Crow et al.: Vertical distribution, diet, and reproduction of Zameus squamulosus 
209 
(1996). Between 2001 and 2008, 19 specimens were col¬ 
lected through the Hawaii Longline Observer Program 
of the National Marine Fisheries Service, and these 
sharks were frozen on board the fishing vessels and re¬ 
tained for later examination. Two additional specimens 
captured by longline vessels were given directly to the 
authors by the fishermen of these longline vessels and 
were frozen until examined. 
Estimated hook depth was based on target set depth 
designated by the boat captain for the deepest part of 
the longline and based on an approximation of individ¬ 
ual hook positions on the longline. The median depth 
of the deepest hook on 266 deep sets (tuna sets) was 
248 m, whereas data for 333 shallow sets (swordfish 
sets) indicated a median depth of 60 m (Bigelow et. ah, 
2006). Previous reports on data from temperature-and- 
depth recorders attached to longlines of fishing ves¬ 
sels in Hawaii indicated that about 70% of deep sets 
were close to targeted set depth (Bigelow et ah, 2006). 
Therefore, records of hook depth of individual longline 
sets provide an estimated target depth for the capture 
of velvet dogfish. 
Frozen specimens were thawed and measured for 
TL, which was measured with calipers to the nearest 
0.1 mm; placing the shark on its side and its tail in 
the “natural” position, we measured in a straight line 
from the tip of the snout along the body axis to a point 
where a straight edge indicated a 90° angle to the body 
axis from the posterior caudal tip. The shark was then 
dissected, stomach and valvular intestine were exam¬ 
ined for food contents, and its entire reproductive anat¬ 
omy was examined. 
Diet 
The primary bait used with tuna longlines was Pacific 
saury (Cololabis saira), which can be recognized in the 
gut and excluded from a dietary analysis. Stomachs 
and valvular intestines were dissected carefully, and 
the contents were retained within a hand net made of a 
800-p fine mesh. Squid beaks, fish remains, and shrimp 
present in stomachs were preserved in 4-10% formalin 
for 3-10 d and then transferred to jars that contained 
40-50% isopropanol for storage until analyzed under a 
dissection microscope and identified to the lowest rec¬ 
ognizable taxa taxon. Squid beaks were identified by 
identified by a coauthor (RY). 
Female reproductive biology 
Reproductive tracts were examined to determine the 
stage of sexual maturity and reproductive status of 
specimens. Presence of ovarian ova, diameter and lo¬ 
cation of eggs, uterus width, and number of embryos 
were used to assess female reproductive biology. Ovar¬ 
ian ova, uterine ova, and uterine embryos were count¬ 
ed, and maximum ova diameter and uterine width 
were measured with calipers to the nearest 0.1 mm. 
The reproductive biology terms that we use were mod¬ 
ified from that used by Tanaka et al. (1990) for the 
frill shark (Chlamydoselachus anguineus) and by Yano 
(1995) for the black dogfish (Centroscyllium fabricii). 
Sexual maturity was determined on the basis of the 
size of ovarian ova and uterine width. Females with 
ovarian ova diameters <10 mm and with thread-like 
uteri (< 10 mm in width) were classified as immature. 
Females with both ovarian ova and uteri >10 mm were 
classified as mature. The reproductive classification 
that we used for mature females had 7 stages: 1) devel¬ 
oping ova, nongravid, with developing ovarian ova and 
expanded uteri (> 10 mm); 2) large ova nongravid, with 
ova and expanded uteri (> 10 mm); 3) ovulating, with 
eggs in body cavity; 4) uterine ova, ova in the uteri af¬ 
ter ovulation; 5) developing embryos, obvious embryos 
with yolk sacs in the uteri; 6) near-term embryos, uter¬ 
ine embryos that have absorbed their external yolk sac; 
and 7) postpartum stage: no uterine embryos and large 
flaccid uteri (>20 mm). Although no near-term embry¬ 
os were observed in utero in velvet dogfish collected 
off Hawaii, size at birth was estimated from lengths 
of near-term embryos (Graham 1 ; Romanov 2 ) and size 
of the smallest free-swimming individuals (Graham 1 ; 
Cadenat and Blache, 1981; Yano and Tanaka, 1984; Ta- 
niuchi and Garrick, 1986). 
Results 
Vertical distribution 
All 21 velvet dogfish obtained from Hawaii longline 
fisheries were females. Estimated target depths of 
longline sets were between 24 and 400 m (median 
depth: 182 m [Table 1]). Females ranged in size from 
576 to 839 mm TL. Specimens were captured within an 
area between 21-34°N and 148-168°W, and the major¬ 
ity of specimens were captured north of the Hawaiian 
Archipelago. The majority of sharks were captured on 
tuna longlines set in the morning between 0550 and 
1144, timing that is characteristic of the deep-set tuna 
fishing method. An exception was a shark caught on a 
longline deployed at 1506. The deep-set tuna longline 
sets had a soak time of 6-11 h and typically were re¬ 
trieved between 1505 and 2331 on the same day that 
they were deployed. 
Diet 
Nine of the 21 sharks examined had recognizable stom¬ 
ach or intestinal contents of mesopelagic and epipelag- 
ic prey (Table 2). Seven sharks contained squid beaks, 
squid eye lenses, and squid gladii. Fish remains (skin, 
eye lenses) were present in 4 sharks, and a shrimp 
1 Graham, A. 2016. Personal commun. Australian Nation¬ 
al Fish Collection, CSIRO, G.P.O. Box 1538, Hobart, Tasma¬ 
nia 7001, Australia. 
2 Romanov, E. 2016. Personal commun. Centre Tech¬ 
nique d'Appui a la Peche ReUNionaise (CAP RUN), Darse 
de Peche, Magasin 10, Port Ouest, 97420 Le Port, He de 
Reunion, France. 
