152 
PACIFIC SCIENCE, Vol. IX, April, 1955 
longer, outer border of the peduncle clothed 
with short single hairs.” There were other 
differences — the telson of T. dorrieni was more 
spinous; the tip of the maxilliped outer plate 
had a long tapering tuft of setae and not a 
simple transverse row of short bristles as in 
T. sylvaticus; the fifth joint of the 1st gnatho- 
pod was prominently rather than minutely 
lobed, and the palp of the maxilla was "smaller 
and without a trace of a second joint.” 
Hunt has some very penetrating remarks to 
make on the "puzzling discrepancy” between 
accounts of the pleopods referred to by Cai- 
man (1912) and shows that specimens re- 
corded by Chevreux (1901) belong to a 
different species. I have examined several 
species of New Zealand terrestrial amphipods 
showing degeneration of pleopods to a marked 
extent and I believe that in these species the 
shape of the pleopods is much more constant 
than the earlier literature would lead one to 
expect. Carl (1934) and Schellenberg (1934) 
also hold this view. Nevertheless, it would 
be foolish to base specific identifications on 
the pleopods alone except where other evi- 
dence is entirely lacking, as in fragmentary 
specimens. 
Hunt notes these differences in pleopods 
between T. sylvaticus and T. dorrieni: (1) the 
peduncle in T. sylvaticus is "clothed with short 
simple hairs,” in T. dorrieni it has "long 
feathered setae;” (2) the rami in T. sylvaticus 
are not distinctly segmented, in T. dorrieni 
they are; (3) the inner ramus of T. dorrieni is 
the longer, in T. sylvaticus the shorter. The 
value of these distinctions is somewhat less- 
ened when one takes into account Barnard’s 
descriptions of the various races of Talitroides 
eastwoodae. His /. typica has "pleopods with 
rami subequal, numerous close-set plumose 
setae along the whole length of the peduncle, 
and the rami more or less distinctly sub- 
jointed.” In f. cylindripes the rami are "more 
or less distinctly unequal (the outer the long- 
er), not jointed.” In/, calva the "peduncle of 
pleopods non-setose, the rami unjointed, but 
the incisions marking the limits of the fused 
segments very deep, so that the ramus be- 
comes a series of subglobose segments.” Here 
we have all the variations necessary to link 
Sayce’s (or Caiman’s) T. sylvaticus with Hunt’s 
T. dorrieni. And also the variation in the pro- 
portions of segments 5 and 6 in gnathopods 
1 and 2 noted by Hunt and Caiman is surely 
very similar to that figured by Barnard for the 
forms of T. eastwoodae. 
Chilton’s Australian specimens also sup- 
port the opinion that T. dorrieni Hunt is 
identical with T. sylvaticus Haswell. My New 
Zealand specimens, like Hunt’s, have plumose 
setae on the peduncle. The Australian spec- 
imens, which agree with mine in all other 
respects, have short simple bristles or nothing 
at all along the peduncle. Only one specimen 
has the rami nonsegmented; it has the outer 
ramus, as in all of the Australian specimens, 
the longest; it lacks plumose setae along the 
peduncle (all Hunt’s conditions for T. syl- 
vaticus) but it agrees with T. dorrieni in both 
maxilliped and uropod 3 (of this more later). 
The Australian specimens also differ from 
Hunt’s in numbers of segments to the rami. 
In short, there seems to be wide variation in 
pleopod ornamentation. I would point out, 
though, that the essential shape of the pleo- 
pods is the same: first and second biramous, 
third one-segmented and vestigial. And, as 
far as I can see, there is no difference in pleo- 
pods in the published accounts of Haswell 
(1880), Sayce (1909), Caiman (1912), Chilton 
(1916) or Hunt (1925). 
The palp of the first maxilla in T. dorrieni 
certainly has not an obvious second segment, 
but Sayce’s "vestige of [a] second [segment]” 
could apply. 
In all the specimens I have examined, the 
number of spines on the telson varies be- 
tween 3 and 5 to each margin. This leaves 
only one point of real value — the shape of 
the maxillipeds. The form of the maxillipeds 
in terrestrial Talitridae is remarkably constant. 
The figure in Haswell’s 1880 paper is very 
poor and of no real use. However, he describes 
the plate as "ending in a single tooth.” If 
