444 DEPARTMENT OF THE NAVAL SERVICE 



during the intervening two months. With S. maxima it extends well up the Laurent- 

 ian channel as shown at stations 85, 86, and 87 and, like that species- it is most abund- 

 ant on the north side of the channel. The decrease in abundance from north to south 

 across the channel is very evident, for the deep vertical hauls eight, four, and two 

 specimens were obtained at stations 85, 86, and 87, respectively, and for the shallow 

 vertical hauls two, one, and zero specimens for the same stations. 



It was not obtained at any of the stations of the Princess cruises and, being a 

 surface form, it could scarcely have been missed. It does not therefore extend into 

 the St. Lawrence gulf, doubtless being killed by the cold brackish surface water. Its 

 absence over the banks south of Newfoundland (St. Pierre and Green banks), and 

 over those north of Sable island (Banquereau, Misaine, etc.) and along the Nova 

 Scotia shore is noteworthy as denoting the absence here of oceanic water. 



Bigelow has shown that it extends well up into the gulf of Maine, keeping to the 

 deep water near the centre, in July and August of both 1912 and 1913 (1914, p. 121, 

 and 1915, p. 299). Our record of September, 1915, at Princess station 3, in the deep 

 water between Grand Manan and Nova Scotia, shows that it extends at least up into 

 the mouth of the Bay of Fundy. It was absent at Prince station 2 off Campobello 

 island, farther up the bay. The small number of specimens obtained at Prince station 

 3, and the lateness in the season, indicates that this is its northern limit in the bay. 



It may perhaps be possible to delimit pure Gulf Stream conditions as exhibited 

 by this species, when we take note of both size and abundance. This is shown in 

 fig. 6, horizontal lines representing the small variety and vertical lin'es the large. 



The outermost stations gave in the vertical hauls comparatively large numbers of 

 individuals, whose upper limit in length was 13mm. Taking these together with the 

 stations that show similar results and we have the following list of " Gulf Stream" 

 stations: Station 41, 200-Om., 6-9mm., twenty-five specimens; station 42, 250-Om., 

 7-llmm., sixty-seven specimens; station 44, 270-Om., 7-llmm., twenty-six specimens; 

 station 45, 270-Om., 6-llmm., thirty-two specimens ; station 56, 250-Om., 7-9mm., forty- 

 two specimens; station 74, 325-0'm., 8-llmm., thirty-five specimens; station 75- 325- 

 Om., 6-1 2mm., 25 x 4 specimens. The line separating these stations from the others 

 nearly coincides with the outer limit of specimens over 15mm. in length. 



There are several stations at which a few small individuals were obtained (station 

 40, 54, 57, and 70), but the degree of sexual maturity shown by them indicates that 

 they belong to the larger cold water form. A line drawn just inside stations 41, 45, 

 56, 74, and 75 would separate the warm water form from the cold water form. It must 

 be emphasized, however, that this line cannot be sharp, as there is a transition from 

 one form to the other. A study of the specimens from stations 44, 45, and 46 shows 

 this. At station 44 a maximum size of 12mm. is attained ; at station 45, of. 15mm. ; 

 and at station 46 of 20mm. There is the same gradation in the change from a stout 

 to a slender body and from early to late maturity. 



Its absence at Acadia stations 59 and 76, where numbers of S. elegans were 

 obtained indicates that at these points definite tongues of coastal water extend out 

 into the boreal or mixed water, which doubtless come from the mass of coastal water 

 present over and between the banks north of Sable island. These two tongues are 

 shown in fig. 6. On the earlier cruise of the Acadia these two tongues are fused as 

 shown in fig. 5, this species not ocurring at Acadia stations 6 to 10. The distribution 

 of *S'. elegans during the same cruise (fig. 8) shows the two tongues but they are close 

 together. 



On neither cruise is there any indication that the cold coastal water of the New- 

 foundland banks (where 8. serratodentata was not found) has to any extent flowed 

 out from the banks into the water outside the continental shelf, but it does appear that 

 the water north of Sable island has done so. 



Though the quantitative data as to the distribution of this species are incomplete, 

 the centre of ai)undance during the first cruise of the Acadia, as shown in fig. 5, was 



