Oceanography. 



^ Department of Oceanography, University of Hawaii, 

 Honolulu, HI 96822. 



' Formerly Department of Oceanography, University 

 of Hawaii. 



(1971). Such attempts should involve samples 

 that represent organisms likely to be coexisting. 

 Grab sampling such as used by Venrick (1971) 

 for diatoms is the obvious method of choice as 

 the scale of coexistence can be specified with pre- 

 cision, but existing grab samplers engulf so little 

 water that they are not likely to afford meaning- 

 ful information on the medium and large zoo- 

 plankton. The probability that a towed net will 

 integrate several communities has also been rec- 

 ognized and several solutions developed with re- 

 spect to subdividing a tow, e.g., Longhurst et al. 

 (1966). But if the tow is subdivided small 

 enough for the dimensions to be meaningful, it 

 will not likely strain enough water to sample the 

 less abundant forms. 



The purse seine (and other nets of similar de- 

 sign) is an extremely effective large volume grab 

 sampler in wide use by commercial fishermen. 

 Essentially, a wall of net is set in a circle and the 

 bottom closed (pursed) by drawing on the purse 

 lines. This kind of grab sampler can only be 

 effective at the surface and, therefore, is of gen- 

 eral application only in shallow water or under 

 circumstances where the surface fauna is of in- 

 terest. 



Manuscript accepted April 1972. 



FISHERY BULLETIN: VOL. 70, NO. 3. 1972. 



789 



