FIJI-NEW ZEALAND EXPEDITION 69 
tensively broken by a series of somewhat deeply incised channels 
forming a rudely reticulate pattern. On the inner surface of the 
branchiostegite is a similar network of what appears to be large 
blood vessels. Inside of this branchiostegite is the highly vascular 
lining of the pulmonary chamber which bears the profusely 
branched bunches of villi. The outer surface of this membrane 
bears complete impressions of the reticulate pattern of blood ves- 
sels between the plates. Dorsally these impressions combine to 
make a very large one running obliquely from the antero-lateral 
edge of the branchiostegite to the posterior border where it joins 
a similar impression from that part of the membrane lining the 
gill chamber proper. The impression made by the combined ves- 
sels then turns abruptly inward and forward and seems to enter 
the body cavity a little to one side of the middle: of the line of 
juncture between the carapace and the first abdominal segment. 
The hairy membrane covers the dorsal and lateral surfaces of the 
gill chamber. The number of gills seems to be eleven in both 
sexes and they lie longitudinally in the branchial chamber. 
The condition of the viscera and other contents of the body 
cavity is such as to preclude investigation of the heart and main 
vessels. 
It is evident that we have in Birgus a form more highly special- 
ized for breathing in air than is found in any other decapod crus- 
tacean. The quantity of highly branched and vascular villi and 
the area that they cover is much greater than in any of the land 
crabs of the West Indies, where gills are still the main organs of 
respiration. In Birgus the gills can by no means be described as 
rudimentary. Although relatively small, they bear every struc- 
tural evidence of being functional. The hairs covering the lining 
of the cavity containing the gills are excellently well adapted to 
retain whatever moisture there is in the branchial chamber and 
thus create a condition in which the gill lamelle can use the 
oxygen in the air even if there is no water in the bottom of the 
gill chamber. Doubtless the soft abdomen aids in respiration, as 
its lower surface is supplied with hairs resembling those in the 
lower branchial chamber. 
I have been unable to learn whether the holes or burrows of 
the robber crab are deep enough to reach water. In the South 
Sea islands the general level of the lowlands, especially near 
beaches, is not far above the water level, and it seems likely that 
Birgus may have convenient access to water to replenish the sup- 
