70 IOWA STUDIES IN NATURAL HISTORY 
ply in the branchial chamber without visiting the shore at all. 
This may, in part, explain the presence of functional gills. In 
any event it seems reasonably sure that the respiratory tufts of 
villi are the principal organs of respiration. 
This system is evidently derived from that of the hermit crab. 
A comparison with Canobita diogenes, a form which is almost as 
terrestrial as Birgus, shows many points of similarity,” among 
which the following may be noted: The gills are of the phyllo- 
branch type and the lamelle are thicker and less numerous than 
in water breathing forms. The number of gills is reduced and 
the individual gills are terete in both species. Similar tufts of 
hairs are found on the abdominal ventral region of Canobita, 
greatly resembling those in the lower branchial cavity of Birgus, 
and channels for blood vessels are found in the lining of the 
branchiostegite of both forms. The most striking difference is in 
the presence of the tufted villi in Birgus while they are absent in 
Cenobita. Thus not only the unsymmetrically placed abdominal 
appendages of the female of Birgus, but also the general gill 
structure, tell the tale of the descent of the robber crabs from 
the hermit crabs. 
Gonodactylus chiragra, one of the mantis shrimps, is quite com- 
mon on the flats of Makuluva where it is very active. This 
species has conspicuous black eyes and some of the mouth parts 
are tipped with scarlet. It resembles the pistol crab in its habit 
of making a clicking noise when confined in a bottle or glass jar 
and sometimes thumping the fingers of its captor. As in the 
common mantis shrimp, the second maxilliped is used as a chela 
but is built on a different plan from that of the lobster, for in- 
stance. It has a remarkably heavy and solid basal joint. The 
pincer is formed by the backward flexing of the distal joint of 
the appendage so that it opposes the anterior border of the pre- 
ceding joint. The inner edge of the flexed finger is very finely 
serrate, and the teeth are much smaller than in the common man- 
tis. Moreover, the abdominal ‘‘hump’’ so conspicuous in the 
mantis is not evident in the Gonodactylus. Back of the hand 
there are three pairs of slender walking legs and six pairs of 
swimmerets, including that of the tail fin. These are very broad 
7I am using in this connection an unpublished thesis, ‘‘ Modification of 
Respiratory Structures involved in transitions from Aquatic to Terrestrial 
Crustacea’’, by Otto Walter, a former graduate student in the Department 
of Zoology, State University of Iowa. 
