FISHERY BULLETIN: VOL. 85, NO. 3 



Information in the literature supports the hy- 

 pothesis that the two species are allopatric. Win- 

 ters ( 1970) found, based on adult samples, that A. 

 dublus occurred exclusively in offshore areas and 

 A. americanus exclusively in inshore bays around 

 Newfoundland. Reay (1970) pointed out that A. 

 dubius is exclusively an offshore species although 

 Richards (1982) indicated that there are offshore 

 components to A. americanus in the New England 

 area. This paper presents information on sand 

 lance in Fortune Bay, Newfoundland, as a further 

 test of the hypothesis that the two species are 

 allopatric. Size and distribution (seasonal and di- 

 urnal) information on Ammodytes larvae both 

 within and at the mouth of Fortune Bay is pre- 

 sented as well as length-frequency information 

 for other parts of the Newfoundland-Labrador 

 area. The development of the definitive number of 

 meristic characteristics is examined to identify 

 the Fortune Bay larvae and, in conjunction with 

 this, aspects of the developmental biology of Am- 

 modytes sp. in the area are also described. 



MATERIALS AND METHODS 



Fortune Bay is a three still fjord located on the 

 south coast of Newfoundland (Figs. 1, 2). Typi- 

 cally it has a two-layered structure with rela- 

 tively warm (1.9°C) deep water in the outer por- 

 tion in winter and cold ( -0.25"-0.50°C) deep 

 water in summer (de Young 1983). Annual sur- 

 face temperatures typically range from 0.0°-1.0°C 

 in February to 12°-16'C in August and Septem- 

 ber. 



From June 1979 to February 1981, several ich- 

 thyoplankton surveys were carried out annually 

 in Fortune Bay. The target species for the surveys 

 was herring, Clupea harengus , but due to low 

 numbers of herring larvae during the first three 

 years the study was relocated, and during 1982- 

 83 only one survey (July) was carried out in each 

 year in Fortune Bay. Sand lance data examined 

 here were collected during these surveys (1979- 

 83). 



Larvae collected m 1979-80 were taken by 

 standard oblique plankton tows (Smith and 

 Richardson 1977) using a 60 cm diameter bongo 

 frame with 333 ixm mesh netting on one side of 

 the frame and 505 p.m on the other. Tows were to 

 a maximum depth of 200 m where possible. Nets 

 were payed out at a speed of 0.77 m/second and 

 retrieved at a speed of 0.38 m/second. After 1980 

 samples were caught in nets when both sides of 

 the bongos were equipped with 333 \j.m netting. 



In February 1981, collections were made using 

 stepped oblique tows (5 minutes at each of 200, 

 150, 100, 50, 20, and m) with a nonclosing N.I.O. 

 rectangular midwater trawl (RMT-8) (Baker et 

 al. 1973). During this survey all stations were 

 fished during 6 hours of daylight with sets on the 

 same stations being repeated after dark. A 10- 

 min surface tow (3/4 m conical plankton net, 333 

 [xm mesh) was carried out during each of the 

 oblique tows with the RMT-8. 



In June 1981, sand lance larvae were collected 

 using bongo nets during the regular survey and 

 also from special stations to investigate diurnal 

 distribution. During each of these special sta- 

 tions, bongos were fished obliquely (to 200 m), 

 and 3/4 m conical plankton nets (333 ixm) were 

 fished for 10 minutes at the surface during day- 

 light hours and again (at the same positions) after 

 dark of the same day. 



Catches from all trips were preserved in a 5% 

 formalin solution buffered with sodium borate. 

 Fish larvae were later sorted, identified, counted, 

 and measured. Total length was recorded to the 

 nearest millimeter. A Macdonald and Pitcher 

 (1979) mixture analysis was performed on the 

 length-frequency distribution of the large sample 

 from June 1979 to investigate the fit of the data 

 to mixtures of normal distributions approximat- 

 ing the data. 



Species designation was determined using 

 meristic characters where possible (June 1979, 

 June 1981), namely, vertebral, anal fin ray, and 

 dorsal fin ray counts. Due to their small size, the 

 specimens were stained for cartilage using Alcian 

 blue and counterstained with alizarin red 

 (Dingerkus and Uhler 1977). Counts were then 

 done under a dissecting microscope (20-40 x mag- 

 nification) with the aid of the camera lucida. 



Additional sand lance length-frequency data 

 are presented, which have been on file at the Sci- 

 ence Branch, Northwest Atlantic Fisheries Cen- 

 ter in St. John's. These sand lance were collected 

 incidentally during tows for other target species 

 around Newfoundland using fishing and plank- 

 ton gear (see Figures 1 and 4). 



RESULTS 



Seasonal and Diurnal Distribution 



Table 1 lists the cruises in Fortune Bay from 

 1979 to 1983. 



Sand lance larvae were encountered during 9 of 

 the 17 cruises in Fortune Bay from 1979 to 1983 



632 



