23 
THE VOYAGE OF H.M.S. CHALLENGER. 
hardened portion, commonly known as the “ cartilaginous shell,” is neither so much 
developed nor so thick. 1 
The pallial gland is asymmetrical (PI. IY. fig. 3, a) and divided into an anterior and 
a posterior portion by an asymmetrical transparent band, which presents near its middle 
a narrow more opaque band. 
Neither buccal mass, jaws, radula, nor salivary glands are to be found ; the oesophagus 
is very extensile. As regards the rest of the visceral anatomy, all that has been noticed 
in Cymbulia holds good here also, and I have nothing to add to what has been published 
by previous authors (van Beneden, delle Chiaje, and chiefly Gegenbaur). 
The Central Nervous System (PI. IY. fig. 4) is constructed on the same plan as that 
of Cymbulia, but it may be noted that the segmentation of the cerebral ganglia (a) is 
more marked than in this latter genus. Besides this, each of the two segments has on 
its posterior aspect a little globular swelling, of the same structure as the superficial layer 
of the ganglia, that is to say, composed of large nervous cells. No nerves arise from these 
swellings. 
The ganglionic elements of the visceral commissure (c, d) are disposed in the same 
manner as in the typical genus of the family, there being three closely approximated 
ganglia. 
The innervation is quite similar to that of Cymbulia. From each cerebral ganglion 
a nerve (i) passes to the tentacle, where it expands into an olfactory or rhinophoral 
ganglion. From each pedal ganglion are given off laterally the two large nerves which 
ramify in the fin, and anteriorly a more slender nerve ( h ) which innervates the retractor 
muscle of the proboscis. 
Finally, the nerves from the visceral ganglia are disposed as in Cymbulia ( 3 , 4). 
Phylogenetic Relations oe the Thecosomata to Each Other. 
From a comparative study of the organisation of the different Thecosomata we ought 
to be able to ascertain which form has preserved the most traces of the primitive struc- 
ture of the group, that is, which of the recent species is the most nearly related to the 
ancestral form from which all the other Mollusca of the group Thecosomata have been 
derived. 
It is our ignorance of this actual primitive form which has led to false conclusions 
regarding the affinities of the Pteropoda. The knowledge of this form will permit us to 
attempt the solution of the problem (which we shall do in the second part of this Report) — 
which of the recent Mollusca are the most nearly related to those forms which have given 
origin to the group Thecosomata. 
1 Compare the systematic Report, Zool. Chall. Exp., part lxv. fig. 1, p. 97 (Cymbulia), and fig. 3, p. 102 (Gleba). 
