Munroe: Systematics of western Atlantic Symphurus 
85 
men are unreliable, occurrence of S. parvus in the West 
Indies, based on this particular specimen, cannot be 
unequivocally verified. Also, because S. parvus is most 
often found on silty substrates in relatively deep ner- 
itic waters ( see below), its occurrence in the West Indies 
would be quite restricted owing to absence of suitable 
soft mud and silt substrates (calcareous sediments are 
more widespread throughout this region) at depths 
where S. parvus usually occurs. 
Bathymetric distribution Symphurus parvus occurs 
over a wide range of depths from 20 to 146 m; one 
unusual deepwater capture of a single specimen 
(UMML 35263) was made at 383 m (Table 10). Most 
specimens ( 107/1 14, 94%), however, have been collected 
between 31 and 110 m. Only five specimens have been 
taken at depths shallower than 30 m, whereas eight 
others were trawled at depths exceeding 100 m. Topp 
and Hoff (1972) reported similar capture depths (37 — 
109 m) on the west Florida shelf for this species. 
Although S. parvus has been collected from depths 
greater than 110 m, such deepwater occurrences are 
apparently unusual. The deepest recorded captures 
are for single specimens collected at 146 m and 383 
m, whereas the next deepest captures are for six 
specimens taken between 101 and 110 m. The speci- 
men captured at 383 m (UMML 35263) is undoubt- 
edly S. parvus , despite a depth of capture consider- 
ably beyond the depth range known for the other 
specimens. Along with S. parvus, specimens of 
Enchelyopus cimbrius, Chaunax pictus, Dibranchus 
atlanticus, Malacocephalus occidentalis, Bembrops 
sp., and Coelorhynchus carminatus, deepwater spe- 
cies occurring to the edge of the continental shelf and 
on the upper continental slope, were also taken in 
the same trawl collection at this station. 
Ecology Little is known concerning the ecology of 
S. parvus. Topp and Hoff (1972) collected this spe- 
cies on the west Florida shelf over a temperature 
range of 18.8-24°C and salinities of 33.8— 36.3%o. They 
also remarked that S. parvus was not abundant any- 
where, because most of their collections were of soli- 
tary individuals as were most collections (51/70) ex- 
amined in this study. Eleven other collections con- 
tained two specimens each, whereas nine additional 
lots contained three to 16 specimens. The rarity of 
this species, based on trawl collections, may reflect 
the general inefficiency of trawls for capturing small 
pleuronectiform fishes, especially those that inhabit 
relatively deep waters, as does S. parvus. 
Remarks Ginsburg (1951) noted that specimens 
from Albatross Station 2318 originally designated by 
Goode and Bean (1886) as “collateral types” (=para- 
types in current usage) of A. pigra were actually speci- 
mens of S. parvus. Ginsburg selected three of these 
(USNM 74330) as paratypes of S. parvus. The fourth 
specimen, contained in the same bottle with the oth- 
ers, was not designated a paratype of S. parvus be- 
cause it was on loan to P. Chabanaud (MNHN) and 
was not included in the original description of S. 
parvus. This specimen has been recatalogued as 
USNM 342239. 
In examining specimens on which the original de- 
scription of A. pigra was based, Ginsburg (1951) 
noted that he could not find specimens from Alba- 
tross station 2405. However, in his description of S. 
parvus, Ginsburg listed six specimens from Albatross 
station 2318. These specimens are now catalogued 
with the following numbers: USNM 84491 one speci- 
men; USNM 152733 one specimen; USNM 74330 
three specimens; and USNM 342239 one specimen. 
Because only four specimens were originally collected 
from Albatross station 2318 (as indicated by Goode 
and Bean, 1886), possibly USNM 84491 and USNM 
152733, now catalogued separately (the specimens 
were originally in the same jar), may represent the 
two paratypes of A. pigra from Albatross station 2405. 
Examination of labels contained with the specimens 
and museum registers at the USNM could not resolve 
this problem. It appears that if labels with the speci- 
mens were mixed-up, or collection data for the speci- 
mens were mislabelled, it happened before the lot be- 
came part of the USNM collection. 
If USNM 84491 (the holotype of S. parvus) and 
USNM 152733 (a paratype of S. parvus) are indeed 
the specimens from Albatross Station 2405 included 
in Goode and Bean’s description of A. pigra (instead 
of Albatross station 2318, as presently labelled), then 
type locality and collection information of S. parvus 
changes from Albatross Station 2318 (24°25'45"N, 
81°46'W; 45 fm; 15 Jan 1885) to Albatross Station 
2405 (28°45'N, 85°02'W; 30 fm; 15 Mar 1885). 
Among western Atlantic tonguefishes, S. parvus 
is most unusual in that two patterns of interdigita- 
tion of dorsal-fin pterygiophores were found in nearly 
equal frequencies in the specimens examined. Thirty- 
three of 82 (40%) specimens had a 1-4-2 pattern, 
whereas 35/82 (43%) had a 1-5-2 pattern. No corre- 
lations between geography, sex, or size and the ob- 
served bimodal distribution of pattern types were 
found. Symphurus parvus is one of only three spe- 
cies in the genus in which two ID patterns occurred 
in nearly equal frequencies (Munroe, 1992). 
Comparisons Of all congeners, S. parvus is most 
similar to the western South Atlantic <S. kyaro- 
pterygium and the western North Atlantic S. minor. 
Symphurus parvus can be distinguished from both 
