Hoff: A nursery site of Bathyra/a parmifera in the eastern Bering Sea 
235 
dominately in early stages of development (newly de- 
posited) and a large percentage were viable and at 
high densities (>50,000 eggs/km 2 ) to allow tracking 
of embryonic development. The index site constituted 
an approximate 1-km 2 area where the highest egg case 
density trawls were located. During each seasonal sam- 
pling period, a single 5 to 10 min bottom trawl targeted 
the index site. The data collected from seasonal sam- 
plings were similar to those previously described for the 
July-August 2004 trawl investigation; however because 
of time limitations, trawl data were limited to bottom 
depth, temperature, distance fished, and start and end 
latitude and longitude during seasonal sampling. 
Collection of biological data 
All skates were weighed and enumerated, or a weighed 
numerical subsample was used to estimate total num- 
bers from weighed samples. All egg cases were identified 
to species and documented as empty (posthatching) or 
full (prehatching), including eggs that may have been 
damaged by the trawl. A random sample of full egg 
cases was fixed in 10% formalin from each sampling 
period for embryo measurements. Density estimates 
for skates and egg cases were calculated as the number 
of eggs encountered per km 2 by using area swept by 
the net and the number of individuals encountered in 
each trawl. 
All skates encountered were identified to species and 
sex, and total lengths (TL, to nearest cm) and weights 
(to nearest 0.1 g) were recorded. Biological data were 
collected on randomly selected Alaska skates during all 
seasonal sampling to determine maturity state, repro- 
ductive state, and diet composition. During the initial 
July 2004 sampling, 67 female and 45 male Alaska 
skates were examined and during all subsequent sam- 
pling periods from 2 to 12 males and 5 to 17 females 
were examined. For each skate sampled the species, 
sex, total fish length, total fish weight, stomach content 
weight, and general diet composition were recorded. Re- 
productive state of males and females were determined 
by following maturity stages detailed in Matta and 
Gunderson (2007). 
Embryo length-frequency measurements 
Formalin-fixed egg cases were neutralized and soaked in 
tap water for up to four days before measurements were 
taken. Egg cases were cut open, embryos excised from 
the yolk, and total lengths (TL) were measured to the 
nearest 0.5 mm. For analysis, lengths were rounded to 
the nearest millimeter. Measurements were taken from 
the anterior tip of the snout or disc to the posterior tip 
of the tail filament. 
Growth rates of Alaska skate embryos were estimated 
by following methods similar to those used for the ju- 
venile English sole ( Plueronectes vetulus) off the Wash- 
ington coast (Shi et ah, 1996). Natural mode breaks 
were used to demarcate cohorts from embryo length 
frequencies. A mean embryo length was estimated for 
each cohort at each sampling period and plotted along 
with the corresponding sampling date. A best-fit linear 
model was used to determine daily growth rates of 
each cohort. Embryonic growth was assumed linear 
throughout development, and the cohort data indicated 
that the linear model was applicable. Hatching dates 
were estimated by using a mean hatching size of 224 
mm TL (mean of all near hatching embryos, n = 39) 
and the average growth rate from the linear regres- 
sion. Hatching-date estimates were defined as the time 
required to reach 224 mm TL based on the length and 
date of capture. Egg-deposition date was obtained for 
each embryo measured by back-calculating the time 
required to reach a size of 1 mm based on the length 
and date of capture and the average growth rate. An 
estimate of time between cohorts was calculated as the 
difference between mean lengths for each cohort divided 
by daily growth rate to obtain average time between 
each depositional event. 
Analysis of developmental period determined 
from the literature 
A review of previously published studies on hatching 
duration and rearing temperatures was synthesized for 
comparison with hatching duration and rearing tempera- 
tures obtained from this study. Species in this analysis 
were limited to oviparous elasmobranchs, which encom- 
passed a diverse group of 13 chondrichthyan fishes: a 
chimera, the spotted ratfish ( Hydrolagus colliei), two 
catshark species ( Scyliorhinus spp.), and ten species of 
skates in three genera (Raja, Leucoraja, and Okamejei) 
(Table 1). Species reviewed were found from subtropi- 
cal to temperate waters spanning a range of tempera- 
tures from 4.6°C to 24°C. Developmental periods and 
rearing temperatures were obtained from the reported 
literature for each study. When a range was reported 
for either developmental period or rearing temperature, 
an arithmetic mean was calculated from those values. 
Temperature and embryonic development period were 
plotted and a nonlinear regression algorithm was applied 
to the data. 
Habitat use 
The distribution of life stages of the Alaska skate was 
investigated by examining bottom trawl survey data 
from the eastern Bering Sea summertime groundfish 
survey of the AFSC for years 2000 through 2007 (Lauth 
and Acuna, 2007). Alaska skate density for each sta- 
tion was estimated as the summed catch per unit of 
effort (CPUE, number of skates/km 2 ) obtained at each 
station for the eight years surveyed. Density estimates 
were calculated at each trawl station for each life his- 
tory stage of the Alaska skate: juvenile (<300 mm TL, 
newly hatched to age +1); immature (301-920 mm TL); 
and adults (>920 mm TL, average maturity size; Matta 
and Gunderson, 2007). Distribution maps were produced 
with ArcMap (vers. 8.3, Environmental System Research 
Institute (ESRI) Redlands, CA). 
