Q2 DE. G. C. BOURNE ON THE RANINID^ : 
the first inaxillipeds. Tliese last are concave dorsally and convex ventrally, 
and fit so closely and accurately to the converging exhalant channels o£ the 
antennary sternum as to constitute an efficient horizontal partition between 
an exhalant passage above and an inhalant passage below. The incurrent 
stream of water, taking the more ventral coursCj must pass right and left 
along the conduits provided by the grooves on the ventral sides of ihe exopods 
of the second maxillipeds ; thence under the edge of the inflated pterygostome 
into the channel which lies parallel to that in which the scaphognathite 
works, but separated from the latter by the vertical partition formed by the 
mastigohranch of the first maxilliped. The floor of these inhalant passages 
is formed by the flat and closely opposed meri and iscliia of the third 
maxillipeds the outer edges of which fit close to the mouth frame. There is 
no " exostegal " groove in the pterygostome forming a definite inhalant canal 
as in the Leucosiidfe, but the relation of the incurrent to the excurrent stream 
of water must otherwise be much the same in the two groups. That is to 
say, in both cases there must be two opposite currents, the one setting- 
inwards to, the other setting outwards from the antennarj^ region of the front. 
In both the Raninidse and the Leucosiidse the excurrent stream is set in 
motion by the " normal '' action of the scaphognathite, and passes forward 
between the converging channels of the antennary sternum and the exopods 
of the first maxillipeds to emerge between the basal joints of the antennides. 
The course of the incurrent stream in the Raninidse is as described above for 
Notopus, but it can liardly have escaped the reader's attention that the 
channels in which the opposite currents flow are not very distinctly separated 
from one another. In the Leucosiidaj the existence of the exostegal canals 
makes the separation much more complete, and the respiratory mechanism is 
in this respect more highly specialised than it is in the Raninida?- In 
this connection it should not be forgotten that the latter have additional 
apertures of ingress to the branchial chambers, viz., the posterior branchial 
apertures of which there is no counterpart in the Leucosiidaj. 
It may be surmised that Notopus makes use of the posterior branchial 
apertures only when it is lightly covered by the loose upper stratum of sand, 
and can make use of the raking action of the chelipeds to clear a passage 
for water below the thorax. As it digs down into the more compact deeper 
layers of sand, this source of water-supply must become more and more 
restricted, and the antennary tabe is then brought into action. It is of 
course possible and even probable that, when the antenmiry tube is used, 
the action of the scaphognathite is reversed and the posterior branchial 
apertures become exhalant, the tVontal exhalant passages being closed and 
thrown out of action so long as the respiratory current is reversed, only to be 
brought into use again when the normal respiratory current, from behind 
forwards, is restored. But the evidence is strongly in favour of my contention 
