76 THE VOYAGE OF H.M.S. CHALLENGER. 



also ; the warts and cirri certainly vary according to the mode of preservation, but they 

 are rarely so masked as not to be perceptible on careful examination. The matter is 

 still open to question, however, and any one who can readily obtain numbers of fresh 

 specimens would do a good work by killing and preserving them in different ways and 

 reporting upon the variations thus produced in their surface markings. 



A character which has been much used by some authors, Gray for instance, is the 

 fact that a certain number of the proximal cups are often arranged in a single series. 

 This, I feel convinced, is of very little importance ; the statement that the suckers in the 

 genus Octopus are in two series is not, strictly speaking, correct, they are placed in a zigzag 

 so close that they often appear to form a succession of pairs, but when the arm is extended 

 {bent away from the mouth) the proximal portion of the zigzag is stretched out and the 

 suckers fall into a more or less nearly straight line, on the other hand, when the arm is 

 flexed (towards the mouth) the zigzag closes up and the two series reappear. The 

 presence or absence of certain large suckers on the lateral arms, used by d'Orbigny, and 

 by Gray and Tryon following him, for specific diagnosis, has long since been shown by 

 Steenstrup to be merely a sexual character. 



To sum up, then, I have paid some attention to the form and proportions of the body, 

 not much to variations in the lengths of the arms (unless these were considerable), some 

 to the colour, and a good deal to the nature of the surface of the body. 



Any peculiarities that presented themselves in the suckers have, of course been noted, 

 and the same is the case with the hectocotylised arm whenever this was present.. 



Of this last structure I have observed what seem to be three fairly distinct types : — 



1. That found in Octopus vulgaris. Octopus marmoratus (PL VI. figs. 2, 3), &c., 

 where the modified extremity is minute in relation to the arm ; it is conical, or rather 

 pyramidal, and has a narrow groove on the inner side. 



. 2. The form seen in Octopus levis (PL II. fig. 2), Octopus januarii (PL VII. fig. 2), 

 &c., which resembles the last in form, but is decidedly more bulky and conspicuous. 

 It shows a tendency to develop transverse ridges on the interior of the spoon-fehaped 

 termination, and reaches its extreme form in Octopus ohesus, VerrUl.^ 



3. The form which has only been found, so far as I am aware, in Octopus punctatus, 

 Gabb. (see PL V. fig. 2). It is slender and very long in proportion to the arm, being 

 nearly one-tenth of its total length in the Challenger specimen. 



There is a temptation to break up the large genus Octopus into groups based 

 upon characters derived from this organ, but I have been unable at present to ascertain 

 that they are correlated with other distinctions between the diff"erent forms, and it 

 may prove impossible to keep these three types separate when the structure of the 

 hectocotylised arm shall become known in a greater number of species than is at 

 present the case. 



1 See Cepli. N. E. Amer., pi. xxxvi. fig. 4. 



