Size, as it affects the price (figure 9) of the sponge, does not 

 follow the expected pattern. Roughly, the price is directly pro- 

 portional to the diameter rather than to the cube of the diameter as 

 one would expect if the price were related to the volume of the 

 sponge, (Volume equals pi divided by 6 and multiplied by the cube of 

 the diameter o.- V = TtIP»7 Except for display purposes, most sponges 

 more than 8 inches in diameter sell slowly and, accordingly, are cut 

 into sizes that are easier to hold. The work of cutting and trimming 

 and the loss of material incurred just about offset the value of the 

 extra volume in a larger sponge. 



The curvature and the slope of the lines in figure 9 will be 

 affected by changes in the supply and demand for different sizes at 

 different times. 



Until quite recently, many sponges were sold on the basis of 

 weight. Sale by size now is recommended by the members of the Sponge 

 Exchange, and measurement of the perimeter of the sponge is preferred 

 over measurement of the maximum diameter used by many, as well as 

 over the three-diameter method suggested here. Details on the above 

 points will be discussed later in this report. 



Demerit System of Grading 



The details of inspection by the demerit system have been placed 

 under proper headings in the following section on Discussion of Faults 

 but an overall picture of the method of grading is described at this 

 point. 



Except for work in the Sponge Exchange or in packing houses, 

 most of the inspections take place after a shipped bale or box of 

 sponges has been opened. These sponges are found to be highly com- 

 pressed and should be sampled according to section F of Federal 

 Specification C-S-631b for "Sponges J Natural," which is in Part $ 

 of Section IV of the Federal Standard Stock Catalog. 



The sponges should be wet and squeezed thoroughly before being 

 inspected. The perimeters should be measured according to the Federal 

 specifications or, if the agreement requires, should be checked for 

 size by a "go - no go" test by means of standard boards with circular 

 holes decreasing in diameter in steps of one-half inch. The three 

 diameter test described later under Miscellaneous Studies may merit 

 consideration, however, since it (1) gives more data than do the "go - 

 no go" boards, (2) is quicker than are perimeter tests. (3) gives one 

 number that approximates the "go - no go" tests, and (It) causes no 

 arguments as to whether the perimeter tape was poorly placed, was too 

 loose, or was too tight. 



It might be worthwhile also to specify a minimum rate of sampling 

 for lots of different size in the manner specified on page 5 and 



Ik 



