Alpheid Shrimp, X: Fiji, Tonga, Samoa — Banner and Banner 
171 
0.65 in the figure (a little shorter than the 
palm); first two carpal articles described as 
"subequal” and shown to have the second 
article about 1.5 times the length of the first 
(the second article varying from 1.1 to 1.6 
times the length of the first); walking legs 
lacking from type (but described by de Man 
to be like those of A. obesomanus); telson 
shown to be about 3 times as long as the 
posterior margin is wide (4 times as long). 
De Man accounted for the difference by saying 
that his specimens were smaller and were not 
broken. Moreover, Bate’s descriptions and 
figures were inaccurate. 
We believe that inasmuch as Bate’s type 
specimen was defective, as his descriptions 
were inadequate and did not match his draw- 
ings, and as de Man’s excellent figures and 
descriptions agree with those of Bate on many 
points, the characteristics of this species should 
be established on the basis of the redescrip- 
tion by de Man. In any case, the species is 
well separated from those placed in Group I 
above by the length of the scaphocerite and 
by the length relationship of the first and 
second carpal articles of the second leg. The 
name A. malleo digitus is the oldest name in 
Group B. 
2. A. malleodigitus var. gracilicarpus de Man. 
This variety was separated by de Man solely 
on the basis of the length: breadth relationship 
of the articles of the second and third legs, 
as given in Table 3, using figures for both 
the parent species and the variety as given 
by de Man (1911:348). 
The only appendages upon which length: 
breadth relations were taken in our study were 
the second antennular article and the merus 
and carpus of the third legs. These show that 
length: breadth ratios are very variable. The 
graph on the merus of A. malleodigitus (Fig. 
17) shows what might have developed into a 
bimodal curve, had hundreds of specimens been 
measured, but there is no reason to believe 
that those with the more slender appendages 
would have shown a valid specific separation 
from those with the heavier. Therefore we 
consider this variety to be merely an extreme 
found in a normal population. 
TABLE 3 
Length : Breadth Ratio of Articles 
ARTICLE 
A. malleo- 
digitus 
A. m. gracili- 
carpus 
Second leg, 
fourth carpal article 
2. 4-3 .9 
5. 0-5. 3 
Third leg, 
merus 
3. 7-4.0 
4.5 
carpus 
4.4-4. 8 
5.6 
3. A. phyrgianus Coutiere. This species was 
characterized by having the second antennular 
article 3 times the length of the first, about 
4 times as long as broad; the carpocerite reach- 
ing to the end of the second antennular article; 
the scaphocerite to the middle of the same 
article, with the squamous portion only one- 
third the length of this article; the sculpturing 
of the large chela slight; the palm of the small 
chela about 1.2 times the length of the fingers; 
the second carpal article of the second legs 
varying from 1.1 to 1.6 times the length of 
the first; the length of the merus of the third 
leg 4 times its breadth, and feebly armed; and 
the telson 4.5 times as long as broad. 
The actual differences between this species 
and A. malleodigitus lie in the slight differ- 
ences in proportions of the appendages, which 
are now shown to be variable; therefore, A. 
phyrgianus must be considered a synonym of 
A. malleodigitus. 
4. A. persicus Nobili. In this species the 
second antennular article was described as being 
3 times as long as the first, 3 times as long 
as broad; the carpocerite reaching almost to 
the end of the third antennular article; the 
scaphocerite reaching to the end of the second 
antennular article, with the squamous portion 
to the middle of the same article; the chelae 
"concerdent avec celles de phyrgianus”; the 
second carpal article of the second legs being 
1.1 times as long as the first (obviously there 
is a typographical error in Nobili’s paper 
[1906:34], as it reads "1.31” instead of 1.13); 
the third legs being similar to those of A. 
