124 BULLETIN OF THE BUEEAU OF FISHEBIES 



Alexander Agassiz's (1866) observation that Limacina often sinks to the bottom 

 suggested to him, and to other students subsequently, that this habit may explain 

 its sudden appearances and disappearances — that is, that it may endure unfavorable 

 periods on the bottom, where salinity would always be sufficiently high for its existence 

 in all parts of the gulf except in very shallow water. However, since this habit has 

 not been observed in European waters, where L. retroversa is often far more abundant 

 than we have ever found it in the Gulf of Maine, probably its disappearance from 

 the coast water reflects either the death of the local stock or a migration out to sea, 

 its reappearance there reflecting an actual immigration from offshore in toward land, 

 which follows more or less closely on the reestablishment of a favorable environment 

 in the coast water and depends on the precise distribution of Limacina at the time 

 relative to the circulation in the central parts of the gulf. 



The upper limit of salinity for Limacina is certainly as high as 36 per mille 

 (35.9 per mille is the most saline water in which I find it actually recorded), and 

 inasmuch as it thrives in water of 34 to 35 per mille in the North Sea region no part 

 of the Gulf of Maine could ever be too salty to afford it a favorable environment. 



Nothing is known of the reproduction of L. retroversa in the Guff of Maine except 

 that young as well as old individuals have been taken repeatedly in spring, summer, 

 autumn, and winter, proving it endemic. Very little information is as yet available 

 as to the actual numbers in which L. retroversa occurs in the gulf, and comparison of 

 the catches of the horizontal nets with those of the verticals shows that whether it 

 be scarce or plentiful, it is so prone to congregate in shoals (which one net may hit 

 but the other miss) that it would take a great number of vertical hauls to yield even 

 an approximation of its actual numerical strength over any considerable area of the 

 sea. For example, the vertical haul from 70 meters yielded none at all at the station 

 where we made our largest catch in the horizontal net (station 10215, northwest part 

 of Georges Bank, 125 cubic centimeters of Limacina in a 50-meter haul of one-half 

 hour's duration). An instance of the opposite sort is afforded by a station in the 

 center of the gulf (March 2, 1920, station 20052), where the quantitative haul yielded 

 enough (58 specimens) to indicate comparative abundance (theoretically 240 Limacina 

 under each square meter of the sea's surface), whereas the surface haul jdelded only a 

 few dozen individuals, the horizontal net, working at 100 meters, none at all, and the 

 closing net only a few at 160 meters. Instances of this sort, which might be multi- 

 plied, make any attempt to plot its actual numbers from the data yet in hand not 

 only idle but apt to prove misleading. However, it can be stated as a general propo- 

 sition that only on the rarest occasions does L. retroversa form any considerable pro- 

 portion of the plankton in any part of the gulf, judged either by numbers of individ- 

 uals or by built. 65 Nor have we ever found it in abundance to compare with the 

 shoals recorded by Paulsen (1910) from the waters south and west of Iceland. There- 

 fore, it is not likely that this pteropod is ever of as much importance as pasturage for 

 the pelagic fishes in the Gulf of Maine as it is in Irish waters, for instance, where, 

 says Massy (1909), it regularly serves as an important item in the diet of both mack- 

 erel and herring. 



w The richest catches of Limacina are noted above (p. H9). 



