SEA-FISHERIES LABORATORY. 151 



purpose. Fastidious method is beside the point in an investi- 

 gation such as this, where the errors of sampling may be very 

 appreciable. 

 Proteid Estimations. 



Kjeldhal's method was used in estimating proteid. The 

 amount of dried mussel powder taken was 0-5 grammes. 

 This was placed in a long-necked flask, with 20 c.c. of strong 

 sulphuric acid and a little sand added to stop bumping. The 

 flask was then heated cautiously until the frothing of the 

 material subsided. After cooling, a teaspoonful of sodium 

 sulphate was added to increase the temperature of boiling, 

 and a small piece of copper sulphate to assist in the oxidation. 

 The flask was then heated vigorously for an hour or more 

 until the contained liquid became a clear green. After diluting 

 the liquid with distilled water, it was added to an excess of 

 caustic soda and distilled, the ammonia evolved being collected 

 in 75 c.c. of decinormal sulphuric acid. The titration was 

 carried out with decinormal sodium hydrate, cochineal being 

 used as an indicator. . 



Two or three samples were estimated at one time, a blank 

 experiment having first been made to calculate the error. 



In order to obtain the percentages of proteid given below 

 the Kjeldhal nitrogen values have been multiplied by the 

 factor 6-25. This number has been used quite provisionally. 

 There is evidence from experiment that the factor ought to be 

 higher so far as the mussel is concerned, but since this part of 

 the work is incomplete, the more commonly accepted factor 

 has been used in the meantime, The proteid values given at 

 any rate show the tendency of the changes throughout the 

 season. 



Estimation of Carbohydrates. 



The large quantities of glycogen found in the mussel render 

 it highly desirable to obtain some method of estimation which 

 will give a sufficient degree of accuracy. 



