in the Cephalopoda. 97 



third arm on the right {Tremoctopus) or left {Argonauta) side of 

 the animal, — is also formed differently to the other arms in the 

 genus OctojMs, and in this case it is always the arm on the right 

 side which has hecome transformed. This arm is always shorter 

 than the left one, even to a considerable extent, as its length in 

 diflFerent species is only one-half to three-fourths of that of the 

 latter ; and as, besides this, it not only often retains the same 

 thickness, but is even more muscular in its outer half, it has 

 frequently a more powerful appearance. It bears far fewer 

 acetabula than the left arm, and is furnished externally at the 

 apex with a peculiar, usually longish plate, which is provided, 

 in most species, with a greater or less number of transverse 

 wrinkles or ribs with intervening pits. This plate is also con- 

 nected with the swimming-membrane at the base of the arm by 

 the agency of a muscular border of skin, which runs down along 

 the dorsal margin of the arm, and this border is very often found 

 with its free margin rolled up towards the inner side of the arm, 

 by which a more or less closed canal is formed, which is un- 

 doubtedly destined to conduct the spermatophora to the apical 

 plate of the arm. As this canal or semicanal is destitute of 

 chromatophora on the inside, and perfectly white in most spe- 

 cies, I conclude that in the living animal this membranous 

 margin will in general be bent towards the side of the arm, as 

 was seen in most of the spirit specimens. 



This is the case at least in Octopus grcenlandicus, Dewhurst 

 (=0. arcticiis, Prosch), in five males of which I find only 41-43 

 acetabula on the right arm of the third pair (whilst on the cor- 

 responding left arm I find 74-79), a spoon-shaped prehensile 

 plate furnished wdth 13-17 transverse ribs at its apex, and a 

 cutaneous margin which extends from the latter to the middle 

 of the connecting membrane between the third and fourth arms, 

 where the semicanal or groove, formed by this margin, also 

 suddenly ceases. Fig. 2, on PI. III., which is of the natural size, 

 is intended to assist the comprehension of this peculiarity : a is 

 the grasping-plate, as I have called it, separated from the sucker- 

 bearing part of the arm by a high angular fold of skin {d) ; 

 b, b is the membranous border; and c, the place where it com- 

 mences or terminates at the margin of the swimming-mem- 

 brane*. In one specimen, a seminal capsule or spermatophore 

 projected from the funnel, and was probably on the way towards 

 this membranous border, to which it is most likely conveyed by 

 the upper extremity of the funnel simply laying itself against 

 the commencement of the fold of skin. 



* The sucking-disks on all the true arms are of ahout the same size ; 

 their appearing larger on the first pair in the figure is owing to the direc- 

 tion in which they were seen by the artist. 



Ann. ^- Mag. N. Hist. Ser. 2. Vol. xx. 7 



