180 



Fishery Bulletin 88(1), 1990 



CONN 

 NEW 

 YORK , ^l- 



MASS ~^ S\ 



ISLAND ,»/;/•• ty^^ 



k y®  





«^" 



.vA^" 



.^^ 



DEL , 



70 



Ot^ 



t>«»«^ 



1. „, ,■ joo' 



50 100 



KILOMETERS 



68" 



Figure 1 



Station locatitms of the four study sites in the 

 Middle Atlantic Bight-Georges Bank area. 

 Other sites examined which did not produce 

 reliable samples oi Erhinarachnius parma are 

 also included ( • ). 



Several studies have examined aspects of £". parma's 

 life history (for review, see Caracciolo and Steimle 

 1983). Only two studies have examined its growth 

 (Cocanour 1969, Graef 1977) and none its production. 

 Since E. parnia can be a major contributor to the 

 overall benthic community biomass, estimates of its 

 production could greatly support an estimate of total 

 community production. Warwick (1980) has shown that 

 a major proportion of total community production is 

 often attributed to one or a few species. 



The purpose of this study was to examine E. parmn 

 size frequencies from a series of at least annual collec- 

 tions to provide information on the species' population 

 structure, dynamics, and growth, and to estimate its 

 production at stations in the Middle Atlantic Bight and 

 on Georges Bank. The term "production" is used in this 

 paper to indicate the amount of biomass added to a unit 

 of bottom area per year. 



Material and methods 



Echinarachnius parma were obtained from archived 

 benthic collections made at three stations (11, 15A, 17) 

 in the Middle Atlantic Bight and one station (23) on 

 southwestern Georges Bank (Fig. 1). Water depths at 

 these stations were 20 m (Stn. 17), 30 m (Stn. 15 A), 

 50 m (Stn. 11), and 70 m (Stn. 23). 



The archived collections consisted of 3-5 replicate 

 0.1-m^ Smith-Mclntyre grab collections at each station 

 sampled on a quarterly to annual basis, 1978 to 1985, 

 as part of the Northeast Monitoring Program (NEMP). 

 Collection periods varied slightly between stations (see 

 Figures 2-5 for specifics). Grabs were handled to 

 minimize loss of sample by surficial material washout; 

 grabs suspected of being excessively disturbed were 



resampled. The samples were washed through l.O-mm 

 mesh sieves prior to 1980, and through 0.5-mm mesh 

 sieves thereafter. While E. parma larvae are reported 

 to undergo metamorphosis and settle to the seabed 

 from the plankton at a width of about 0.4 mm (Coca- 

 nour 1969), the l.O-mm sieve collected most postlar- 

 vae because few were collected on the 0.5-mm sieve 

 when both sieve sizes were used together in 1980 and 

 1981. The sieved benthic samples were fixed in lO'Fo 

 buffered formalin, transferred to 70% ethanol within 

 a few days, and later sorted. 



Size-frequency distributions were based on measur- 

 ing the diameter of each individual, parallel to the anal 

 pore margin of the test, using either an ocular micro- 

 meter or vernier caliper. To estimate the ash-free dry 

 weight (AFDW) of different cohort size classes, addi- 

 tional E. parma specimens were collected at the above 

 sites and frozen, then sorted into 5-mm size groups 

 covering the 5-50 mm size range, with each group con- 

 taining 3-10 individuals, as available. Each size group 

 was dried for 12 hours at 60°C and ashed for 4 hours 

 at 500°C to produce a mean AFDW value per mean 

 width (L in mm) and the regression; AFDW = 0.02L- '•'' 

 (/•- =0.991). Overlapping cohorts were initially sepa- 

 rated by probit graphical analysis (Cassie 1954), and 

 these estimates were refined by the NORMSEP pro- 

 gram (Tomlinson 1971) to separate overlapping dis- 

 tributions. Growth rates were estimated by modal 

 progression analysis of the time series of size-frequen- 

 cy histograms (Figs. 2-5) for collections at each sta- 

 tion. Production estimates are based on Crisp's (1971 ) 

 growth-increment survivorship-curve method for po])U- 

 lations with distinguishable recruitment tuid age classes 

 (cohorts). Recruitment events in the Middle Atlantic 

 Bight were assumed to l)e the result of fall-winter 

 spawning, based on when recruits are first detected in 



