.Iiitcrican Fisheries Socidv. 71 



flatfish collected near the Laboratory. On this diagram each of 

 the red marks represents a fisii. and the marks are arranged in 

 rows according to the nnml^er of fin-rays. Thus at the left of 

 the diagram it will be noted that one fish had only 62 diirsal 

 fin-rays, seven fishes had 63 fin-rays, twelve fishes had 64 fin- 

 rays, twenty-two (the largest number of individuals) had 65, 

 eighteen had 66, twenty-one had 67, and from there on the num- 

 ber of individuals almost constantly decrease, nine having 68. 

 six having 69, one having but 70, one iiaving 71, and two having 

 y2. A curve, then, drawn through the culminating {wints of 

 the several columns is a curve that represents, at least roughly, 

 the variation in the number of dorsal fin-rays for this specific 

 locality. The curve indicates that no matter how- many flatfish 

 may be collected at Woods Hole, specimens having less than 

 62 fin-rays will be extremely infrequent, while those having 

 slightly more than 72 fin-rays may occasionally occur. The 

 variation is about an average which lies near the column 66. 



If w'e now tabulate the fin-rays of an equal number of flat- 

 fish from another locality, it is evident that if the fishes in both 

 localities are alike, the curves will coincide. If. however, the 

 fishes are different, even slightly so, the lack of coincidence in 

 the curves will indicate the difference. 



The diagram marked "Waquoit" is based on the variation in 

 the number of the dorsal fin-rays of one hundred flatfish taken at 

 Waquoit. from a small bay only eight miles east of Woods Hole. 

 Compared with the first curve, the Waquoit curve lies further 

 to the left, has a shorter base and a less altitude. The Waquoit 

 collection contains fifteen fishes which have a less nmnber of 

 fin-rays than any fish collected at Woods Hole, a striking differ- 

 ence when one considers the small number of fish examined. 

 Moreover, the right side of the Waquoit curve is almost equally 

 characteristic, and the average number of fin-rays in the W^aquoit 

 fish is very evidently less than the average numl)er at Woods 

 Hole. The Waquoit fish are more variable, the ampliture at 

 Woods Hole being 62 to 72 (11 points), while the ampliture at 

 W'aquoit is from 60 to 71 (12 points). 



These curves of distribution bring out certain characters 

 that it would be quite impracticable for one to detect bv the 

 mere examination of a few representative fish, and it w^ould be 

 quite possible for one to decide by such curves w'hich of two 

 baskets of fish come from Woods Hole and which from Wa- 

 quoit. even though the fish bore no other mark than that pro- 

 vided bv nature. 



