BIGELOW: EXPLORATIONS OF THE COAST WATERS. 249 



subchordal cells. Its very large size (rump-length upwards of 4 mm.) 

 is likewise fliagnostic, while tlie red margin of the tail makes it a 

 conspicuous object in the water. 



None of these Arctic forms were encountered in the Gulf on our 

 later cruises (June-October), their absence coinciding with the 

 shrinkage of the Cabot Current (p. 242). But they were all present 

 in June across the whole breadth of the shelf off Shelburne (]\Ier- 

 tensia, Station 10291, 10294: Oikopleura vanhoffeni, Station 10291, 

 10293: Limacina helicina, Station 10295). 



Our records indicate a further shrinkage in the southward and 

 westward extension of the Arctic faunal community, as the summer 

 advances, for it hardly extended beyond Halifax in August, 1914 

 {Limacina helicina, Station 10236, 10237; Mertensia, Station 10236). 



Previous records of Arctic organisms in the Gulf of Maine are too 

 scanty to show whether there is an annual spring invasion from the 

 Cabot Current, as was certainly the case in 1915. But judging from 

 the hydrography of the Gulf in general and of the northern and south- 

 ern currents in particular, this is probable. The mere fact that 

 neither Mertensia, the Arctic Oikopleura nor Limacina helicina were 

 found in the Gulf in the summers of 1912, 1913, or 1914, argues 

 nothing against this view, for they were equally wanting there in the 

 summer of 1915. And experience shows that Mertensia (A. Agassiz, 

 1865), apparently also Limacina, is so delicate, and so rapidly killed 

 by unduly high temperature, that its absence at any particular time 

 has no necessary bearing on conditions one or two months previous. 

 The Grampus records do not indicate any greater invasion of Arctic 

 animals in 1912, 1913, or 1914, than in 1915, though the occasional 

 appearance of Mertensia in abundance at Eastport in September 

 (A. Agassiz, 18Q5, p. 29; Fewkes, 1888), and of Mertensia and Ptycho- 

 gena in ^Massachusetts Bay in the autumn (A. Agassiz, 1865) suggests 

 that they are occasionally more important in the economy of the Gulf 

 than we have found them in recent years. Unfortunately we have 

 no records of the Gulf plankton for the year 1884, interesting because 

 of its low temperature (1915, p. 243). 



The Grampus hauls of 1915 demonstrate that, as a general rule, the 

 plankton at any given part of the Gulf changes very little in its com- 

 position during the summer. Perhaps the most striking example of 

 the permanence of the general plankton type is afforded by German 

 Bank, where a swarm of Pleurobrachia (p. 306) prevailed, almost to 

 the exclusion of other forms, from May (Station 10271) to September 

 (Station 10290, 10311); and where we have usually found Pleuro- 



