PRESIDENTIAL ADDRESS. by7/ 
similar in size and form to the other arms. The suckers, according to Steenstrup, 
are reduced to one on each arm ; but so small a number has not been found in any 
of the specimens subsequently examined; and in any case diminution in the 
number of suckers in the hectocotylised arm is by no means an uncommon 
phenomenon. Compare, for example, Sepia, Todaropsis, Sepioloidea. When we con- 
sider the absence of the shell in Jdiosepius, and its size and character in Spirwla, 
we must, I think, conclude that, at any rate, these two forms cannot be so closely 
related to each other as to belong to the same sub-family, or even family. 
With regard to the question at issue between Pelseneer and Lonnberg as to 
the Cigopsid or Myopsid nature of Spzruda, I think, on the whole, that its resem- 
blance is to the former rather than to the latter; but I believe that the branch 
of the ancestral tree which terminates in Spirwla was given off from the main 
Cephalopod stem before the Gigopsida and Myopsida, as we now know them, had 
been separately evolved. Paleontology reveals a possible descent of Spirula from 
a Belemnitoid through such an intermediate form as Sperwlirostra ; and from this, 
on the other hand, it is easy to conceive of the descent of Sepia through a form 
resembling Belosepia. Such a relation could be expressed by the following diagram, 
which is, however, only a rough illustration of possibilities, for Spzrulirostra is a 
miocene form and Belosepia an eocene, so that the former could hardly be the 
ancestor of the latter. It is only contended that these forms indicate a possible 
line of descent. 
Sepia 
Spirula Belosepia 
Spirulirostra 
Belemnites 
Unfortunately, in the present state of our knowledge, it is impossible to corre- 
late the above diagram with one based upon the study of the soft parts of recent 
forms. It is sufficient if they do not contradict each other. We know nothing 
of the soft parts of the fossils, and there is no recent form, which exhibits shell 
cbaracters, bridging over the gulf between Sepia and Spirula. To sum up, 
Spirula must be regarded as, at all events, the representative of a distinct family : it 
it is not unlikely that it may one day become the type of a division coequal with 
Myopsida and Cgopsida, and it does not appear to me that the structure of its 
hectocotylised arms would be any argument against such a view. 
We may now consider the genera Idiosepius, Sepiadartum, and Sepioloidea. 
Idioseprus has the following characters :— 
1. The shell is wanting. 
2. The connection between the mantle and head resembles that of Sepiola, but 
is in a transition state. 
3. The fins are circular ; rather posterior in position. 
4, There are no siphonal adductors visible externally, 
5. Pallial adductors are present. 
6. The pallial nerve is divided into two branches for a short distance; the 
stellate ganglion is elongated. 
7. The liver capsule is completely closed. 
8. There are small anterior salivary glands, as in Rossia; the posterior are 
separated, as in Sepzola. 
9. The testicular canals are radially disposed about a round superficial 
depression. 
10. Both ventral arms are hectocotylised, the suckers being reduced to 4, 3, 2, 
or 1, and the spermatophores are attached to the buccal membrane, 
