A STUDY IN CARCmOLOGY. 57 
crabs. Most commonly the edges of the branchiostegite arc fitted closely to 
the epimera just above the ooxa3 oi: the pereiopods and to the tergum of the 
first abdominal segment, so that no water can find ingress or egress that way. 
As a consequence the podobranchs or arthrobranchs of the three posterior 
pereiopods disappear ; tliere is no longer room for them, and as the branchial 
cavity is reduced posteriorly by the great development of the muscle-cavities 
of the digging legs, the posterior pleurobranchs follow suit. The branchiae 
are reduced in number and those that remain are massed in the anterior half 
of the branchial cavitj^, nearest to the newly developed entrance of the 
respiratory current in i'ront of the chelipeds. In the Raninidse, however, 
the arrangements differ from those usually observed in crabs. There is, in 
most of them, a pair of posterior respiratory orifices situated between the 
tergum of the first abdominal segment and the coxse of the last pair of 
pereipods. When tlie abdomen is extended or only slightly flexed, water 
can pass freely into or out of these orifices, but they are pretty effectively 
closed when the abdomen is closely flexed under the thorax. The orifices 
were accurately described and figured in Ranina by Milne Edwards^ but 
many subsequent authors, particularly Ortmann (42), seem to have failed to 
recognise them. Borrodaile (14) gives an accurate account of their relations. 
These posterior branchial orifices are not, however, peculiar to the RaninidfB : 
I have found them in precisely the same position in Cori/stes, in which genus 
they can easily be seen by bending the abdomen upwards and looking between 
the lower side of the tergum of the first abdominal segment and the coxa of 
the last pair of legs. G-arstang (30) failed to observe these apertures in 
Cori/stes, and I suspect, but have not yet had the opiportunity of proving it 
by experiment, that they form the main entrance for water into the branchial 
chamber during the operation of the normal current. There is also a pair of 
posterior apertures in Thia polka, but these are in a somewhat different 
position, close under the edge of the branchiostegite in front and to the out- 
side of the articulation of the first abdominal segment withthe carapiace, and 
well in front of the coxse of the last pair of pereiopods. The persistence of a 
posterior branchial opening is a primitive feature, never to be seen, as far as 
my observations go, in crabs in which the abdomen is permanently flexed 
and kept closely applied to the sternal plastron. 
To return to the Raninidse : the posterior part of the branchial chamber, 
into which the posterior orifice opens, is reduced to a narrow passage by the 
reduction in vertical depth of the branchiostegite and the close adherence of 
its edge to the thoracic epimera. The edge of the branchiostegite is held in 
place by two prominences, one on the anterior edge of the eleventh, the 
other on the posterior edge of the thirteenth epimeron, and a close fit is 
ensured by a well-marked ridge running upwards and backwards along the 
eleventh, twelfth, and thirteenth epimera (figs. 7 & 17). Thus the entrance 
of water into or its egress from the branchial chamber at the sides of the 
