1 1 



The two lateral ridges c and d have not a straight course as the two ridges (/ and Jc) 

 which they unite are not of equal length. Their course is gradual!)' divergent from top to bottom. 



The position of facet C is rather difficult to make clear. It is not a true facet but serves 

 at the same time to protect siphon, tentacle and gonocalyx. Proximally it is a true cavity, 

 hollowed out between the dorsal facets, facet D and E, and the ventral facet B. The upper 

 ridge (we may call it ridge, as long as it is a clear definite line separating two facets one from 

 the other) is called ;;/. It is curved, meets the ridg-es c and d in the angles d" and e" and 

 goes gradually to the centre of the bract, leaving the surface and is quite imbedded in the 

 centre of the gelatinous substance (see PI. I, fig. 4). 



It is only ridge 111 which is a true ridge, not belonging to any other facet. So the other 

 outline of facet C will be described by the dorsal facets D and E. 



It is clear that facet B and C together constitute the ventral side of the bract. 



The dorsal side cannot be described without the lateral ones, as there are three facets 

 and the asymmetry is very marked. 



As was said above the asymmetry appears on the right side of the bract. 



The ridge going down from the angle b' is called b. It is the dorsal ridge of a four- 

 sided facet called F. This facet is limited on the opposite (the ventral) side proximally by rido-e 

 c (angles c' and c") and distally by ridge /. Ridge j is very much longer and curved ; it unites 

 the much longer ridge b to ridge c which is shorter. 



Ridge b is at the same time the left lateral ridge of the fifth facet E (textfig. 10), 

 which is also irregularly four-sided, its proximal ridge // being the shortest. The other ridges 

 are a (on the right lateral (= dorsal) side) and posteriorly /. They are both slightly convex 

 and a little serrated at the angle (a") where they meet. The angle /' (connecting ridge b and /) 

 is also a little serrated. 



Of course there exists the same difterence in position between the angles b' and a\ 

 and j' and a" . 



The last facet D (textfigure 9) is one composed quite probably of two. 



It consists of ridge a on the left (dorsal) side, ridge e proximally and ridge d on the 

 right (ventral) side; distally we see a ridge n which is very distinctly curved in the middle 

 whereas on the opposite side (facet E) there is the angle j' . It is this total absence of a ridge 

 that should have be<:run in the centre of rid^e e and have run backwards to meet the curved 

 ridge n somewhere in the middle of its length (thus constituting a double facet such as we 

 notice on the left) which makes this bract so absolutely asymmetrical. 



We have carefully observed the bracts of this species and we did not find the slightest 

 deviation from this particularity in any one of them. It will now have become a great deal 

 easier to understand the shape of the bract as we have sketched it in PI. I, figg. 2, 3, 4. We 

 see the different facets as we have described them above. 

 The Phyllocyst. 



The phyllocyst (PI. I, figg. 2, 3, 4 phyl.) bears much resemblance to that of Ainphiroa 

 alata Les. It has the same two wings although longer and narrower ; laterally it is flattened 

 on the ventral side of the bract, and the wings begin at the ventral proximal side. They go 



