PART L FURTHER LIGHT ON THE OCTOPUS RUGOSUS PROBLEM By GRACE E. PICKFORD IN spite of the efforts of Robson (1929), which form a landmark in the history of octopodan taxonomy, the Octopus rugosus problem remains unsolved, an outstanding source of confusion obscuring the correct identification of small, rough, vulgaris-\ike specimens from all parts of the world. Some years ago the writer reviewed the western Atlantic forms which had been, or could be, assigned to Octopus rugosus. It was possible to show that no such species can be found in American waters and that western Atlantic specimens previously referred to it were either 0. vulgaris or, occasionally, belonged to one of the other Caribbean species (Pickford, 1945, 1946). It is believed that the problem can best be solved by a zoogeographical approach, pan passu with increasing knowledge of each local fauna to the point that makes the recognition of its component species a matter of certainty. Through the work of Voss (1950, 1951, etc.) and the author we have now arrived at this state of know-ledge of the American east coast fauna. However, the situation in other parts of the world is less satisfactory. Octopodan taxonomy has always been hampered by the lack of adequate series of specimens collected from any one locality and this, together with the confusing effects of different states of preservation, has made recognition of the limits of variation within a single species all but impossible. In the present contribution an attempt has been made to clarify the status of those specimens from the eastern Atlantic and South Africa which Robson identified as 0. rugosus and which are now in the collections of the British Museum. A survey of bodily characteristics, irrespective of species assignation, is given in Table I. It will be seen that only 7 of the 22 specimens are sexually mature, 4 are well grown but immature, and n are juvenile. In a difficult problem of this nature it is obvious that no definite conclusions can be based on juvenile, nor even on sexually immature specimens, except in so far as they can reasonably be assigned to adult forms with well established characteristics. Octopus rugosus, as it is generally understood, is a species closely resembling 0. vulgaris but smaller in size ; the head is broader, the arms shorter, and the rough, finely granulated skin is darkly reticulated with black. None of these characters, taken singly, can be used with any degree of certainty for the separation of the two species. Taken together the resulting picture is indistinguishable from that of juvenile and immature specimens of 0. vulgaris, so that it may be conjectured that 0. rugosus is merely the young of 0. vulgaris (Pickford, 1945). This point of view is that taken by the writer who believes that it is necessary to prove that an Atlantic specimen assigned to 0. rugosus is not a young 0. vulgaris before it can be considered to belong to any other species ; a rugosus-like specimen should be assigned to 0. ZOOL. 3, 3. 13
BioStor
