AVERAGE NUMBER OF DEMERITS CHARACTERISTIC OF EACH TYPE 

 AND GRADE OF SPONGE 



■When the system of demerit grading described in the preceding 

 section is applied to sponges, the average number of demerits assigned 

 to a lot varies both according to the type of sponge and to the grade 

 of sponge. With Rock Island Sheepswool sponges, for example. No. 1 

 Forms will average 50 demerits and No. 3 Forms -will average l\2$ de- 

 merits. On the other hand, with Mediterranean Deepwater sponges, 

 No. 1 Forms will average 350 demerits and No. 3 Forms will average 

 630 demerits. Thus, the average number of demerits varies according 

 to both the "type and the grade of sponge under consideration. 



In practice, we find that the number of demerits assigned to an 

 individual sponge varies widely from the average for its type and pur- 

 ported grade. The question naturally arises as to what is a reason- 

 able variation. It is suggested that a good basis of judgment would 

 be to consider the magnitude of the variation in relationship to the 

 midpoint between the average number of demerits characteristic of the 

 purported grade and the average number characteristic of the next grade. 



The fact that the number of demerits assigned to a particular 

 sponge deviates widely from the average for its grade shows the need 

 for careful sampling in the grading of sponges. 



In the event that the demerits assigned to individual sponges in 

 a lot are found to deviate too widely from the average for the pur- 

 ported grade of the lot, there are two possible solutions to the prob- 

 lem: (1) regrade the individual sponges or (2) assign a different 

 grade to the lot as a whole. In either case, the basis for reassign- 

 ment of grade could be the midpoint between the average number of de- 

 merits characteristic of the purported grade and the average number 

 characteristic of the next grade. 



It thus becomes important, in the demerit system of grading, 

 accurately to determine the average number of demerits characteristic 

 for each type and grade of sponge and the midpoints between these 

 characteristic numbers. 



Accordingly, the various types of sponges were graded by the 

 demerit system in order that the characteristic number of demerits 

 for each type and grade could be determined. The results are reported 

 in the following subsections. 



Rock Island 



Table 3 gives the average number of demerits characteristic of 

 each grade of Rock Island Sheepswool sponge. Inasmuch as the number 

 of demerits found by actual grading will fluctuate, depending on the 

 lot of sponges and upon the grader, this number is subject to some 

 variation. Accordingly, since round numbers are more convenient 



27 



