Alpheid Shrimp, V — Banner and Banner 
gobranch was on the second maxilliped and the 
last setobranch (at least in those specimens 
where it was large enough to be seen) was on 
the third maxilliped. The two specimens with 
the more complete set did not differ in any 
other way from the specimens with the reduced 
branchial formula. Other specimens from Can- 
ton and from other localities were spot checked; 
most were found with the reduced formula, but 
occasional individuals carried the more complete 
set. Thus the more common branchial formula 
in Athanas djiboutensis is the same as that for 
Arete. 
Athanas verrucosa Banner, described below, 
is definitely related to the species known as 
Arete borradailei and Arete ghardaqensis in the 
form of the rostrum, in the form of the large 
chela, which is unlike those found in other spe- 
cies of either Athanas or Arete, and in the gen- 
eral configuration of the body and appendages. 
In fact, even the specific distinctions between 
these nominal species is in doubt. However, 
the second legs of Arete borradailei and Arete 
ghardaqensis have the four- jointed carpus while 
Athanas verrucosa a five-jointed carpus. It would 
be unacceptable to place such closely related 
species in separate genera. 
The distinctions, therefore, between Athanas 
and Arete have been entirely lost, and Arete 
must be regarded as a synonym of Athanas. 
Further supporting the combination of the two 
genera is the fact that species in both genera 
show parallel variation in the form of the large 
chela with sex and maturity. 
Athanas 
redefinition: Alpheidae in which the ros- 
trum is well developed; supra-, extra- and in- 
fracorneal spines usually present with varying 
degrees of development; corneas of eyes exposed 
anteriorly and largely dorsally and laterally; 
chelae of the first legs well developed, either 
carried extended or flexed against an expanded 
merus, usually asymmetric and sexually dimor- 
phic, always without the cylinder and plunger 
characteristic of Alpheus; carpus of second legs 
with four or five articles; dactylus of third legs 
simple or bifid; pleura of sixth abdominal seg- 
ment articulated; telson without anal tubercles, 
tip arcuate. Branchial formula variable. 
135 
DIAGNOSIS: The exposure of the eyes and the 
form of the large chela separate this genus from 
Alpheus, Thunor, and Synalpheus; the articula- 
tion of the sixth pleura from Pterocaris, Auto- 
mate, Salmoneus, Amphibetaeus, Betaeus (in 
part), Batella, Metabetaeus, Racilius, and Po- 
rno gnat hus. The presence of a rostrum separates 
the genus from Betaeus and Parabetaeus, and the 
exposure of the eyes in dorsal view from Alphe- 
opsis, Neoalpheopsis and Athanopsis. Finally the 
lack of high dorsal and ventral keel on the 
rostrum separates this genus from Aretopsis. 
TYPE SPECIES: Palaemon nitescens Leach, 
1814 - 
variation IN FORM 
Because previous workers usually had only a 
few specimens of each nominal species, most 
were unaware of the variation that might occur. 
In the present collections the following species 
are present in great enough numbers to at least 
tentatively delimit the extent of variation: A. 
areteformis Coutiere, A. parvus de Man, A. dji- 
boutensis Coutiere, A. marshallensis Chace, A. 
rhotionastes Banner, A. dorsalis (Stimpson), 
and A. indicus (Coutiere). Of these only one 
species, A. djiboutensis, the most widespread 
species both ecologically and geographically, 
seems to show no marked variation except in 
the branchial formula, as discussed above, and 
in the chela. In specimens of this species the 
form of the rostrum, the teeth about the orbits, 
and the legs are quite constant. The chelae do 
show changes in proportion with maturity and 
sex, but these differences are more slight than in 
other species. The other species all show marked 
variation. It is the chelae that have the greatest 
variation. For example, in the two closely re- 
lated species, A. marshallensis and A. rhothi- 
onastes, the chelae in the females start out as 
slender appendages where the merus is not ex- 
panded, the palm scarcely broader than the 
merus, the carpus almost as long as the palm; 
in this condition the fingers are straight and 
unarmed ( Fig. 2 ) . As the individual grows 
larger the merus expands to accommodate the 
chela; the palm grows heavy, the fingers develop 
strong teeth, the carpus either remains the 
same length, shortening in relation to adjacent 
articles, or becomes shorter by actual measure- 
