FISHERY BULLETIN VOL 77, NO .' 



low, but these samples were used in length- 

 frequency distributions and reproduction 

 analyses. 



All specimens were identified by me from Firth 

 and Pequegnat's (see footnote 4) key to the 

 Polychelidae. Short carapace length (SCLi, i.e., 

 the distance from the median posterior margin of 

 the carapace to the orbit, was measured to the 

 nearest millimeter. 



Sex and gonad condition were recorded for all 

 polychelids, and gonads representative of stages of 

 development were obtained for histological 

 examination and placed in Davidson's fixative 

 (Humason 1972). Validity of female gonad stages 

 was determined by gross ovarian morphology, 

 ovarian histology, and oocyte diameter. The 

 longest horizontal diameter of 15 oocytes ran- 

 domly chosen from excised ovaries of each lobster 

 was measured with an ocular micrometer. 



Fecundity was estimated from total external 

 egg number. I stripped eggs from the pleopods, 

 placed them in a graduated tube, and adjusted the 

 volume to 10 ml w^ith water. After mixing, I took 

 three 0.5 ml aliquots and counted eggs from the 

 aliquots on a gridded Petri dish. I then noted the 

 degree of embryological development of eggs, 

 similar to descriptions by Meredith (1952) and 

 Allen ( 1966), and measured the longest horizontal 

 diameter of 15 randomly chosen eggs. 



I also removed stomachs from preserved lobsters 

 and sorted and identified their contents where 

 possible. The importance of food taxa was then 

 determined from their numerical abundance. 



RESULTS 



Stereoniastis tniiui (Smith) 



the Middle Atlantic Bight collected 459 S. nana 

 from depths of approximately 613-2,642 m and 

 temperatures of 2.4°-5.0° C. Analysis of variance 

 showed a significant difference (Table 1 ) between 

 abundance of S. nana for depth intervals shown in 

 Figure 1. Scheffe's multiple mean comparison test 

 iSnedecor and Cochran 1967) showed the mean 

 catch rate, expressed as log,,, (.v + 1) 0.5 h tow, to 

 be significantly higher at depths of 1 ,400-2,599 m. 

 There was no discernible change in depth distribu- 

 tion of this species with season. 



There was also no apparent segregation of sex 

 with depth since both male and female S. nana 

 occurred throughout the depth range. Chi-square 

 analysis using Yates correction (Woolf 1968) 

 showed females and ovigerous females to be sig- 

 nificantly more numerous than males at arbitrar- 

 ily chosen depth strata of 1,200-1,999 and 2,000- 

 2,800 m (Tables 2, 3). There was no significant 

 relationship between average size of S. nana and 

 depth of capture (F = 0.056, df = 1,460). 



Males (mean = 22 mm SCL), females (mean = 

 25 mm SCL), and ovigerous females (mean = 2(S 

 mm SCL) differed significantly from each other in 

 size by analysis of variance (Table 4) and Scheffe's 

 multiple mean comparison test. Sex ratios varied 

 significantly with size, with females predominat- 

 ing at lengths >26 mm SCL and males at lengths 

 <22 mm SCL (Table 5). 



Among the Polychelidae, sperm transfer is 

 accomplished by attachment of spermatophores to 

 the surface of the posterior sterna of the females 

 ( Andrews 1911). Most ovigerous and nonovigerous 

 females >23 mm had externally attached sper- 

 matophores (Figure 2). All ovigerous females 

 except five damaged individuals had spermato- 

 phores attached. It is probable that spermato- 



Stereomastis nana is found in the three major 

 oceans but not in the Mediterranean and Carib- 

 bean Seas or the Gulf of Mexico (Firth and 

 Pequegnat see footnote 4). Its bathymetric dis- 

 tribution within the western North Atlantic off 

 the east coast of the United States was reported to 

 be 1,289-3,506 m (Smith 1884, 1887); off Green- 

 land and Iceland, specimens have been taken from 

 1,271 to 2,271 m (Hansen 1908). 



Abundance data based on our 13.7 m otter trawl 

 catches show that S. nana constitutes 20^t by 

 number of the total benthic decapod fauna at 

 depths below 1,200 m. Its importance within the 

 benthic decapod community diminishes to 0.3'/^ at 

 depths between 400 and 1,199 m. Trawls within 



Table I — One-way analysis of variance on abundance, expres- 

 sed as logio (* -^ 11 per 0.5 h tow, of Stereomastts nana and .S. 

 ^culpta by depth interval (see Figure 1 for description of depth 

 intervals). 



Source of variation 



S nana 



Among groups (depth 



interval) 

 Wittiin groups 

 Total 



S sculpta 



Among groups (depth 



interval) 

 Within groups 

 Total 



'SS - sum of squares. 

 'MS = mean squares, 

 ■■P- 01 



dl 



SS' 



MS' 



436 



