64 DE. G. C. BOURNE ON THE EANINID^ : 
laterally by the smooth and somewhat concaye inner surfaces of I he third 
peduncular segment; ventrally by the meri of the third maxillipeds; dorsally 
by the densely setose expods of the first maxillipeds. This may be called the 
inhalant chamber. 
The antennules are inserted to the inside of and somewhat above the 
antonnse. As shown in figs. 34 and 34 a, their basal joints are expanded 
distally into a flabelliform lobe very similar to that of the fourth peduncular 
segment of the antennae and, like the latter, their margins are garnished 
with a fringe of long setpe. The slightly concave inner surfaces of these 
flabelliform expansions of the antennules, when opposed in the middle line, 
form a second vertical cleft or passage within the larger passage formed by 
the antennre, and this internal and somewhat dorsal antennulary passage is 
so disposed that its hinder opening coincides with the spout formed by the 
convergence of the exhalant canals of the aniennary sternum. The cleft 
between the basal joints of the antennules therefore serves as a conduit for 
the escurrent stream of water. Thus far the apparatus for directing the 
courses of the excurrent and incnrrent respiratory streams differs from that 
of Notoims chiefly in the fact that in llanina the long tube formed by the 
antennary flagella is absent, but in the latter genus a large part of the 
incurrent stream must find its way into the above-mentioned irdralant 
chamber through the orbital cavities. I have already shown how these 
narrow and elongated cavities are covered in by fringes of setse which meet 
over the retracted eyes and eye-stalks but are pushed aside when the latter 
are extended. Though the orbit appears to be closed below by the crest of 
the third seoment of the antennary peduncle, a sufficiently wide space is left 
between this segment and the basal joint of the antennule to allow of the 
free passage of water from the orbit into the inhalant chamber, and much of 
the inhalant current must pass through the orbits. It is evident that when 
the animal is deeply dug in, the broad front of the carapace just breaks the 
surface of the sand ; the tips of the largest frontal spines perhaps project 
very slightly above it. The hairy fringes of the orbits and of the fourth 
joints of the antennje are then just awash between sand and water, and are 
admirably adapted for filtering solid particles from the streams setting- 
inwards through the orbits and between the flabelliform fourth segments of 
the antennfe. Water is also admitted into the orbits on either side through 
a considerable cleft Ij^ing between the long and narrow sub-orbital lobe of 
the pterygostome and the innermost of the large frontal spines. These clefts 
are ouarded by dense hairy fringes. The arrangements by which the 
incurrent stream is directed from the inhalant chamber into the branchial 
chamber by the channel formed by the exopod of the second maxilliped are 
the same in Ranina as in Notopus. 
I would not have presumed to give an account unsupported by direct 
