FISHERY BULLETIN: VOL. 72, NO. 3 



The number of spines in the ventral comblike 

 row on the inner uropod first suggested that 

 northern populations might have fewer spines 

 than those in the south. This proves not to be true; 

 rather, the number of spines is apparently a func- 

 tion of body size (Figure 1). A regression analysis 

 shows that the relationship is nonlinear {Y = 

 3.39X + 6.08, r = 0.579, z - 0.661, P>0.05), 

 falling short of statistical significance. Inspection 

 of the scatter of points indicates that number of 

 spines more or less levels off at upper size limits, 

 but distribution is fairly broad at all sizes. 



Two other relationships do yield statistically 

 significant correlations but show no geographic 

 association. Length of the antennal scale is cor- 

 related with its width (Figure 2) {Y = 9.726X 

 - 0.177, r = 0.967,2 = 2.043, P<0.05) and with 

 carapace length (Figure 3) (Y = 0.813X - 0.264, 

 r = 0.822,2 = 1.162, P<0.05). 



Other plots analyzed but not discussed in detail 

 here show similar relationships which further 

 substantiate the facts given above: length of comb- 

 like row of spines on inner uropod plotted against 

 length of carapace; length of comblike row of 

 spines on inner uropod plotted against longest 

 spine in row; length of inner uropod plotted 

 against length of carapace; number of spines per 

 0.01 mm in comblike row of spines on inner 

 uropod plotted against length of carapace. Size of 

 cornea, shape of rostrum, shape of antennal scale, 

 spination of telson, and relative lengths of uropods 

 on specimens from throughout the range were 

 compared. No constant differences were noted 

 that would define geographic races. 



Supplemental Description 



Despite its abundance, A/^eom_ysis americana has 

 not been described and illustrated fully. The 

 reason for this omission is obvious: of the 16 

 known species of Neomysis, it is the only one that 

 occurs in the western North Atlantic and is not 

 likely to be confused with any other mysid within 

 its range. Three mysid genera are similar to 

 Neomysis: Acanthomysis, Paracanthomysis , and 

 Proneomysis. The last two are confined to the 

 North Pacific, and the only Atlantic representa- 

 tive of the 13 known species of Acanthomysis, 

 A. longicornis (Milne Edwards), is limited to 

 European waters. 



However, the lack of an adequate description 

 of N. americana gave li ( 1964) some difficulty 

 before he decided not to identify a Neomysis from 



Korea asN. americana but to describe it as a new 

 species, N. orientalis. To obviate problems such as 

 this, we offer Figures 4A-K and 5A-J and com- 

 ments on some characters for which adequate 

 illustrations are not available. 



The rostrum (Figure 4A) is broadly rounded as 

 inA^. intermedia and N.japonica. The eyes are as 

 described by W. M. Tattersall (1951), with a broad 

 cornea occupying half the eyestalk. Medially, just 

 before the eyestalk narrows, it is produced into a 

 low protuberance armed with short setae. W. M. 

 Tattersall described the telson (Figure 4K) as 

 having about 40 spines on each lateral margin; the 

 number of spines depends on the body size and 

 ranges from about 20 to 40. The peduncle of 

 antenna 1 is shown in Figure 4B, C, D, the latter 

 showing the male lobe. W. M. Tattersall's (1951) 

 Figure 77 A shows the antennal scale without a 

 suture, which would be unique for the genus. Our 

 specimens have a distinct suture (Figure 4D) 

 setting off a short distal segment. As in other 

 species of Neomysis the labrum is produced an- 

 teriorly into a spiniform process (Figure 4E, F), 

 and the terminal segment of the mandibular palp 

 (Figure 4G) is relatively slender. 



According to W. M. Tattersall (1951) the 

 "tarsus" of pereopods 2-7 is 8-9 segmented. We 

 find that the number of segments increases with 

 body size (Figure 5A, C) as inN. patagona (O. S. 

 Tattersall, 1955b; Holmquist, 1957). The proxi- 

 mal suture may be incomplete, not present medi- 

 ally; in such pereopods an additional tarsal seg- 

 ment would be counted when viewed laterally. 

 Our counts are in lateral view and include both 

 partly fused proximal segments. Small juveniles 

 have 6 segments; small adults from Pamlico 

 Sound, N.C., 7 segments; large adults from Woods 

 Hole, Mass., 8 segments; and large adults from 

 the York River, Va., 9 segments. The number 

 appears to be rather constant among the pereo- 

 pods of an individual, but may differ by 1 segment 

 in one or two of the pereopods. 



The genns Neomysis is unique in having median 

 fingerlike papillae on the last two or three 

 pereonal sterna of gravid females (W. M. Tatter- 

 sall, 1932). W. M. Tattersall (1951) found these 

 papillae on the last two sternae of A'^. americana, 

 and we show them in Figure 4J. Their function 

 is unknown. Other characters of Neomysis are 

 the presence of a bailing lobe on the posterior 

 margin of the oostegite of pereopod 6 (Figure 

 5E) and a rudimentary oostegite on pereopod 5 

 (Figure 5D). 



838 



