52 BULLETIN OF THE BUREAU OF FISHERIES. 



miles from Marthas Vineyard. The tides, likewise, are of much greater amplitude 

 north of Cape Cod, insuring a far more rapid intermingling of the coastal waters with 

 those of the open sea. South of Cape Cod there is an extensive area of shoal water, 

 much of which is pretty definitely bounded off from the open ocean. Reference has 

 already been made to the occurrence of a net westerly tidal movement through Vine- 

 yard Sound. This implies, of course, that the latter derives much of its water from 

 Nantucket Sound, a broad and on the whole very shallow area of sea, pretty well shut 

 in by land and by shoals. 



5. SALINITY. 



Salinity or, more properly, density determinations were made along with those for 

 temperature. The Sigsbee water cup was employed for obtaining samples from the 

 bottom, while the surface water was merely dra\vn up in a pail. The salinometers 

 employed were of the Hilgard pattern and were previously tested by the Bureau of 

 Standards. Great care was taken to prevent the soiling of the stem by the hands, 

 which was found to exert a marked effect upon the level reached by the instrument. 

 A bottle of caustic soda solution, or a mixture of sulphuric acid and potassium bichro- 

 mate, was kept at hand, and used from time to time for cleaning the stem. It was 

 found more practicable to read from the summit of the meniscus, or cone of fluid sur- 

 rounding the salinometer stem, than to read from the actual water level. The value 

 of the meniscus in terms of the scale was later determined. Since the temperature of 

 the water is an all important factor in determining its specific gravity, as referred to 

 distilled water at maximum density, careful record was kept of the water temperature 

 at the time of taking the reading for density. Knowing these two factors, reduction 

 was easily accompUshed by the aid of a table furnished by the Bureau of Standards.'' 



The figures, as presented, represent the specific gravities which would have been 

 obtained had the water samples in all cases been at a temperature of 15° C. Thus each 

 figure represents the relative weight of a given sample at 15° C. compared with an equal 

 volume of distilled water at 4° C. The density of a solution depends, of course, upon 

 two factors, its temperature and its concentration. Having eliminated all differences 

 due to the former factor, the figures here given represent the concentration, i. e., the 

 salinity of the water. 



The density readings here recorded were in nearly all cases made aboard ship. 

 More precise determinations would of course have been possible if the water samples 

 had been bottled and brought back to the laboratory where the ship's motion would not 

 have disturbed the obser^'ations.* And our results would have been still more precise 

 had we resorted to the method of titration \nth nitrate of silver, as employed in recent 

 hydrographic studies.'' The latter method has, however, been used by us as a check 

 upon our specific gravity determinations, and the results of the two accord so well on 

 the whole (see p. 54) that the figures here presented are probably exact enough to meet 

 the demands of the present work. Our figures for density are recorded to the fotirth 

 decimal place. From comparison with the chlorine tests it seems likely that in 



a Various tables of this sort have been published; e. g., Libbey. 1S91, p. 397; Tanner, 1S97, p. 337. 



>> In five cases, in which this was done, and the results of the two independent determinations were compared, a mean differ- 

 ence of 0.00024 was found; i. e.. the error affected only the fourth decimal place, or last one considered in making the reading, 



"See Pettersson, 1S94, p. 296; also account of International Conference for the Exploration of the Sea. in Journal of the Marine 

 Biological Association, vol. vi, pp. 101-114. 



