56 DE. G. C. BOURNE ON THE RANINID^ : 
noiiniil or reduced in size ; sternal can:il present ; thoracic nerve ganglion- 
chain elongate ; antennary sterimm triangular, spout-shaped ; branchise 8 on 
each side Family : Raninidw. 
We may now deal with sonic features of general and special interest 
presented by the tribe Gyninopleura. It has been shown to have originated, 
independently of other " crabs,'' from the Astacura. It is but a small 
assemblage of animals, comprising some eight or nine living genera, 
exhibiting a very small range of variation, and evidently very well adapted 
to their environment, for the tribe emerges in the Cretaceous with its s|!ecial 
adaptive characters fully established, and two of the recent genera date back 
to the Eocene. But although not descended from crabs, but from lobsters, 
the Raninidse have been so similarly modified, in one direction and another, 
_that they must be classed with the crabs with which they have no relation- 
ship save that of a remote common ancestor which was not itself a crab. 
It is by no means an isolated phenomenon. We know of several crab-like 
forms, Porcellana, LitJiodea, I:/ippa, and the interesting P orcellanopagurus, 
whose assumption of a crab-like form has been described in a very lively 
manner by Borrodaile (30), which have undoubtedly been derived, quite 
independently in each individual case, from Macrurous ancestors. The 
question is, what causes have operated to produce so great similarities in 
animals so remotely related to one another ? Without doubt, I think, the 
answer is that tlie efficient cause has been the assumption of the habit of 
burrowing in sand or mud. Bohn (7) has shown that various members of 
the Nephropsidea, when living on sandy bottoms, protect themselves bj' an 
investment of sand, and to a limited extent take cover in it. But their 
pereiopods are ill-adapted for digging ; they must retreat backwards into the 
sand to leave their antennse and eyes free to keep watch against enemies, 
and the elongnted abdomen is a great obstacle to the effective and rapid 
concealment of the posterior parts of their bodies. A suitable modification 
of the pereiopods and reduction and infolding of the obstructive abdomen 
are prerequisite to retrofossorial efRciency, and any mutations in (,hese 
directions must have had a high selective value. It is not surprising that 
favourable modifications should have presented themselves and have been 
selected several times over, nor is it surprising that, once the burrowing 
habit was adopted, similar adaptive modifications to the new condition of life 
should have established themselves. In the first place the Decapod, buried 
in sand or mud, must adapt its respiratory mechanism to the changed 
conditions. The inhalant spaces extending along the posterior and ventral 
edges of the branchiostegite would be choked and rendered useless unless 
some provision against the entrance of sand were developed, and the 
researches of Grarstang and Bohn iiave shown by what various means this 
form of sutfocation is obviateii in ditferent genera of the sand-burrowino- 
